
From nobody Sun Mar  1 11:01:39 2015
Return-Path: <adrian@olddog.co.uk>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 84F961A1BD9; Sun,  1 Mar 2015 11:01:37 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.9
X-Spam-Level: 
X-Spam-Status: No, score=-101.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id glei0fpchsHg; Sun,  1 Mar 2015 11:01:36 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 4AC1E1A1BCC; Sun,  1 Mar 2015 11:01:36 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
From: "Adrian Farrel" <adrian@olddog.co.uk>
To: "The IESG" <iesg@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150301190136.14291.13104.idtracker@ietfa.amsl.com>
Date: Sun, 01 Mar 2015 11:01:36 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/uFo71I2iI1Foqjv7biVrIgxMtBs>
Cc: draft-ietf-bess-orf-covering-prefixes.all@ietf.org, bess-chairs@ietf.org, martin.vigoureux@alcatel-lucent.com, bess@ietf.org
Subject: [bess] Adrian Farrel's Yes on draft-ietf-bess-orf-covering-prefixes-04: (with COMMENT)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 01 Mar 2015 19:01:37 -0000

Adrian Farrel has entered the following ballot position for
draft-ietf-bess-orf-covering-prefixes-04: Yes

When responding, please keep the subject line intact and reply to all
email addresses included in the To and CC lines. (Feel free to cut this
introductory paragraph, however.)


Please refer to http://www.ietf.org/iesg/statement/discuss-criteria.html
for more information about IESG DISCUSS and COMMENT positions.


The document, along with other ballot positions, can be found here:
http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/



----------------------------------------------------------------------
COMMENT:
----------------------------------------------------------------------

Some minor editorial bits and pieces we can pickup after the telechat.
(Thanks to Dave Black and Alvaro Retana for the reviews).

---

Something can be added to the Introduction (which is a bit short) to
explain what a covering prefix is. Probably breaking the second paragraph
after the first sentence and adding a new sentence that begins "A
covering prefix is..."

---

The fields in Figures 1 are briefly described after the figure, but in
some cases the description is exceeding brief. For example, minlen.

Can you look at the text and see where you can beef it up either with a
forward reference to where the field is described, or with a simple
statement about what the field is used for.

---

Section 3 has  "RECOMMENDS" which is not a 2119 word. To use
"RECOMMENDED" you will need to reword slightly.

---

Section 8 has a paragraph that begins  "Section 5…” I think this applies
specifically to the first bullet in the list (of two) immediately above.
If I'm right, can you move the paragraph into that bullet?

---

Can you move RFC 4364 to be an Informative reference.



From nobody Sun Mar  1 11:01:40 2015
Return-Path: <adrian@olddog.co.uk>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id A75681A1BDE; Sun,  1 Mar 2015 11:01:37 -0800 (PST)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 84F961A1BD9; Sun,  1 Mar 2015 11:01:37 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.9
X-Spam-Level: 
X-Spam-Status: No, score=-101.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id glei0fpchsHg; Sun,  1 Mar 2015 11:01:36 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 4AC1E1A1BCC; Sun,  1 Mar 2015 11:01:36 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
From: "Adrian Farrel" <adrian@olddog.co.uk>
To: "The IESG" <iesg@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150301190136.14291.13104.idtracker@ietfa.amsl.com>
Date: Sun, 01 Mar 2015 11:01:36 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/uFo71I2iI1Foqjv7biVrIgxMtBs>
Cc: draft-ietf-bess-orf-covering-prefixes.all@ietf.org, bess-chairs@ietf.org, martin.vigoureux@alcatel-lucent.com, bess@ietf.org
Subject: [bess] Adrian Farrel's Yes on draft-ietf-bess-orf-covering-prefixes-04: (with COMMENT)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 01 Mar 2015 19:01:37 -0000

Adrian Farrel has entered the following ballot position for
draft-ietf-bess-orf-covering-prefixes-04: Yes

When responding, please keep the subject line intact and reply to all
email addresses included in the To and CC lines. (Feel free to cut this
introductory paragraph, however.)


Please refer to http://www.ietf.org/iesg/statement/discuss-criteria.html
for more information about IESG DISCUSS and COMMENT positions.


The document, along with other ballot positions, can be found here:
http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/



----------------------------------------------------------------------
COMMENT:
----------------------------------------------------------------------

Some minor editorial bits and pieces we can pickup after the telechat.
(Thanks to Dave Black and Alvaro Retana for the reviews).

---

Something can be added to the Introduction (which is a bit short) to
explain what a covering prefix is. Probably breaking the second paragraph
after the first sentence and adding a new sentence that begins "A
covering prefix is..."

---

The fields in Figures 1 are briefly described after the figure, but in
some cases the description is exceeding brief. For example, minlen.

Can you look at the text and see where you can beef it up either with a
forward reference to where the field is described, or with a simple
statement about what the field is used for.

---

Section 3 has  "RECOMMENDS" which is not a 2119 word. To use
"RECOMMENDED" you will need to reword slightly.

---

Section 8 has a paragraph that begins  "Section 5…” I think this applies
specifically to the first bullet in the list (of two) immediately above.
If I'm right, can you move the paragraph into that bullet?

---

Can you move RFC 4364 to be an Informative reference.



From nobody Mon Mar  2 02:07:22 2015
Return-Path: <xuxiaohu@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8204A1A7013 for <bess@ietfa.amsl.com>; Mon,  2 Mar 2015 02:07:21 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.211
X-Spam-Level: 
X-Spam-Status: No, score=-4.211 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id npZBM9mbgd5d for <bess@ietfa.amsl.com>; Mon,  2 Mar 2015 02:07:16 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8F5D01A700D for <bess@ietf.org>; Mon,  2 Mar 2015 02:07:15 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BTE11527; Mon, 02 Mar 2015 10:07:14 +0000 (GMT)
Received: from nkgeml409-hub.china.huawei.com (10.98.56.40) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Mon, 2 Mar 2015 10:07:12 +0000
Received: from NKGEML512-MBS.china.huawei.com ([169.254.8.115]) by nkgeml409-hub.china.huawei.com ([10.98.56.40]) with mapi id 14.03.0158.001; Mon, 2 Mar 2015 18:07:00 +0800
From: Xuxiaohu <xuxiaohu@huawei.com>
To: Xuxiaohu <xuxiaohu@huawei.com>, "adrian@olddog.co.uk" <adrian@olddog.co.uk>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE: Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
Thread-Index: AQHQSLr3DiuxdRaR30WUQ5uVlB6ZeJzw6wsggBgiBbA=
Date: Mon, 2 Mar 2015 10:07:00 +0000
Message-ID: <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE0830EF91@NKGEML512-MBS.china.huawei.com>
References: <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE083064D3@NKGEML512-MBS.china.huawei.com> <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE083066B6@NKGEML512-MBS.china.huawei.com> <00d801d048ba$e4eb09b0$aec11d10$@olddog.co.uk> 
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.99.55]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/Qhs85AY12IFSyWbMDWYf1Wfiefo>
Subject: Re: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE: Why transform draft-xu-softwire-encaps-udp to	draft-xu-bess-encaps-udp
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 02 Mar 2015 10:07:21 -0000

Besides the reasons as mentioned below, since the BGP Encapsulation SAFI fo=
r UDP or MPLS-in-UDP is not only applicable to BGP/MPLS-based L2VPN (includ=
ing VPLS and EVPN) and BGP/MPLS-based L3VPN, but also be applicable to othe=
r use cases (e.g., incremental deployment of BGP-based Segment Routing), it=
 seems better to specify it in a separate doc rather than in the end-system=
-l3vpn draft (http://datatracker.ietf.org/doc/draft-ietf-l3vpn-end-system/)=
 which is dedicated for how to use XMPP as a replacement of the BGP-L3VPN s=
ignaling, IMHO.

Best regards,
Xiaohu

> -----Original Message-----
> From: Xuxiaohu
> Sent: Sunday, February 15, 2015 11:31 AM
> To: 'adrian@olddog.co.uk'; bess@ietf.org
> Subject: RE: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//=
RE:
> Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
>=20
> Hi Adrian,
>=20
> > -----Original Message-----
> > From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Adrian Farrel
> > Sent: Sunday, February 15, 2015 9:01 AM
> > To: bess@ietf.org
> > Subject: Re: [bess] Request for WG adoption of draft-xu-bess-encaps-udp=
//RE:
> > Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
> >
> > Just for clarity on this point:
> >
> > I suggested that BESS was the right place to discuss whether and how
> > to use BGP to indicate tunnel types.
>=20
> Thanks for making that suggestion. Although RFC5512 (i.e.,
> draft-ietf-softwire-encaps-safi) and RFC5566 (i.e.,
> draft-ietf-softwire-encaps-ipsec) have been originated from Softwire WG a=
nd
> the current Softwire WG charter (https://tools.ietf.org/wg/softwire/chart=
ers)
> still state that "BGP and other routing and signaling protocols developed=
 in this
> group will be reviewed jointly with the proper working groups and other
> workings that may take interest (e.g. IDR, L3VPN, PIM, LDP, SAAG, etc)", =
your
> above suggestion, together with Softwire co-chairs' claim of last Friday =
that the
> softwire WG is going to shut down and therefore it would not be a right p=
lace to
> pursue this draft anymore, seem to be a joint statement that any future w=
ork
> related to BGP Encapsulation SAFI should be discussed in the BESS WG, rat=
her
> than the Softwire WG. Thanks again for clarifying the confusion.
>=20
> > I also observed that a code point already exists in the BGP Tunnel
> > Encapsulation Attribute Tunnel Types registry at
> > http://www.iana.org/assignments/bgp-parameters/bgp-parameters.xhtml#tu
> > n
> > nel-types
>=20
> > Therefore the debate the WG needs to have (perhaps before debating
> > adoption of this document) is:
> >
> > - whether the existing code point is enough for the purpose of identify=
ing
> >   MPLS-in-UDP tunnels or whether more work is needed
> > - whether a foo-in-UDP tunnel type should be used instead with sub-TLVs
> >   to identify the different cases of foo.
>=20
> IMHO, it depends on whether the tunnel type of the Tunnel Encapsulation T=
LV as
> specified in RFC5512 should be interpreted as UDP or MPLS-in-UDP.
>=20
> See the following text quoted from RFC 5512 (draft-ietf-softwire-encaps-s=
afi):
>=20
> ****************
> 4.2.  Protocol Type Sub-TLV
>=20
>    The protocol type sub-TLV MAY be encoded to indicate the type of the
>    payload packets that will be encapsulated with the tunnel parameters
>    that are being signaled in the TLV.  The value field of the sub-TLV
>    contains a 2-octet protocol type that is one of the types defined in
>    [IANA-AF] as ETHER TYPEs.
>=20
>    For example, if we want to use three L2TPv3 sessions, one carrying
>    IPv4 packets, one carrying IPv6 packets, and one carrying MPLS
>    packets, the egress router will include three TLVs of L2TPv3
>    encapsulation type, each specifying a different Session ID and a
>    different payload type.  The protocol type sub-TLV for these will be
>    IPv4 (protocol type =3D 0x0800), IPv6 (protocol type =3D 0x86dd), and
>    MPLS (protocol type =3D 0x8847), respectively.
>=20
> 4.4.  Tunnel Type Selection
>=20
>    A BGP speaker may include multiple tunnel TLVs in the tunnel
>    attribute.  The receiving speaker MAY have local policies defined to
>    choose different tunnel types for different sets/types of payload
>    prefixes received from the same BGP speaker.  For instance, if a BGP
>    speaker includes both L2TPv3 and GRE tunnel types in the tunnel
>    attribute and it also advertises IPv4 and IPv6 prefixes, the ingress
>    router may have local policy defined to choose L2TPv3 for IPv4
>    prefixes (provided the protocol type received in the tunnel attribute
>    matches) and GRE for IPv6 prefixes.
>=20
> ***************
>=20
> To keep it in accordance with the usage of the tunnel type as specified i=
n
> RFC5512, I believe the tunnel type of the Tunnel Encapsulation TLV should=
 be
> interpreted as UDP, rather than MPLS-in-UDP. That's to say, what type of
> payload will be carried over the UDP tunnel cloud be explicitly indicated=
 by the
> protocol type sub-TLV.
>=20
> Best regards,
> Xiaohu
>=20
> > Adrian
> >
> > > -----Original Message-----
> > > From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Xuxiaohu
> > > Sent: 13 February 2015 06:10
> > > To: Xuxiaohu; bess@ietf.org; bess-chairs@tools.ietf.org
> > > Cc: draft-xu-bess-encaps-udp@tools.ietf.org;
> > > softwire-chairs@tools.ietf.org
> > > Subject: [bess] Request for WG adoption of
> > > draft-xu-bess-encaps-udp//RE: Why transform
> > > draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
> > >
> > > Hi BESS WG co-chairs,
> > >
> > > The BGP tunnel type for MPLS-in-UDP has been mentioned in the
> > > MPLS-in-UDP draft since the 00 version
> > > (https://tools.ietf.org/html/draft-xu-mpls-in-udp-
> > > 00#page-4) which was published on April 28, 2012. However, according
> > > to the WG consensus during the WG adoption poll period, that section
> > > of "Signaling for Encapsulation in UDP" was removed from the
> > > MPLS-in-UDP draft and accordingly it was specified in a separate
> > > draft
> > (https://tools.ietf.org/html/draft-xu-softwire-
> > > encaps-udp-00) which was published on February 12, 2013.
> > >
> > > Adrian has suggested me to post this draft to BESS. Furthermore,
> > > Suresh has indicated that the Softwire WG would not be a right place
> > > for any BGP tunnel type related works anymore since the WG is going
> > > to shut down in the very near future. Hence, would you please start
> > > a WG adoption for draft-xu-bess-encaps- udp which is transformed
> > > from
> > draft-xu-softwire-encaps-udp?
> > >
> > > Best regards,
> > > Xiaohu
> > >
> > > > -----Original Message-----
> > > > From: Xuxiaohu
> > > > Sent: Friday, February 13, 2015 12:13 PM
> > > > To: Xuxiaohu; bess@ietf.org
> > > > Cc: Softwires WG
> > > > Subject: RE: [bess] Why transform draft-xu-softwire-encaps-udp to
> > > > draft-xu-bess-encaps-udp
> > > >
> > > > Hi all,
> > > >
> > > > I noticed the following text from RFC 5512
> > (draft-ietf-softwire-encaps-safi):
> > > >
> > > > ****************
> > > > 4.2.  Protocol Type Sub-TLV
> > > >
> > > >    The protocol type sub-TLV MAY be encoded to indicate the type of=
 the
> > > >    payload packets that will be encapsulated with the tunnel parame=
ters
> > > >    that are being signaled in the TLV.  The value field of the sub-=
TLV
> > > >    contains a 2-octet protocol type that is one of the types define=
d in
> > > >    [IANA-AF] as ETHER TYPEs.
> > > >
> > > >    For example, if we want to use three L2TPv3 sessions, one carryi=
ng
> > > >    IPv4 packets, one carrying IPv6 packets, and one carrying MPLS
> > > >    packets, the egress router will include three TLVs of L2TPv3
> > > >    encapsulation type, each specifying a different Session ID and a
> > > >    different payload type.  The protocol type sub-TLV for these wil=
l be
> > > >    IPv4 (protocol type =3D 0x0800), IPv6 (protocol type =3D 0x86dd)=
, and
> > > >    MPLS (protocol type =3D 0x8847), respectively.
> > > > ***************
> > > >
> > > > It seems that RFC5512 is not only talking about IP-in-GRE, but
> > > > also
> > MPLS-in-GRE.
> > > >
> > > > I also noticed an expired IDR WG doc (see
> > > > https://tools.ietf.org/html/draft-ietf-idr-encaps-safi-00) which
> > > > was a predecessor of draft-ietf-softwire-encaps-safi. Besides,
> > > > RFC5566
> > > > (draft-ietf-softwire-encaps-ipsec) is also originated from the Soft=
wire WG.
> > > Does
> > > > it mean which WG should be responsible for the BGP Tunnel
> > > > Encapsulation Attribute related work had ever been discussed and
> > > > finally determined that
> > the
> > > > Softwire WG is the right place for it?
> > > >
> > > > Best regards,
> > > > Xiaohu
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Xuxiaohu
> > > > > Sent: Friday, February 13, 2015 9:21 AM
> > > > > To: bess@ietf.org
> > > > > Cc: Softwires WG
> > > > > Subject: [bess] Why transform draft-xu-softwire-encaps-udp to
> > > > > draft-xu-bess-encaps-udp
> > > > >
> > > > > Hi all,
> > > > >
> > > > > According to the suggestion from Adrian as a Routing co-AD,
> > > > > draft-xu-softwire-encaps-udp
> > > > > (https://tools.ietf.org/html/draft-xu-softwire-encaps-udp) which
> > > > > was posted to the Softwire WG is now posted to the BESS WG.
> > > > >
> > > > > Any comments and suggestions are welcome.
> > > > >
> > > > > Best regards,
> > > > > Xiaohu
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Xuxiaohu
> > > > > > Sent: Friday, February 13, 2015 8:45 AM
> > > > > > To: 'adrian@olddog.co.uk'; 'Black, David';
> > > > > > rcallon@juniper.net;
> > > > > > draft-ietf-l3vpn-end-system@tools.ietf.org;
> > > > > > bess-chairs@tools.ietf.org; softwire-chairs@tools.ietf.org
> > > > > > Cc: 'Alvaro Retana'; akatlas@gmail.com; 'Loa Andersson'
> > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > draft-ietf-mpls-in-udp
> > > > > >
> > > > > > Hi Adrian,
> > > > > >
> > > > > > Thanks a lot for your response. Although RFC5512 (i.e.,
> > > > > > draft-ietf-softwire-encaps-safi) and RFC5566 (i.e.,
> > > > > > draft-ietf-softwire-encaps-ipsec) which specify the BGP Tunnel
> > > > > > Encapsulation Attribute Tunnel Types for GRE, L2TPv3 and IPsec
> > > > > > respectively are all originated from Softwire, and further the
> > > > > > Softwire WG co-chairs didn't state that
> > > > > > draft-xu-softwire-encaps-udp doesn't belong to their WG, if
> > > > > > the BESS and Softwire WG co-chairs could reach an agreement
> > > > > > that any future work related to BGP Tunnel Encapsulation
> > > > > > Attribute should be done in the BESS WG, it looks fine to me.
> > > > > > I would submit the same draft to the BESS WG as
> > > > > draft-xu-bess-encaps-udp.
> > > > > >
> > > > > > Best regards,
> > > > > > Xiaohu
> > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: Adrian Farrel [mailto:adrian@olddog.co.uk]
> > > > > > > Sent: Thursday, February 12, 2015 10:50 PM
> > > > > > > To: Xuxiaohu; 'Black, David'; rcallon@juniper.net;
> > > > > > > draft-ietf-l3vpn-end-system@tools.ietf.org;
> > > > > > > bess-chairs@tools.ietf.org; softwire-chairs@tools.ietf.org
> > > > > > > Cc: 'Alvaro Retana'; akatlas@gmail.com; 'Loa Andersson'
> > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > draft-ietf-mpls-in-udp
> > > > > > >
> > > > > > > Hello all,
> > > > > > >
> > > > > > > 1. Why softwire? That is strictly IP-in-IP with a particular
> > > > > > > intention of 4-over-6 and 6-over-4. Why would MPLS-in-UDP
> > > > > > > fall into their
> > > > > charter?
> > > > > > > You say "all the specifications for the BGP signaling for
> > > > > > > GRE, IPsec and etc were all defined in separate drafts
> > > > > > > belonging to the Softwire WG" but I see no evidence of this.
> > > > > > > The only vaguely related draft I can see is
> > > > > > > draft-xu-softwire-ip-in-udp which is a specific
> > > > > > > IP-over-UDP-over-IP mechanism about which I will reserve
> > > > > > > judgement except to say that I that softwire really needs
> > > > > > > yet another transition mechanism and that I believe IP-in-IP
> > > > > > > can be hashed by existing ECMP
> > > > > > hardware.
> > > > > > >
> > > > > > > You also referenced draft-xu-softwire-encaps-udp but I
> > > > > > > believe this document expired over 12 months ago. I would
> > > > > > > not say that it was the most substantive or technical
> > > > > > > document I have ever read
> > > > > > > :-)
> > > > > > >
> > > > > > > I have not removed the chairs from this thread, but I really
> > > > > > > hate spamming people's in-boxes.
> > > > > > >
> > > > > > > 2. While Xiaohu has correctly pointed at the current version
> > > > > > > of the I-D, it might be better to look at the status in the
> > > > > > > Datatracker via
> > > > > > > http://datatracker.ietf.org/doc/draft-ietf-l3vpn-end-system/
> > > > > > > - You'll see that the status is Waiting for AD
> > > > > > > Go-Ahead::Revised I-D Needed which means it has completed
> > > > > > > IETF last call and is waiting for a
> > > > > > revision.
> > > > > > >
> > > > > > > 3. If you are following the BESS mailing list, you'll see
> > > > > > > that there is text agreed with IANA to fix the "empty" IANA
> > > > > > > considerations
> > > > section.
> > > > > > > http://www.ietf.org/mail-archive/web/bess/current/msg00233.h
> > > > > > > tm l This will be in the next revision of the draft.
> > > > > > >
> > > > > > > 4. I am sure we can involve the BESS chairs any time we note
> > > > > > > some work that they need to do. At the moment, they may be
> > > > > > > interested to know there is a conversation, but I don't know
> > > > > > > that we have identified any actions for them. I have not
> > > > > > > removed them from this thread, but I really hate spamming
> > > > > > > people's
> > in-boxes.
> > > > > > >
> > > > > > >
> > > > > > > I believe there are two pieces of work:
> > > > > > >
> > > > > > > A. Assign a BGP Tunnel Encapsulation Attribute Tunnel Type.
> > > > > > > This has already been done. No amount of effort to change
> > > > > > > documents or advance one document or another will change
> > > > > > > this fact. The code point has already been assigned. The
> > > > > > > registry is "First Come First Served" and no particular
> > > > > > > process was required except an application to IANA. No
> > > > > > > further
> > > > > > action is desirable.
> > > > > > >
> > > > > > > B. Specify necessary protocol work to utilise this code point=
.
> > > > > > > This is a matter for the BESS WG. They may consider that
> > > > > > > everything needed has already been documented, they may
> > > > > > > consider that they do not want to specify anything, they may
> > > > > > > consider that further work is needed and can be based on
> > > > > > > your I-D, they may consider that further work is needed and
> > > > > > > can needs a different starting point. The correct way to
> > > > > > > handle this is to post your I-D and take the discussion to
> > > > > > > the BESS
> > > > > mailing list.
> > > > > > You may ask the BESS chairs for advice.
> > > > > > >
> > > > > > >
> > > > > > > What am I missing here?
> > > > > > > What do you want to achieve and why?
> > > > > > >
> > > > > > > What action are you actually asking for?
> > > > > > >
> > > > > > > Adrian
> > > > > > >
> > > > > > > > -----Original Message-----
> > > > > > > > From: Xuxiaohu [mailto:xuxiaohu@huawei.com]
> > > > > > > > Sent: 12 February 2015 05:56
> > > > > > > > To: Xuxiaohu; Black, David; adrian@olddog.co.uk;
> > > > > > > > rcallon@juniper.net;
> > > > > > > > draft-ietf- l3vpn-end-system@tools.ietf.org;
> > > > > > > > bess-chairs@tools.ietf.org; softwire-
> > > > > > > > chairs@tools.ietf.org
> > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com; Loa Andersson
> > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > >
> > > > > > > > By the way, I think it would be better to allow the BESS
> > > > > > > > and Softwire WG co-chairs to be involved.
> > > > > > > >
> > > > > > > > Best regards,
> > > > > > > > Xiaohu
> > > > > > > >
> > > > > > > > > -----Original Message-----
> > > > > > > > > From: Xuxiaohu
> > > > > > > > > Sent: Thursday, February 12, 2015 9:47 AM
> > > > > > > > > To: 'Black, David'; adrian@olddog.co.uk;
> > > > > > > > > rcallon@juniper.net; 'draft-ietf-l3vpn-end-system@tools.i=
etf.org'
> > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com; 'Loa Andersson'
> > > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > >
> > > > > > > > > (cced to the authors of the end-system draft)
> > > > > > > > >
> > > > > > > > > Hi all,
> > > > > > > > >
> > > > > > > > > I thinks there must be some avoidable mistaken IANA
> > > > > > > > > action
> > request.
> > > > > > > > > The IANA Considerations of the latest version of the
> > > > > > > > > end-system draft
> > > > > > > > > (https://tools.ietf.org/html/draft-ietf-l3vpn-end-system
> > > > > > > > > -0
> > > > > > > > > 4#pa
> > > > > > > > > ge
> > > > > > > > > -2
> > > > > > > > > 1)
> > > > > > > > > which
> > > > > > > > was
> > > > > > > > > published on October 2, 2014 clearly states that " This
> > > > > > > > > document has no IANA actions." Furthermore, the -03
> > > > > > > > > version
> > > > > > > > > (https://tools.ietf.org/html/draft-ietf-l3vpn-end-system
> > > > > > > > > -0
> > > > > > > > > 3) which was published on September 18, 2014 and all the
> > > > > > > > > previous versions didn't mention the BGP tunnel type
> > > > > > > > > matter at all. On the contrary, the BGP tunnel type for
> > > > > > > > > MPLS-in-UDP has been mentioned since the
> > > > > > > > > 00 version of the MPLS-in-UDP draft
> > > > > > > > > (https://tools.ietf.org/html/draft-xu-mpls-in-udp-00#pag
> > > > > > > > > e-
> > > > > > > > > 4) which was published April 28, 2012. However,
> > > > > > > > > According to the WG consensus during the WG adoption
> > > > > > > > > poll period, that section about "Signaling for Encapsulat=
ion in
> UDP"
> > > > > > > > > was removed and accordingly be specified in a separate
> > > > > > > > > draft
> > > > > > > > > (https://tools.ietf.org/html/draft-xu-softwire-encaps-ud
> > > > > > > > > p-
> > > > > > > > > 00) which was published on February 12, 2013.
> > > > > > > > >
> > > > > > > > > Since the WG consensus during the adoption poll of the
> > > > > > > > > MPLS-in-UDP draft is to specify the signaling for
> > > > > > > > > encapsulation in UDP in a separate draft and all the
> > > > > > > > > specifications for the BGP signaling for GRE, IPsec and
> > > > > > > > > etc were all defined in separate drafts belonging to the
> > > > > > > > > Softwire WG, I do believe we should define
> > > > > > > > the
> > > > > > > > > signaling for UDP tunnel in a separate draft belonging
> > > > > > > > > to the Softwire
> > > > > WG.
> > > > > > > > >
> > > > > > > > > Since authors of the end-system draft believe the BGP
> > > > > > > > > tunnel type for MPLS-in-UDP is necessary and the
> > > > > > > > > MPLS-in-UDP draft is going to be published soon, the
> > > > > > > > > normative way is to move forward
> > > > > > > > > draft-xu-softwire-encaps-udp as
> > quickly as possible, IMHO.
> > > > > > > > >
> > > > > > > > > Best regards,
> > > > > > > > > Xiaohu
> > > > > > > > >
> > > > > > > > > > -----Original Message-----
> > > > > > > > > > From: Black, David [mailto:david.black@emc.com]
> > > > > > > > > > Sent: Wednesday, February 11, 2015 9:58 PM
> > > > > > > > > > To: adrian@olddog.co.uk; Xuxiaohu; rcallon@juniper.net
> > > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com; Black, David
> > > > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > > >
> > > > > > > > > > Adrian,
> > > > > > > > > >
> > > > > > > > > > Ok, that's roughly what I expected - between IANA and
> > > > > > > > > > the RFC Editor, the l3vpn-end-system draft will record
> > > > > > > > > > IANA's
> > actions
> > > > here.
> > > > > > > > > >
> > > > > > > > > > I had included you as the (ir)responsible AD for the
> > > > > > > > > > l3vpn-end-system draft, and indeed what is transpiring
> > > > > > > > > > is a version of "ADs can make many
> > > > > > > > > things happen."
> > > > > > > > > >
> > > > > > > > > > The good news is that we don't need another draft to
> > > > > > > > > > allocate that BGP tunnel type code point, which was
> > > > > > > > > > where this whole thread started, so chalk this up as a
> > > > > > > > > > small victory in the never-ending battle to reduce
> > > > > > > > > > IESG
> > > > > > > > > workload ;-).
> > > > > > > > > >
> > > > > > > > > > Alvaro - welcome, and congratulations on your new role!
> > > > > > > > > >
> > > > > > > > > > Thanks,
> > > > > > > > > > --David
> > > > > > > > > >
> > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > From: Adrian Farrel [mailto:adrian@olddog.co.uk]
> > > > > > > > > > > Sent: Wednesday, February 11, 2015 4:53 AM
> > > > > > > > > > > To: Black, David; 'Xuxiaohu'; rcallon@juniper.net
> > > > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com
> > > > > > > > > > > Subject: draft-ietf-l3vpn-end-system and
> > > > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > > > >
> > > > > > > > > > > Hi and sorry,
> > > > > > > > > > >
> > > > > > > > > > > I should have looked more deeply *before* sending my
> > > > > > > > > > > previous
> > > > > email.
> > > > > > > > > > >
> > > > > > > > > > > Here is the resolution to IANA's issue with
> > > > > > > > > > > draft-ietf-l3vpn-end-system that I proposed and they
> accepted.
> > > > > > > > > > >
> > > > > > > > > > > We're just waiting for the authors of
> > > > > > > > > > > draft-ietf-l3vpn-end-system to do something.
> > > > > > > > > > >
> > > > > > > > > > > A
> > > > > > > > > > >
> > > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > > From: Pearl Liang via RT
> > > > > > > > > > > > [mailto:iana-issues@iana.org]
> > > > > > > > > > > > Sent: 09 December 2014 17:40
> > > > > > > > > > > > Cc: thomas.morin@orange.com; adrian@olddog.co.uk;
> > > > > > > > > > > > bess@ietf.org;
> > > > > > > > > > > > draft-ietf- l3vpn-end-system.all@tools.ietf.org
> > > > > > > > > > > > Subject: [IANA #798045] IANA's comments on
> > > > > > > > > > > > draft-ietf-l3vpn-end-system
> > > > > > > > > > > >
> > > > > > > > > > > > Hi Adrian,
> > > > > > > > > > > >
> > > > > > > > > > > > This makes it clear whether or not that assignment
> > > > > > > > > > > > needs to be updated when this draft is approved
> > > > > > > > > > > > for publication as
> > RFC:
> > > > > > > > > > > >
> > > > > > > > > > > > [[[
> > > > > > > > > > > > I think that might be valuable. So the IANA
> > > > > > > > > > > > section should
> > read...
> > > > > > > > > > > >
> > > > > > > > > > > >    IANA has previously made an allocation from the
> > > > > > > > > > > > "BGP Tunnel
> > > > > > > > > > Encapsulation
> > > > > > > > > > > >    Attribute Tunnel Types" registry that reads:
> > > > > > > > > > > >
> > > > > > > > > > > >    Value  | Name                      | Reference
> > > > > > > > > > > >
> > --------+---------------------------+-------------------------------
> > > > > > > > > > > >        13  | MPLS in UDP Encapsulation |
> > > > > > > > > > > > [draft-ietf-l3vpn-end-system]
> > > > > > > > > > > >
> > > > > > > > > > > >    IANA is requested to change the reference to
> > > > > > > > > > > > point to the RFC
> > > > > > > number
> > > > > > > > > > > >    of this document when it is published.
> > > > > > > > > > > >
> > > > > > > > > > > > ]]]
> > > > > > > > > > > >
> > > > > > > > > > > > The current text  "This document has no IANA action=
s."
> > > > > > > > > > > > provides no instructions and incorrectly tell
> > > > > > > > > > > > people there is no actions
> > > > > > > requested.
> > > > > > > > > > > >
> > > > > > > > > > > > Thanks
> > > > > > > > > > > > ~pl
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > On Tue Dec 09 13:20:57 2014, adrian@olddog.co.uk wr=
ote:
> > > > > > > > > > > > > Hi,
> > > > > > > > > > > > >
> > > > > > > > > > > > > Replying to myself and keeping the same IANA
> > > > > > > > > > > > > tracking
> > > number.
> > > > > > > > > > > > >
> > > > > > > > > > > > > > > IESG/Authors/WG Chairs:
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > IANA has reviewed draft-ietf-l3vpn-end-system=
-04.
> > > > > > > > > > > > > > > Authors should
> > > > > > > > > > > > > review
> > > > > > > > > > > > > > > the comments and/or questions below.  Please
> > > > > > > > > > > > > > > report any
> > > > > > > > > > > > > inaccuracies and
> > > > > > > > > > > > > > > respond to any questions as soon as possible.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > IANA's reviewer has the following
> comments/questions:
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > IANA has a question about the IANA
> > > > > > > > > > > > > > > Considerations section of this
> > > > > > > > > > > > > document.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Previously, an early assignment has been
> > > > > > > > > > > > > > > made to support this
> > > > > > > > > > > > > draft. The
> > > > > > > > > > > > > > > original request for an assignment is below:
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > >> <begin request=3D""> Contact Name:
> > > > > > > > > > > > > > >> Thomas Morin
> > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > >> Contact Email:
> > > > > > > > > > > > > > >> thomas.morin@orange.com
> > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > >> Type of Assignment:
> > > > > > > > > > > > > > >> Assignement of a BGP parameter in a FCFS reg=
istry.
> > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > >> Registry:
> > > > > > > > > > > > > > >> BGP Tunnel Encapsulation Attribute Tunnel
> > > > > > > > > > > > > > >> Types
> > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > >> See:
> > > > > > > > > > > > > > >> https://www.iana.org/assignments/bgp-parame
> > > > > > > > > > > > > > >> te
> > > > > > > > > > > > > > >> rs
> > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > >> Description:
> > > > > > > > > > > > > > >> Needed for draft-ietf-l3vpn-end-system, to
> > > > > > > > > > > > > > >> allow the use of an MPLS-over-UDP
> > > > > > > > > > > > > > >> encapsulation as specified in
> > > > > > > > > > > > > > >> draft-ietf-mpls-in-
> > > > > > > > > > > > > udp .
> > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > >> No value has been proposed yet, next
> > > > > > > > > > > > > > >> available value
> > > > > > > > > > > > > > >> 13 would be
> > > > > > > > > > > > > fine.
> > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > >> Additional Info:
> > > > > > > > > > > > > > >> draft-ietf-l3vpn-end-system </end>
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > IANA Question --> The IANA Considerations
> > > > > > > > > > > > > > > section said "This
> > > > > > > > > > > > > document has
> > > > > > > > > > > > > > > no IANA actions."  and, as a result, the
> > > > > > > > > > > > > > > assignment made through
> > > > > > > > > > > > > the request
> > > > > > > > > > > > > > > above would not be made permanent. Is this
> > > > > > > > > > > > > > > the author's intent? If
> > > > > > > > > > > > > not,
> > > > > > > > > > > > > could
> > > > > > > > > > > > > > > the draft be revised to indicate that the
> > > > > > > > > > > > > > > assignment made based on
> > > > > > > > > > > > > the
> > > > > > > > > > > > > request
> > > > > > > > > > > > > > > above be changed from an initial assignment
> > > > > > > > > > > > > > > to a permanent
> > > > > > > > > > > > > assignment.
> > > > > > > > > > > > >
> > > > > > > > > > > > > How do you mean?
> > > > > > > > > > > > > The registry is FCFS for which *any* document is
> > sufficient.
> > > > > > > > > > > > > The assignment has been made and is as permanent
> > > > > > > > > > > > > as any FCFS assignment ever is.
> > > > > > > > > > > > >
> > > > > > > > > > > > > > > Please note that IANA cannot reserve specific=
 values.
> > > > > > > > > > > > > > > However,
> > > > > > > > > > > > > early
> > > > > > > > > > > > > > > allocation is available for some types of
> > registrations.
> > > > > > > > > > > > > > > For more
> > > > > > > > > > > > > information,
> > > > > > > > > > > > > > > please see RFC 7120.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Yes, but this is a FCFS registry to which 7120
> > > > > > > > > > > > > does not apply, and nor does "reservation of valu=
es".
> > > > > > > > > > > > > With FCFS the value is assigned when requested
> > > > > > > > > > > > > and that's
> > it.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Now, it is a different question whether this
> > > > > > > > > > > > > document should ask for the registry to be
> > > > > > > > > > > > > updated to point to the consequent RFC instead of=
 the I-D.
> > > > > > > > > > > > >
> > > > > > > > > > > > > I think that might be valuable. So the IANA
> > > > > > > > > > > > > section should
> > > > read...
> > > > > > > > > > > > >
> > > > > > > > > > > > >    IANA has previously made an allocation from
> > > > > > > > > > > > > the "BGP Tunnel Encapsulation
> > > > > > > > > > > > >    Attribute Tunnel Types" registry that reads:
> > > > > > > > > > > > >
> > > > > > > > > > > > >    Value  | Name                      | Reference
> > > > > > > > > > > > >    --------+---------------------------
> > > > > > > > > > > > > +-------------------------------
> > > > > > > > > > > > >        13  | MPLS in UDP Encapsulation |
> > > > > > > > > > > > > [draft-ietf-l3vpn-end-system]
> > > > > > > > > > > > >
> > > > > > > > > > > > >    IANA is requested to change the reference to
> > > > > > > > > > > > > point to the RFC number
> > > > > > > > > > > > >    of this document when it is published.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Cheers,
> > > > > > > > > > > > > Adrian
> > > > > > > > > > >
> > > > > > >
> > > > >
> > > > > _______________________________________________
> > > > > BESS mailing list
> > > > > BESS@ietf.org
> > > > > https://www.ietf.org/mailman/listinfo/bess
> > >
> > > _______________________________________________
> > > BESS mailing list
> > > BESS@ietf.org
> > > https://www.ietf.org/mailman/listinfo/bess
> >
> > _______________________________________________
> > BESS mailing list
> > BESS@ietf.org
> > https://www.ietf.org/mailman/listinfo/bess


From nobody Tue Mar  3 05:13:59 2015
Return-Path: <adrian@olddog.co.uk>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 96BA11A6FF7 for <bess@ietfa.amsl.com>; Tue,  3 Mar 2015 05:13:58 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.6
X-Spam-Level: 
X-Spam-Status: No, score=-102.6 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YymB2yCS5vK7 for <bess@ietfa.amsl.com>; Tue,  3 Mar 2015 05:13:52 -0800 (PST)
Received: from asmtp1.iomartmail.com (asmtp1.iomartmail.com [62.128.201.248]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 162E61A6F7C for <bess@ietf.org>; Tue,  3 Mar 2015 05:13:51 -0800 (PST)
Received: from asmtp1.iomartmail.com (localhost.localdomain [127.0.0.1]) by asmtp1.iomartmail.com (8.13.8/8.13.8) with ESMTP id t23DDjr7003037; Tue, 3 Mar 2015 13:13:45 GMT
Received: from 950129200 (089144234089.atnat0043.highway.a1.net [89.144.234.89]) (authenticated bits=0) by asmtp1.iomartmail.com (8.13.8/8.13.8) with ESMTP id t23DDhcC003014 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=NO); Tue, 3 Mar 2015 13:13:44 GMT
From: "Adrian Farrel" <adrian@olddog.co.uk>
To: "'Xuxiaohu'" <xuxiaohu@huawei.com>, <bess@ietf.org>
References: <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE083064D3@NKGEML512-MBS.china.huawei.com> <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE083066B6@NKGEML512-MBS.china.huawei.com> <00d801d048ba$e4eb09b0$aec11d10$@olddog.co.uk> <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE0830EF91@NKGEML512-MBS.china.huawei.com>
In-Reply-To: <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE0830EF91@NKGEML512-MBS.china.huawei.com>
Date: Tue, 3 Mar 2015 13:13:46 -0000
Message-ID: <031601d055b3$e283cda0$a78b68e0$@olddog.co.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQNXzOpNxBkaCSTp4d7RfEjKii6N5wE0fVnkAf/UiHoB8RG7pZnSogTw
Content-Language: en-gb
X-TM-AS-MML: disable
X-TM-AS-Product-Ver: IMSS-7.1.0.1576-7.5.0.1018-21368.007
X-TM-AS-Result: No--35.891-10.0-31-10
X-imss-scan-details: No--35.891-10.0-31-10
X-TMASE-MatchedRID: jFqw+1pFnMzW/bDrA6VrLfzu9Lw9C7fAUAjrAJWsTe/JYIv7y0tu9q7m YDa7MoGalhR/IwJ8LR2vh3ewfsNrXNgwSqFJytAFQ1OcCEvT+be8Phgqj1K23ADqzaYhcjeQvL4 BuAuBsSmUH8OkPUsxAfFPDHhQkgxy5Ls9mw/0hkGVUcz8XpiS9BSRa9qpSosfimI35BFBfTlH+B /ecAqGmbhInuDKffiIUOBwG9PYW5EUgj4BME3CEu9VsdrlGzy3MdC/JSQ0MpGe9toQ6h6LE7Pxi a6nVyzwuowu8uR80H/VwfmM7kNGWKuThJWNM5SqU+OjsPhIWDhKKnAlDnKDuiTiYfGXnq6hIBX2 rYVgtOfvt3nUS5MECXSmZByVeIeXds1VAj9mPJ0wiJTf3kjwfaHXXqNw58ALnLVhzy0+RX2xZmm VcDDWJ1DbG7RccYzQfdTAWJ8WVbvdKXydXwjND/OHbIp2eXtYI6PHNDZGGCLxH97w0zz7de+/8g DYc+T4yobudGO4ujGbOEOnvX9KDXay9L2QAf8Y5Lm3wIkdqWMmOHJ0aBcO1AQLUvnBZL2KOXUFK +IJNsLijYHtimVeHtVMl411STI/xzvHgMQjo/YC3Zc9wAelquSBrGHYyfa6eNDQ2odiEiFTxL79 8Vu5k1QmQ4oHEtT5Jm73Sy88oZr6SrizO3rf+YWs+rMQjb4f0vNkFv6xUCNX14Hy+eYp7+J1fnQ QXoSOYRx2gVSq/vIKC+uoO2oeiBdr6zI6nTrteCXTmxf351KNxR0omuRmzcUmcSma304Tdh1a52 zlmP9WP8b5kLaf3tsG2+F/ohOmvoOwdMywpfOeAiCmPx4NwLTrdaH1ZWqCpvI8UZOf47hck9lvx NcntccVmXYwAkhKseWplitmp0j6C0ePs7A07QKmARN5PTKc
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/wMIGLq3xIkxKPMuZ8cjSIzh8f58>
Subject: Re: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE: Why transform draft-xu-softwire-encaps-udp to	draft-xu-bess-encaps-udp
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: adrian@olddog.co.uk
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 03 Mar 2015 13:13:58 -0000

OK,

So you want an I-D that describes how to use the code point to achieve a variety
of things. That is potentially useful, but I haven't read that text in your I-D.

But please understand that the code point is already allocated (and I shouldn't
be surprised to hear it is coded and deployed). So it is way too late to say
that the wrong document asked for the code point.

Adrian

> -----Original Message-----
> From: Xuxiaohu [mailto:xuxiaohu@huawei.com]
> Sent: 02 March 2015 10:07
> To: Xuxiaohu; adrian@olddog.co.uk; bess@ietf.org
> Subject: RE: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE:
> Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
> 
> Besides the reasons as mentioned below, since the BGP Encapsulation SAFI for
> UDP or MPLS-in-UDP is not only applicable to BGP/MPLS-based L2VPN (including
> VPLS and EVPN) and BGP/MPLS-based L3VPN, but also be applicable to other use
> cases (e.g., incremental deployment of BGP-based Segment Routing), it seems
> better to specify it in a separate doc rather than in the end-system-l3vpn
draft
> (http://datatracker.ietf.org/doc/draft-ietf-l3vpn-end-system/) which is
> dedicated for how to use XMPP as a replacement of the BGP-L3VPN signaling,
> IMHO.
> 
> Best regards,
> Xiaohu
> 
> > -----Original Message-----
> > From: Xuxiaohu
> > Sent: Sunday, February 15, 2015 11:31 AM
> > To: 'adrian@olddog.co.uk'; bess@ietf.org
> > Subject: RE: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE:
> > Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
> >
> > Hi Adrian,
> >
> > > -----Original Message-----
> > > From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Adrian Farrel
> > > Sent: Sunday, February 15, 2015 9:01 AM
> > > To: bess@ietf.org
> > > Subject: Re: [bess] Request for WG adoption of draft-xu-bess-encaps-
> udp//RE:
> > > Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
> > >
> > > Just for clarity on this point:
> > >
> > > I suggested that BESS was the right place to discuss whether and how
> > > to use BGP to indicate tunnel types.
> >
> > Thanks for making that suggestion. Although RFC5512 (i.e.,
> > draft-ietf-softwire-encaps-safi) and RFC5566 (i.e.,
> > draft-ietf-softwire-encaps-ipsec) have been originated from Softwire WG and
> > the current Softwire WG charter
(https://tools.ietf.org/wg/softwire/charters)
> > still state that "BGP and other routing and signaling protocols developed in
this
> > group will be reviewed jointly with the proper working groups and other
> > workings that may take interest (e.g. IDR, L3VPN, PIM, LDP, SAAG, etc)",
your
> > above suggestion, together with Softwire co-chairs' claim of last Friday
that the
> > softwire WG is going to shut down and therefore it would not be a right
place to
> > pursue this draft anymore, seem to be a joint statement that any future work
> > related to BGP Encapsulation SAFI should be discussed in the BESS WG, rather
> > than the Softwire WG. Thanks again for clarifying the confusion.
> >
> > > I also observed that a code point already exists in the BGP Tunnel
> > > Encapsulation Attribute Tunnel Types registry at
> > > http://www.iana.org/assignments/bgp-parameters/bgp-
> parameters.xhtml#tu
> > > n
> > > nel-types
> >
> > > Therefore the debate the WG needs to have (perhaps before debating
> > > adoption of this document) is:
> > >
> > > - whether the existing code point is enough for the purpose of identifying
> > >   MPLS-in-UDP tunnels or whether more work is needed
> > > - whether a foo-in-UDP tunnel type should be used instead with sub-TLVs
> > >   to identify the different cases of foo.
> >
> > IMHO, it depends on whether the tunnel type of the Tunnel Encapsulation TLV
> as
> > specified in RFC5512 should be interpreted as UDP or MPLS-in-UDP.
> >
> > See the following text quoted from RFC 5512
(draft-ietf-softwire-encaps-safi):
> >
> > ****************
> > 4.2.  Protocol Type Sub-TLV
> >
> >    The protocol type sub-TLV MAY be encoded to indicate the type of the
> >    payload packets that will be encapsulated with the tunnel parameters
> >    that are being signaled in the TLV.  The value field of the sub-TLV
> >    contains a 2-octet protocol type that is one of the types defined in
> >    [IANA-AF] as ETHER TYPEs.
> >
> >    For example, if we want to use three L2TPv3 sessions, one carrying
> >    IPv4 packets, one carrying IPv6 packets, and one carrying MPLS
> >    packets, the egress router will include three TLVs of L2TPv3
> >    encapsulation type, each specifying a different Session ID and a
> >    different payload type.  The protocol type sub-TLV for these will be
> >    IPv4 (protocol type = 0x0800), IPv6 (protocol type = 0x86dd), and
> >    MPLS (protocol type = 0x8847), respectively.
> >
> > 4.4.  Tunnel Type Selection
> >
> >    A BGP speaker may include multiple tunnel TLVs in the tunnel
> >    attribute.  The receiving speaker MAY have local policies defined to
> >    choose different tunnel types for different sets/types of payload
> >    prefixes received from the same BGP speaker.  For instance, if a BGP
> >    speaker includes both L2TPv3 and GRE tunnel types in the tunnel
> >    attribute and it also advertises IPv4 and IPv6 prefixes, the ingress
> >    router may have local policy defined to choose L2TPv3 for IPv4
> >    prefixes (provided the protocol type received in the tunnel attribute
> >    matches) and GRE for IPv6 prefixes.
> >
> > ***************
> >
> > To keep it in accordance with the usage of the tunnel type as specified in
> > RFC5512, I believe the tunnel type of the Tunnel Encapsulation TLV should be
> > interpreted as UDP, rather than MPLS-in-UDP. That's to say, what type of
> > payload will be carried over the UDP tunnel cloud be explicitly indicated by
the
> > protocol type sub-TLV.
> >
> > Best regards,
> > Xiaohu
> >
> > > Adrian
> > >
> > > > -----Original Message-----
> > > > From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Xuxiaohu
> > > > Sent: 13 February 2015 06:10
> > > > To: Xuxiaohu; bess@ietf.org; bess-chairs@tools.ietf.org
> > > > Cc: draft-xu-bess-encaps-udp@tools.ietf.org;
> > > > softwire-chairs@tools.ietf.org
> > > > Subject: [bess] Request for WG adoption of
> > > > draft-xu-bess-encaps-udp//RE: Why transform
> > > > draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
> > > >
> > > > Hi BESS WG co-chairs,
> > > >
> > > > The BGP tunnel type for MPLS-in-UDP has been mentioned in the
> > > > MPLS-in-UDP draft since the 00 version
> > > > (https://tools.ietf.org/html/draft-xu-mpls-in-udp-
> > > > 00#page-4) which was published on April 28, 2012. However, according
> > > > to the WG consensus during the WG adoption poll period, that section
> > > > of "Signaling for Encapsulation in UDP" was removed from the
> > > > MPLS-in-UDP draft and accordingly it was specified in a separate
> > > > draft
> > > (https://tools.ietf.org/html/draft-xu-softwire-
> > > > encaps-udp-00) which was published on February 12, 2013.
> > > >
> > > > Adrian has suggested me to post this draft to BESS. Furthermore,
> > > > Suresh has indicated that the Softwire WG would not be a right place
> > > > for any BGP tunnel type related works anymore since the WG is going
> > > > to shut down in the very near future. Hence, would you please start
> > > > a WG adoption for draft-xu-bess-encaps- udp which is transformed
> > > > from
> > > draft-xu-softwire-encaps-udp?
> > > >
> > > > Best regards,
> > > > Xiaohu
> > > >
> > > > > -----Original Message-----
> > > > > From: Xuxiaohu
> > > > > Sent: Friday, February 13, 2015 12:13 PM
> > > > > To: Xuxiaohu; bess@ietf.org
> > > > > Cc: Softwires WG
> > > > > Subject: RE: [bess] Why transform draft-xu-softwire-encaps-udp to
> > > > > draft-xu-bess-encaps-udp
> > > > >
> > > > > Hi all,
> > > > >
> > > > > I noticed the following text from RFC 5512
> > > (draft-ietf-softwire-encaps-safi):
> > > > >
> > > > > ****************
> > > > > 4.2.  Protocol Type Sub-TLV
> > > > >
> > > > >    The protocol type sub-TLV MAY be encoded to indicate the type of
the
> > > > >    payload packets that will be encapsulated with the tunnel
parameters
> > > > >    that are being signaled in the TLV.  The value field of the sub-TLV
> > > > >    contains a 2-octet protocol type that is one of the types defined
in
> > > > >    [IANA-AF] as ETHER TYPEs.
> > > > >
> > > > >    For example, if we want to use three L2TPv3 sessions, one carrying
> > > > >    IPv4 packets, one carrying IPv6 packets, and one carrying MPLS
> > > > >    packets, the egress router will include three TLVs of L2TPv3
> > > > >    encapsulation type, each specifying a different Session ID and a
> > > > >    different payload type.  The protocol type sub-TLV for these will
be
> > > > >    IPv4 (protocol type = 0x0800), IPv6 (protocol type = 0x86dd), and
> > > > >    MPLS (protocol type = 0x8847), respectively.
> > > > > ***************
> > > > >
> > > > > It seems that RFC5512 is not only talking about IP-in-GRE, but
> > > > > also
> > > MPLS-in-GRE.
> > > > >
> > > > > I also noticed an expired IDR WG doc (see
> > > > > https://tools.ietf.org/html/draft-ietf-idr-encaps-safi-00) which
> > > > > was a predecessor of draft-ietf-softwire-encaps-safi. Besides,
> > > > > RFC5566
> > > > > (draft-ietf-softwire-encaps-ipsec) is also originated from the
Softwire WG.
> > > > Does
> > > > > it mean which WG should be responsible for the BGP Tunnel
> > > > > Encapsulation Attribute related work had ever been discussed and
> > > > > finally determined that
> > > the
> > > > > Softwire WG is the right place for it?
> > > > >
> > > > > Best regards,
> > > > > Xiaohu
> > > > >
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Xuxiaohu
> > > > > > Sent: Friday, February 13, 2015 9:21 AM
> > > > > > To: bess@ietf.org
> > > > > > Cc: Softwires WG
> > > > > > Subject: [bess] Why transform draft-xu-softwire-encaps-udp to
> > > > > > draft-xu-bess-encaps-udp
> > > > > >
> > > > > > Hi all,
> > > > > >
> > > > > > According to the suggestion from Adrian as a Routing co-AD,
> > > > > > draft-xu-softwire-encaps-udp
> > > > > > (https://tools.ietf.org/html/draft-xu-softwire-encaps-udp) which
> > > > > > was posted to the Softwire WG is now posted to the BESS WG.
> > > > > >
> > > > > > Any comments and suggestions are welcome.
> > > > > >
> > > > > > Best regards,
> > > > > > Xiaohu
> > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: Xuxiaohu
> > > > > > > Sent: Friday, February 13, 2015 8:45 AM
> > > > > > > To: 'adrian@olddog.co.uk'; 'Black, David';
> > > > > > > rcallon@juniper.net;
> > > > > > > draft-ietf-l3vpn-end-system@tools.ietf.org;
> > > > > > > bess-chairs@tools.ietf.org; softwire-chairs@tools.ietf.org
> > > > > > > Cc: 'Alvaro Retana'; akatlas@gmail.com; 'Loa Andersson'
> > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > draft-ietf-mpls-in-udp
> > > > > > >
> > > > > > > Hi Adrian,
> > > > > > >
> > > > > > > Thanks a lot for your response. Although RFC5512 (i.e.,
> > > > > > > draft-ietf-softwire-encaps-safi) and RFC5566 (i.e.,
> > > > > > > draft-ietf-softwire-encaps-ipsec) which specify the BGP Tunnel
> > > > > > > Encapsulation Attribute Tunnel Types for GRE, L2TPv3 and IPsec
> > > > > > > respectively are all originated from Softwire, and further the
> > > > > > > Softwire WG co-chairs didn't state that
> > > > > > > draft-xu-softwire-encaps-udp doesn't belong to their WG, if
> > > > > > > the BESS and Softwire WG co-chairs could reach an agreement
> > > > > > > that any future work related to BGP Tunnel Encapsulation
> > > > > > > Attribute should be done in the BESS WG, it looks fine to me.
> > > > > > > I would submit the same draft to the BESS WG as
> > > > > > draft-xu-bess-encaps-udp.
> > > > > > >
> > > > > > > Best regards,
> > > > > > > Xiaohu
> > > > > > >
> > > > > > > > -----Original Message-----
> > > > > > > > From: Adrian Farrel [mailto:adrian@olddog.co.uk]
> > > > > > > > Sent: Thursday, February 12, 2015 10:50 PM
> > > > > > > > To: Xuxiaohu; 'Black, David'; rcallon@juniper.net;
> > > > > > > > draft-ietf-l3vpn-end-system@tools.ietf.org;
> > > > > > > > bess-chairs@tools.ietf.org; softwire-chairs@tools.ietf.org
> > > > > > > > Cc: 'Alvaro Retana'; akatlas@gmail.com; 'Loa Andersson'
> > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > >
> > > > > > > > Hello all,
> > > > > > > >
> > > > > > > > 1. Why softwire? That is strictly IP-in-IP with a particular
> > > > > > > > intention of 4-over-6 and 6-over-4. Why would MPLS-in-UDP
> > > > > > > > fall into their
> > > > > > charter?
> > > > > > > > You say "all the specifications for the BGP signaling for
> > > > > > > > GRE, IPsec and etc were all defined in separate drafts
> > > > > > > > belonging to the Softwire WG" but I see no evidence of this.
> > > > > > > > The only vaguely related draft I can see is
> > > > > > > > draft-xu-softwire-ip-in-udp which is a specific
> > > > > > > > IP-over-UDP-over-IP mechanism about which I will reserve
> > > > > > > > judgement except to say that I that softwire really needs
> > > > > > > > yet another transition mechanism and that I believe IP-in-IP
> > > > > > > > can be hashed by existing ECMP
> > > > > > > hardware.
> > > > > > > >
> > > > > > > > You also referenced draft-xu-softwire-encaps-udp but I
> > > > > > > > believe this document expired over 12 months ago. I would
> > > > > > > > not say that it was the most substantive or technical
> > > > > > > > document I have ever read
> > > > > > > > :-)
> > > > > > > >
> > > > > > > > I have not removed the chairs from this thread, but I really
> > > > > > > > hate spamming people's in-boxes.
> > > > > > > >
> > > > > > > > 2. While Xiaohu has correctly pointed at the current version
> > > > > > > > of the I-D, it might be better to look at the status in the
> > > > > > > > Datatracker via
> > > > > > > > http://datatracker.ietf.org/doc/draft-ietf-l3vpn-end-system/
> > > > > > > > - You'll see that the status is Waiting for AD
> > > > > > > > Go-Ahead::Revised I-D Needed which means it has completed
> > > > > > > > IETF last call and is waiting for a
> > > > > > > revision.
> > > > > > > >
> > > > > > > > 3. If you are following the BESS mailing list, you'll see
> > > > > > > > that there is text agreed with IANA to fix the "empty" IANA
> > > > > > > > considerations
> > > > > section.
> > > > > > > > http://www.ietf.org/mail-archive/web/bess/current/msg00233.h
> > > > > > > > tm l This will be in the next revision of the draft.
> > > > > > > >
> > > > > > > > 4. I am sure we can involve the BESS chairs any time we note
> > > > > > > > some work that they need to do. At the moment, they may be
> > > > > > > > interested to know there is a conversation, but I don't know
> > > > > > > > that we have identified any actions for them. I have not
> > > > > > > > removed them from this thread, but I really hate spamming
> > > > > > > > people's
> > > in-boxes.
> > > > > > > >
> > > > > > > >
> > > > > > > > I believe there are two pieces of work:
> > > > > > > >
> > > > > > > > A. Assign a BGP Tunnel Encapsulation Attribute Tunnel Type.
> > > > > > > > This has already been done. No amount of effort to change
> > > > > > > > documents or advance one document or another will change
> > > > > > > > this fact. The code point has already been assigned. The
> > > > > > > > registry is "First Come First Served" and no particular
> > > > > > > > process was required except an application to IANA. No
> > > > > > > > further
> > > > > > > action is desirable.
> > > > > > > >
> > > > > > > > B. Specify necessary protocol work to utilise this code point.
> > > > > > > > This is a matter for the BESS WG. They may consider that
> > > > > > > > everything needed has already been documented, they may
> > > > > > > > consider that they do not want to specify anything, they may
> > > > > > > > consider that further work is needed and can be based on
> > > > > > > > your I-D, they may consider that further work is needed and
> > > > > > > > can needs a different starting point. The correct way to
> > > > > > > > handle this is to post your I-D and take the discussion to
> > > > > > > > the BESS
> > > > > > mailing list.
> > > > > > > You may ask the BESS chairs for advice.
> > > > > > > >
> > > > > > > >
> > > > > > > > What am I missing here?
> > > > > > > > What do you want to achieve and why?
> > > > > > > >
> > > > > > > > What action are you actually asking for?
> > > > > > > >
> > > > > > > > Adrian
> > > > > > > >
> > > > > > > > > -----Original Message-----
> > > > > > > > > From: Xuxiaohu [mailto:xuxiaohu@huawei.com]
> > > > > > > > > Sent: 12 February 2015 05:56
> > > > > > > > > To: Xuxiaohu; Black, David; adrian@olddog.co.uk;
> > > > > > > > > rcallon@juniper.net;
> > > > > > > > > draft-ietf- l3vpn-end-system@tools.ietf.org;
> > > > > > > > > bess-chairs@tools.ietf.org; softwire-
> > > > > > > > > chairs@tools.ietf.org
> > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com; Loa Andersson
> > > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > >
> > > > > > > > > By the way, I think it would be better to allow the BESS
> > > > > > > > > and Softwire WG co-chairs to be involved.
> > > > > > > > >
> > > > > > > > > Best regards,
> > > > > > > > > Xiaohu
> > > > > > > > >
> > > > > > > > > > -----Original Message-----
> > > > > > > > > > From: Xuxiaohu
> > > > > > > > > > Sent: Thursday, February 12, 2015 9:47 AM
> > > > > > > > > > To: 'Black, David'; adrian@olddog.co.uk;
> > > > > > > > > > rcallon@juniper.net; 'draft-ietf-l3vpn-end-
> system@tools.ietf.org'
> > > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com; 'Loa Andersson'
> > > > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > > >
> > > > > > > > > > (cced to the authors of the end-system draft)
> > > > > > > > > >
> > > > > > > > > > Hi all,
> > > > > > > > > >
> > > > > > > > > > I thinks there must be some avoidable mistaken IANA
> > > > > > > > > > action
> > > request.
> > > > > > > > > > The IANA Considerations of the latest version of the
> > > > > > > > > > end-system draft
> > > > > > > > > > (https://tools.ietf.org/html/draft-ietf-l3vpn-end-system
> > > > > > > > > > -0
> > > > > > > > > > 4#pa
> > > > > > > > > > ge
> > > > > > > > > > -2
> > > > > > > > > > 1)
> > > > > > > > > > which
> > > > > > > > > was
> > > > > > > > > > published on October 2, 2014 clearly states that " This
> > > > > > > > > > document has no IANA actions." Furthermore, the -03
> > > > > > > > > > version
> > > > > > > > > > (https://tools.ietf.org/html/draft-ietf-l3vpn-end-system
> > > > > > > > > > -0
> > > > > > > > > > 3) which was published on September 18, 2014 and all the
> > > > > > > > > > previous versions didn't mention the BGP tunnel type
> > > > > > > > > > matter at all. On the contrary, the BGP tunnel type for
> > > > > > > > > > MPLS-in-UDP has been mentioned since the
> > > > > > > > > > 00 version of the MPLS-in-UDP draft
> > > > > > > > > > (https://tools.ietf.org/html/draft-xu-mpls-in-udp-00#pag
> > > > > > > > > > e-
> > > > > > > > > > 4) which was published April 28, 2012. However,
> > > > > > > > > > According to the WG consensus during the WG adoption
> > > > > > > > > > poll period, that section about "Signaling for Encapsulation
in
> > UDP"
> > > > > > > > > > was removed and accordingly be specified in a separate
> > > > > > > > > > draft
> > > > > > > > > > (https://tools.ietf.org/html/draft-xu-softwire-encaps-ud
> > > > > > > > > > p-
> > > > > > > > > > 00) which was published on February 12, 2013.
> > > > > > > > > >
> > > > > > > > > > Since the WG consensus during the adoption poll of the
> > > > > > > > > > MPLS-in-UDP draft is to specify the signaling for
> > > > > > > > > > encapsulation in UDP in a separate draft and all the
> > > > > > > > > > specifications for the BGP signaling for GRE, IPsec and
> > > > > > > > > > etc were all defined in separate drafts belonging to the
> > > > > > > > > > Softwire WG, I do believe we should define
> > > > > > > > > the
> > > > > > > > > > signaling for UDP tunnel in a separate draft belonging
> > > > > > > > > > to the Softwire
> > > > > > WG.
> > > > > > > > > >
> > > > > > > > > > Since authors of the end-system draft believe the BGP
> > > > > > > > > > tunnel type for MPLS-in-UDP is necessary and the
> > > > > > > > > > MPLS-in-UDP draft is going to be published soon, the
> > > > > > > > > > normative way is to move forward
> > > > > > > > > > draft-xu-softwire-encaps-udp as
> > > quickly as possible, IMHO.
> > > > > > > > > >
> > > > > > > > > > Best regards,
> > > > > > > > > > Xiaohu
> > > > > > > > > >
> > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > From: Black, David [mailto:david.black@emc.com]
> > > > > > > > > > > Sent: Wednesday, February 11, 2015 9:58 PM
> > > > > > > > > > > To: adrian@olddog.co.uk; Xuxiaohu; rcallon@juniper.net
> > > > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com; Black, David
> > > > > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > > > >
> > > > > > > > > > > Adrian,
> > > > > > > > > > >
> > > > > > > > > > > Ok, that's roughly what I expected - between IANA and
> > > > > > > > > > > the RFC Editor, the l3vpn-end-system draft will record
> > > > > > > > > > > IANA's
> > > actions
> > > > > here.
> > > > > > > > > > >
> > > > > > > > > > > I had included you as the (ir)responsible AD for the
> > > > > > > > > > > l3vpn-end-system draft, and indeed what is transpiring
> > > > > > > > > > > is a version of "ADs can make many
> > > > > > > > > > things happen."
> > > > > > > > > > >
> > > > > > > > > > > The good news is that we don't need another draft to
> > > > > > > > > > > allocate that BGP tunnel type code point, which was
> > > > > > > > > > > where this whole thread started, so chalk this up as a
> > > > > > > > > > > small victory in the never-ending battle to reduce
> > > > > > > > > > > IESG
> > > > > > > > > > workload ;-).
> > > > > > > > > > >
> > > > > > > > > > > Alvaro - welcome, and congratulations on your new role!
> > > > > > > > > > >
> > > > > > > > > > > Thanks,
> > > > > > > > > > > --David
> > > > > > > > > > >
> > > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > > From: Adrian Farrel [mailto:adrian@olddog.co.uk]
> > > > > > > > > > > > Sent: Wednesday, February 11, 2015 4:53 AM
> > > > > > > > > > > > To: Black, David; 'Xuxiaohu'; rcallon@juniper.net
> > > > > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com
> > > > > > > > > > > > Subject: draft-ietf-l3vpn-end-system and
> > > > > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > > > > >
> > > > > > > > > > > > Hi and sorry,
> > > > > > > > > > > >
> > > > > > > > > > > > I should have looked more deeply *before* sending my
> > > > > > > > > > > > previous
> > > > > > email.
> > > > > > > > > > > >
> > > > > > > > > > > > Here is the resolution to IANA's issue with
> > > > > > > > > > > > draft-ietf-l3vpn-end-system that I proposed and they
> > accepted.
> > > > > > > > > > > >
> > > > > > > > > > > > We're just waiting for the authors of
> > > > > > > > > > > > draft-ietf-l3vpn-end-system to do something.
> > > > > > > > > > > >
> > > > > > > > > > > > A
> > > > > > > > > > > >
> > > > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > > > From: Pearl Liang via RT
> > > > > > > > > > > > > [mailto:iana-issues@iana.org]
> > > > > > > > > > > > > Sent: 09 December 2014 17:40
> > > > > > > > > > > > > Cc: thomas.morin@orange.com; adrian@olddog.co.uk;
> > > > > > > > > > > > > bess@ietf.org;
> > > > > > > > > > > > > draft-ietf- l3vpn-end-system.all@tools.ietf.org
> > > > > > > > > > > > > Subject: [IANA #798045] IANA's comments on
> > > > > > > > > > > > > draft-ietf-l3vpn-end-system
> > > > > > > > > > > > >
> > > > > > > > > > > > > Hi Adrian,
> > > > > > > > > > > > >
> > > > > > > > > > > > > This makes it clear whether or not that assignment
> > > > > > > > > > > > > needs to be updated when this draft is approved
> > > > > > > > > > > > > for publication as
> > > RFC:
> > > > > > > > > > > > >
> > > > > > > > > > > > > [[[
> > > > > > > > > > > > > I think that might be valuable. So the IANA
> > > > > > > > > > > > > section should
> > > read...
> > > > > > > > > > > > >
> > > > > > > > > > > > >    IANA has previously made an allocation from the
> > > > > > > > > > > > > "BGP Tunnel
> > > > > > > > > > > Encapsulation
> > > > > > > > > > > > >    Attribute Tunnel Types" registry that reads:
> > > > > > > > > > > > >
> > > > > > > > > > > > >    Value  | Name                      | Reference
> > > > > > > > > > > > >
> > > --------+---------------------------+-------------------------------
> > > > > > > > > > > > >        13  | MPLS in UDP Encapsulation |
> > > > > > > > > > > > > [draft-ietf-l3vpn-end-system]
> > > > > > > > > > > > >
> > > > > > > > > > > > >    IANA is requested to change the reference to
> > > > > > > > > > > > > point to the RFC
> > > > > > > > number
> > > > > > > > > > > > >    of this document when it is published.
> > > > > > > > > > > > >
> > > > > > > > > > > > > ]]]
> > > > > > > > > > > > >
> > > > > > > > > > > > > The current text  "This document has no IANA actions."
> > > > > > > > > > > > > provides no instructions and incorrectly tell
> > > > > > > > > > > > > people there is no actions
> > > > > > > > requested.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Thanks
> > > > > > > > > > > > > ~pl
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > > > On Tue Dec 09 13:20:57 2014, adrian@olddog.co.uk
wrote:
> > > > > > > > > > > > > > Hi,
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Replying to myself and keeping the same IANA
> > > > > > > > > > > > > > tracking
> > > > number.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > IESG/Authors/WG Chairs:
> > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > IANA has reviewed
draft-ietf-l3vpn-end-system-04.
> > > > > > > > > > > > > > > > Authors should
> > > > > > > > > > > > > > review
> > > > > > > > > > > > > > > > the comments and/or questions below.  Please
> > > > > > > > > > > > > > > > report any
> > > > > > > > > > > > > > inaccuracies and
> > > > > > > > > > > > > > > > respond to any questions as soon as possible.
> > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > IANA's reviewer has the following
> > comments/questions:
> > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > IANA has a question about the IANA
> > > > > > > > > > > > > > > > Considerations section of this
> > > > > > > > > > > > > > document.
> > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > Previously, an early assignment has been
> > > > > > > > > > > > > > > > made to support this
> > > > > > > > > > > > > > draft. The
> > > > > > > > > > > > > > > > original request for an assignment is below:
> > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > >> <begin request=""> Contact Name:
> > > > > > > > > > > > > > > >> Thomas Morin
> > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > >> Contact Email:
> > > > > > > > > > > > > > > >> thomas.morin@orange.com
> > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > >> Type of Assignment:
> > > > > > > > > > > > > > > >> Assignement of a BGP parameter in a FCFS
registry.
> > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > >> Registry:
> > > > > > > > > > > > > > > >> BGP Tunnel Encapsulation Attribute Tunnel
> > > > > > > > > > > > > > > >> Types
> > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > >> See:
> > > > > > > > > > > > > > > >> https://www.iana.org/assignments/bgp-parame
> > > > > > > > > > > > > > > >> te
> > > > > > > > > > > > > > > >> rs
> > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > >> Description:
> > > > > > > > > > > > > > > >> Needed for draft-ietf-l3vpn-end-system, to
> > > > > > > > > > > > > > > >> allow the use of an MPLS-over-UDP
> > > > > > > > > > > > > > > >> encapsulation as specified in
> > > > > > > > > > > > > > > >> draft-ietf-mpls-in-
> > > > > > > > > > > > > > udp .
> > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > >> No value has been proposed yet, next
> > > > > > > > > > > > > > > >> available value
> > > > > > > > > > > > > > > >> 13 would be
> > > > > > > > > > > > > > fine.
> > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > >> Additional Info:
> > > > > > > > > > > > > > > >> draft-ietf-l3vpn-end-system </end>
> > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > IANA Question --> The IANA Considerations
> > > > > > > > > > > > > > > > section said "This
> > > > > > > > > > > > > > document has
> > > > > > > > > > > > > > > > no IANA actions."  and, as a result, the
> > > > > > > > > > > > > > > > assignment made through
> > > > > > > > > > > > > > the request
> > > > > > > > > > > > > > > > above would not be made permanent. Is this
> > > > > > > > > > > > > > > > the author's intent? If
> > > > > > > > > > > > > > not,
> > > > > > > > > > > > > > could
> > > > > > > > > > > > > > > > the draft be revised to indicate that the
> > > > > > > > > > > > > > > > assignment made based on
> > > > > > > > > > > > > > the
> > > > > > > > > > > > > > request
> > > > > > > > > > > > > > > > above be changed from an initial assignment
> > > > > > > > > > > > > > > > to a permanent
> > > > > > > > > > > > > > assignment.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > How do you mean?
> > > > > > > > > > > > > > The registry is FCFS for which *any* document is
> > > sufficient.
> > > > > > > > > > > > > > The assignment has been made and is as permanent
> > > > > > > > > > > > > > as any FCFS assignment ever is.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > Please note that IANA cannot reserve specific
values.
> > > > > > > > > > > > > > > > However,
> > > > > > > > > > > > > > early
> > > > > > > > > > > > > > > > allocation is available for some types of
> > > registrations.
> > > > > > > > > > > > > > > > For more
> > > > > > > > > > > > > > information,
> > > > > > > > > > > > > > > > please see RFC 7120.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Yes, but this is a FCFS registry to which 7120
> > > > > > > > > > > > > > does not apply, and nor does "reservation of
values".
> > > > > > > > > > > > > > With FCFS the value is assigned when requested
> > > > > > > > > > > > > > and that's
> > > it.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Now, it is a different question whether this
> > > > > > > > > > > > > > document should ask for the registry to be
> > > > > > > > > > > > > > updated to point to the consequent RFC instead of
the I-
> D.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > I think that might be valuable. So the IANA
> > > > > > > > > > > > > > section should
> > > > > read...
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >    IANA has previously made an allocation from
> > > > > > > > > > > > > > the "BGP Tunnel Encapsulation
> > > > > > > > > > > > > >    Attribute Tunnel Types" registry that reads:
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >    Value  | Name                      | Reference
> > > > > > > > > > > > > >    --------+---------------------------
> > > > > > > > > > > > > > +-------------------------------
> > > > > > > > > > > > > >        13  | MPLS in UDP Encapsulation |
> > > > > > > > > > > > > > [draft-ietf-l3vpn-end-system]
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >    IANA is requested to change the reference to
> > > > > > > > > > > > > > point to the RFC number
> > > > > > > > > > > > > >    of this document when it is published.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Cheers,
> > > > > > > > > > > > > > Adrian
> > > > > > > > > > > >
> > > > > > > >
> > > > > >
> > > > > > _______________________________________________
> > > > > > BESS mailing list
> > > > > > BESS@ietf.org
> > > > > > https://www.ietf.org/mailman/listinfo/bess
> > > >
> > > > _______________________________________________
> > > > BESS mailing list
> > > > BESS@ietf.org
> > > > https://www.ietf.org/mailman/listinfo/bess
> > >
> > > _______________________________________________
> > > BESS mailing list
> > > BESS@ietf.org
> > > https://www.ietf.org/mailman/listinfo/bess


From nobody Tue Mar  3 17:29:54 2015
Return-Path: <xuxiaohu@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EDB3A1A8A28 for <bess@ietfa.amsl.com>; Tue,  3 Mar 2015 17:29:51 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.211
X-Spam-Level: 
X-Spam-Status: No, score=-4.211 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 75gfAh0xwU_i for <bess@ietfa.amsl.com>; Tue,  3 Mar 2015 17:29:45 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E78741A8A43 for <bess@ietf.org>; Tue,  3 Mar 2015 17:29:43 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BPV94619; Wed, 04 Mar 2015 01:29:42 +0000 (GMT)
Received: from NKGEML408-HUB.china.huawei.com (10.98.56.39) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Wed, 4 Mar 2015 01:29:41 +0000
Received: from NKGEML512-MBS.china.huawei.com ([169.254.8.115]) by nkgeml408-hub.china.huawei.com ([10.98.56.39]) with mapi id 14.03.0158.001; Wed, 4 Mar 2015 09:29:35 +0800
From: Xuxiaohu <xuxiaohu@huawei.com>
To: "adrian@olddog.co.uk" <adrian@olddog.co.uk>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE: Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
Thread-Index: AQHQSLr3DiuxdRaR30WUQ5uVlB6ZeJzw6wsggBgiBbCAAUMgAIABT/yg
Date: Wed, 4 Mar 2015 01:29:34 +0000
Message-ID: <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE0830F69F@NKGEML512-MBS.china.huawei.com>
References: <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE083064D3@NKGEML512-MBS.china.huawei.com> <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE083066B6@NKGEML512-MBS.china.huawei.com> <00d801d048ba$e4eb09b0$aec11d10$@olddog.co.uk> <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE0830EF91@NKGEML512-MBS.china.huawei.com> <031601d055b3$e283cda0$a78b68e0$@olddog.co.uk>
In-Reply-To: <031601d055b3$e283cda0$a78b68e0$@olddog.co.uk>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.99.55]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/U4WatXV8LtvVd2aOvrmWx1z07Mw>
Subject: Re: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE: Why transform draft-xu-softwire-encaps-udp to	draft-xu-bess-encaps-udp
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Mar 2015 01:29:52 -0000

Hi Adrian,

Thanks a lot for your quick response.

> -----Original Message-----
> From: Adrian Farrel [mailto:adrian@olddog.co.uk]
> Sent: Tuesday, March 03, 2015 9:14 PM
> To: Xuxiaohu; bess@ietf.org
> Subject: RE: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//=
RE:
> Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
>=20
> OK,
>=20
> So you want an I-D that describes how to use the code point to achieve a =
variety
> of things. That is potentially useful, but I haven't read that text in yo=
ur I-D.

I can update the text accordingly soon. Of course, it would be great if you=
 could propose any text.

> But please understand that the code point is already allocated (and I sho=
uldn't
> be surprised to hear it is coded and deployed). So it is way too late to =
say that
> the wrong document asked for the code point.

How about proceeding draft-xu-bess-encaps-udp quickly and then making it to=
 take over the allocated code point? After all, the draft is very short and=
 more importantly the IETF community has reached a consensus on the necessi=
ty of allocating a code point for UDP, it seems not a problem to proceed it=
 quickly, IMHO.

Best regards,
Xiaohu

> Adrian
>=20
> > -----Original Message-----
> > From: Xuxiaohu [mailto:xuxiaohu@huawei.com]
> > Sent: 02 March 2015 10:07
> > To: Xuxiaohu; adrian@olddog.co.uk; bess@ietf.org
> > Subject: RE: [bess] Request for WG adoption of draft-xu-bess-encaps-udp=
//RE:
> > Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
> >
> > Besides the reasons as mentioned below, since the BGP Encapsulation
> > SAFI for UDP or MPLS-in-UDP is not only applicable to BGP/MPLS-based
> > L2VPN (including VPLS and EVPN) and BGP/MPLS-based L3VPN, but also be
> > applicable to other use cases (e.g., incremental deployment of
> > BGP-based Segment Routing), it seems better to specify it in a
> > separate doc rather than in the end-system-l3vpn
> draft
> > (http://datatracker.ietf.org/doc/draft-ietf-l3vpn-end-system/) which
> > is dedicated for how to use XMPP as a replacement of the BGP-L3VPN
> > signaling, IMHO.
> >
> > Best regards,
> > Xiaohu
> >
> > > -----Original Message-----
> > > From: Xuxiaohu
> > > Sent: Sunday, February 15, 2015 11:31 AM
> > > To: 'adrian@olddog.co.uk'; bess@ietf.org
> > > Subject: RE: [bess] Request for WG adoption of
> draft-xu-bess-encaps-udp//RE:
> > > Why transform draft-xu-softwire-encaps-udp to
> > > draft-xu-bess-encaps-udp
> > >
> > > Hi Adrian,
> > >
> > > > -----Original Message-----
> > > > From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Adrian
> > > > Farrel
> > > > Sent: Sunday, February 15, 2015 9:01 AM
> > > > To: bess@ietf.org
> > > > Subject: Re: [bess] Request for WG adoption of
> > > > draft-xu-bess-encaps-
> > udp//RE:
> > > > Why transform draft-xu-softwire-encaps-udp to
> > > > draft-xu-bess-encaps-udp
> > > >
> > > > Just for clarity on this point:
> > > >
> > > > I suggested that BESS was the right place to discuss whether and
> > > > how to use BGP to indicate tunnel types.
> > >
> > > Thanks for making that suggestion. Although RFC5512 (i.e.,
> > > draft-ietf-softwire-encaps-safi) and RFC5566 (i.e.,
> > > draft-ietf-softwire-encaps-ipsec) have been originated from Softwire
> > > WG and the current Softwire WG charter
> (https://tools.ietf.org/wg/softwire/charters)
> > > still state that "BGP and other routing and signaling protocols
> > > developed in
> this
> > > group will be reviewed jointly with the proper working groups and
> > > other workings that may take interest (e.g. IDR, L3VPN, PIM, LDP,
> > > SAAG, etc)",
> your
> > > above suggestion, together with Softwire co-chairs' claim of last
> > > Friday
> that the
> > > softwire WG is going to shut down and therefore it would not be a
> > > right
> place to
> > > pursue this draft anymore, seem to be a joint statement that any
> > > future work related to BGP Encapsulation SAFI should be discussed in
> > > the BESS WG, rather than the Softwire WG. Thanks again for clarifying=
 the
> confusion.
> > >
> > > > I also observed that a code point already exists in the BGP Tunnel
> > > > Encapsulation Attribute Tunnel Types registry at
> > > > http://www.iana.org/assignments/bgp-parameters/bgp-
> > parameters.xhtml#tu
> > > > n
> > > > nel-types
> > >
> > > > Therefore the debate the WG needs to have (perhaps before debating
> > > > adoption of this document) is:
> > > >
> > > > - whether the existing code point is enough for the purpose of iden=
tifying
> > > >   MPLS-in-UDP tunnels or whether more work is needed
> > > > - whether a foo-in-UDP tunnel type should be used instead with sub-=
TLVs
> > > >   to identify the different cases of foo.
> > >
> > > IMHO, it depends on whether the tunnel type of the Tunnel
> > > Encapsulation TLV
> > as
> > > specified in RFC5512 should be interpreted as UDP or MPLS-in-UDP.
> > >
> > > See the following text quoted from RFC 5512
> (draft-ietf-softwire-encaps-safi):
> > >
> > > ****************
> > > 4.2.  Protocol Type Sub-TLV
> > >
> > >    The protocol type sub-TLV MAY be encoded to indicate the type of t=
he
> > >    payload packets that will be encapsulated with the tunnel paramete=
rs
> > >    that are being signaled in the TLV.  The value field of the sub-TL=
V
> > >    contains a 2-octet protocol type that is one of the types defined =
in
> > >    [IANA-AF] as ETHER TYPEs.
> > >
> > >    For example, if we want to use three L2TPv3 sessions, one carrying
> > >    IPv4 packets, one carrying IPv6 packets, and one carrying MPLS
> > >    packets, the egress router will include three TLVs of L2TPv3
> > >    encapsulation type, each specifying a different Session ID and a
> > >    different payload type.  The protocol type sub-TLV for these will =
be
> > >    IPv4 (protocol type =3D 0x0800), IPv6 (protocol type =3D 0x86dd), =
and
> > >    MPLS (protocol type =3D 0x8847), respectively.
> > >
> > > 4.4.  Tunnel Type Selection
> > >
> > >    A BGP speaker may include multiple tunnel TLVs in the tunnel
> > >    attribute.  The receiving speaker MAY have local policies defined =
to
> > >    choose different tunnel types for different sets/types of payload
> > >    prefixes received from the same BGP speaker.  For instance, if a B=
GP
> > >    speaker includes both L2TPv3 and GRE tunnel types in the tunnel
> > >    attribute and it also advertises IPv4 and IPv6 prefixes, the ingre=
ss
> > >    router may have local policy defined to choose L2TPv3 for IPv4
> > >    prefixes (provided the protocol type received in the tunnel attrib=
ute
> > >    matches) and GRE for IPv6 prefixes.
> > >
> > > ***************
> > >
> > > To keep it in accordance with the usage of the tunnel type as
> > > specified in RFC5512, I believe the tunnel type of the Tunnel
> > > Encapsulation TLV should be interpreted as UDP, rather than
> > > MPLS-in-UDP. That's to say, what type of payload will be carried
> > > over the UDP tunnel cloud be explicitly indicated by
> the
> > > protocol type sub-TLV.
> > >
> > > Best regards,
> > > Xiaohu
> > >
> > > > Adrian
> > > >
> > > > > -----Original Message-----
> > > > > From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Xuxiaohu
> > > > > Sent: 13 February 2015 06:10
> > > > > To: Xuxiaohu; bess@ietf.org; bess-chairs@tools.ietf.org
> > > > > Cc: draft-xu-bess-encaps-udp@tools.ietf.org;
> > > > > softwire-chairs@tools.ietf.org
> > > > > Subject: [bess] Request for WG adoption of
> > > > > draft-xu-bess-encaps-udp//RE: Why transform
> > > > > draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
> > > > >
> > > > > Hi BESS WG co-chairs,
> > > > >
> > > > > The BGP tunnel type for MPLS-in-UDP has been mentioned in the
> > > > > MPLS-in-UDP draft since the 00 version
> > > > > (https://tools.ietf.org/html/draft-xu-mpls-in-udp-
> > > > > 00#page-4) which was published on April 28, 2012. However,
> > > > > according to the WG consensus during the WG adoption poll
> > > > > period, that section of "Signaling for Encapsulation in UDP" was
> > > > > removed from the MPLS-in-UDP draft and accordingly it was
> > > > > specified in a separate draft
> > > > (https://tools.ietf.org/html/draft-xu-softwire-
> > > > > encaps-udp-00) which was published on February 12, 2013.
> > > > >
> > > > > Adrian has suggested me to post this draft to BESS. Furthermore,
> > > > > Suresh has indicated that the Softwire WG would not be a right
> > > > > place for any BGP tunnel type related works anymore since the WG
> > > > > is going to shut down in the very near future. Hence, would you
> > > > > please start a WG adoption for draft-xu-bess-encaps- udp which
> > > > > is transformed from
> > > > draft-xu-softwire-encaps-udp?
> > > > >
> > > > > Best regards,
> > > > > Xiaohu
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Xuxiaohu
> > > > > > Sent: Friday, February 13, 2015 12:13 PM
> > > > > > To: Xuxiaohu; bess@ietf.org
> > > > > > Cc: Softwires WG
> > > > > > Subject: RE: [bess] Why transform draft-xu-softwire-encaps-udp
> > > > > > to draft-xu-bess-encaps-udp
> > > > > >
> > > > > > Hi all,
> > > > > >
> > > > > > I noticed the following text from RFC 5512
> > > > (draft-ietf-softwire-encaps-safi):
> > > > > >
> > > > > > ****************
> > > > > > 4.2.  Protocol Type Sub-TLV
> > > > > >
> > > > > >    The protocol type sub-TLV MAY be encoded to indicate the
> > > > > > type of
> the
> > > > > >    payload packets that will be encapsulated with the tunnel
> parameters
> > > > > >    that are being signaled in the TLV.  The value field of the =
sub-TLV
> > > > > >    contains a 2-octet protocol type that is one of the types
> > > > > > defined
> in
> > > > > >    [IANA-AF] as ETHER TYPEs.
> > > > > >
> > > > > >    For example, if we want to use three L2TPv3 sessions, one ca=
rrying
> > > > > >    IPv4 packets, one carrying IPv6 packets, and one carrying MP=
LS
> > > > > >    packets, the egress router will include three TLVs of L2TPv3
> > > > > >    encapsulation type, each specifying a different Session ID a=
nd a
> > > > > >    different payload type.  The protocol type sub-TLV for
> > > > > > these will
> be
> > > > > >    IPv4 (protocol type =3D 0x0800), IPv6 (protocol type =3D 0x8=
6dd), and
> > > > > >    MPLS (protocol type =3D 0x8847), respectively.
> > > > > > ***************
> > > > > >
> > > > > > It seems that RFC5512 is not only talking about IP-in-GRE, but
> > > > > > also
> > > > MPLS-in-GRE.
> > > > > >
> > > > > > I also noticed an expired IDR WG doc (see
> > > > > > https://tools.ietf.org/html/draft-ietf-idr-encaps-safi-00)
> > > > > > which was a predecessor of draft-ietf-softwire-encaps-safi.
> > > > > > Besides,
> > > > > > RFC5566
> > > > > > (draft-ietf-softwire-encaps-ipsec) is also originated from the
> Softwire WG.
> > > > > Does
> > > > > > it mean which WG should be responsible for the BGP Tunnel
> > > > > > Encapsulation Attribute related work had ever been discussed
> > > > > > and finally determined that
> > > > the
> > > > > > Softwire WG is the right place for it?
> > > > > >
> > > > > > Best regards,
> > > > > > Xiaohu
> > > > > >
> > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of
> > > > > > > Xuxiaohu
> > > > > > > Sent: Friday, February 13, 2015 9:21 AM
> > > > > > > To: bess@ietf.org
> > > > > > > Cc: Softwires WG
> > > > > > > Subject: [bess] Why transform draft-xu-softwire-encaps-udp
> > > > > > > to draft-xu-bess-encaps-udp
> > > > > > >
> > > > > > > Hi all,
> > > > > > >
> > > > > > > According to the suggestion from Adrian as a Routing co-AD,
> > > > > > > draft-xu-softwire-encaps-udp
> > > > > > > (https://tools.ietf.org/html/draft-xu-softwire-encaps-udp)
> > > > > > > which was posted to the Softwire WG is now posted to the BESS=
 WG.
> > > > > > >
> > > > > > > Any comments and suggestions are welcome.
> > > > > > >
> > > > > > > Best regards,
> > > > > > > Xiaohu
> > > > > > >
> > > > > > > > -----Original Message-----
> > > > > > > > From: Xuxiaohu
> > > > > > > > Sent: Friday, February 13, 2015 8:45 AM
> > > > > > > > To: 'adrian@olddog.co.uk'; 'Black, David';
> > > > > > > > rcallon@juniper.net;
> > > > > > > > draft-ietf-l3vpn-end-system@tools.ietf.org;
> > > > > > > > bess-chairs@tools.ietf.org; softwire-chairs@tools.ietf.org
> > > > > > > > Cc: 'Alvaro Retana'; akatlas@gmail.com; 'Loa Andersson'
> > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > >
> > > > > > > > Hi Adrian,
> > > > > > > >
> > > > > > > > Thanks a lot for your response. Although RFC5512 (i.e.,
> > > > > > > > draft-ietf-softwire-encaps-safi) and RFC5566 (i.e.,
> > > > > > > > draft-ietf-softwire-encaps-ipsec) which specify the BGP
> > > > > > > > Tunnel Encapsulation Attribute Tunnel Types for GRE,
> > > > > > > > L2TPv3 and IPsec respectively are all originated from
> > > > > > > > Softwire, and further the Softwire WG co-chairs didn't
> > > > > > > > state that draft-xu-softwire-encaps-udp doesn't belong to
> > > > > > > > their WG, if the BESS and Softwire WG co-chairs could
> > > > > > > > reach an agreement that any future work related to BGP
> > > > > > > > Tunnel Encapsulation Attribute should be done in the BESS W=
G, it
> looks fine to me.
> > > > > > > > I would submit the same draft to the BESS WG as
> > > > > > > draft-xu-bess-encaps-udp.
> > > > > > > >
> > > > > > > > Best regards,
> > > > > > > > Xiaohu
> > > > > > > >
> > > > > > > > > -----Original Message-----
> > > > > > > > > From: Adrian Farrel [mailto:adrian@olddog.co.uk]
> > > > > > > > > Sent: Thursday, February 12, 2015 10:50 PM
> > > > > > > > > To: Xuxiaohu; 'Black, David'; rcallon@juniper.net;
> > > > > > > > > draft-ietf-l3vpn-end-system@tools.ietf.org;
> > > > > > > > > bess-chairs@tools.ietf.org;
> > > > > > > > > softwire-chairs@tools.ietf.org
> > > > > > > > > Cc: 'Alvaro Retana'; akatlas@gmail.com; 'Loa Andersson'
> > > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > >
> > > > > > > > > Hello all,
> > > > > > > > >
> > > > > > > > > 1. Why softwire? That is strictly IP-in-IP with a
> > > > > > > > > particular intention of 4-over-6 and 6-over-4. Why would
> > > > > > > > > MPLS-in-UDP fall into their
> > > > > > > charter?
> > > > > > > > > You say "all the specifications for the BGP signaling
> > > > > > > > > for GRE, IPsec and etc were all defined in separate
> > > > > > > > > drafts belonging to the Softwire WG" but I see no evidenc=
e of
> this.
> > > > > > > > > The only vaguely related draft I can see is
> > > > > > > > > draft-xu-softwire-ip-in-udp which is a specific
> > > > > > > > > IP-over-UDP-over-IP mechanism about which I will reserve
> > > > > > > > > judgement except to say that I that softwire really
> > > > > > > > > needs yet another transition mechanism and that I
> > > > > > > > > believe IP-in-IP can be hashed by existing ECMP
> > > > > > > > hardware.
> > > > > > > > >
> > > > > > > > > You also referenced draft-xu-softwire-encaps-udp but I
> > > > > > > > > believe this document expired over 12 months ago. I
> > > > > > > > > would not say that it was the most substantive or
> > > > > > > > > technical document I have ever read
> > > > > > > > > :-)
> > > > > > > > >
> > > > > > > > > I have not removed the chairs from this thread, but I
> > > > > > > > > really hate spamming people's in-boxes.
> > > > > > > > >
> > > > > > > > > 2. While Xiaohu has correctly pointed at the current
> > > > > > > > > version of the I-D, it might be better to look at the
> > > > > > > > > status in the Datatracker via
> > > > > > > > > http://datatracker.ietf.org/doc/draft-ietf-l3vpn-end-sys
> > > > > > > > > tem/
> > > > > > > > > - You'll see that the status is Waiting for AD
> > > > > > > > > Go-Ahead::Revised I-D Needed which means it has
> > > > > > > > > completed IETF last call and is waiting for a
> > > > > > > > revision.
> > > > > > > > >
> > > > > > > > > 3. If you are following the BESS mailing list, you'll
> > > > > > > > > see that there is text agreed with IANA to fix the
> > > > > > > > > "empty" IANA considerations
> > > > > > section.
> > > > > > > > > http://www.ietf.org/mail-archive/web/bess/current/msg002
> > > > > > > > > 33.h tm l This will be in the next revision of the
> > > > > > > > > draft.
> > > > > > > > >
> > > > > > > > > 4. I am sure we can involve the BESS chairs any time we
> > > > > > > > > note some work that they need to do. At the moment, they
> > > > > > > > > may be interested to know there is a conversation, but I
> > > > > > > > > don't know that we have identified any actions for them.
> > > > > > > > > I have not removed them from this thread, but I really
> > > > > > > > > hate spamming people's
> > > > in-boxes.
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > I believe there are two pieces of work:
> > > > > > > > >
> > > > > > > > > A. Assign a BGP Tunnel Encapsulation Attribute Tunnel Typ=
e.
> > > > > > > > > This has already been done. No amount of effort to
> > > > > > > > > change documents or advance one document or another will
> > > > > > > > > change this fact. The code point has already been
> > > > > > > > > assigned. The registry is "First Come First Served" and
> > > > > > > > > no particular process was required except an application
> > > > > > > > > to IANA. No further
> > > > > > > > action is desirable.
> > > > > > > > >
> > > > > > > > > B. Specify necessary protocol work to utilise this code p=
oint.
> > > > > > > > > This is a matter for the BESS WG. They may consider that
> > > > > > > > > everything needed has already been documented, they may
> > > > > > > > > consider that they do not want to specify anything, they
> > > > > > > > > may consider that further work is needed and can be
> > > > > > > > > based on your I-D, they may consider that further work
> > > > > > > > > is needed and can needs a different starting point. The
> > > > > > > > > correct way to handle this is to post your I-D and take
> > > > > > > > > the discussion to the BESS
> > > > > > > mailing list.
> > > > > > > > You may ask the BESS chairs for advice.
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > What am I missing here?
> > > > > > > > > What do you want to achieve and why?
> > > > > > > > >
> > > > > > > > > What action are you actually asking for?
> > > > > > > > >
> > > > > > > > > Adrian
> > > > > > > > >
> > > > > > > > > > -----Original Message-----
> > > > > > > > > > From: Xuxiaohu [mailto:xuxiaohu@huawei.com]
> > > > > > > > > > Sent: 12 February 2015 05:56
> > > > > > > > > > To: Xuxiaohu; Black, David; adrian@olddog.co.uk;
> > > > > > > > > > rcallon@juniper.net;
> > > > > > > > > > draft-ietf- l3vpn-end-system@tools.ietf.org;
> > > > > > > > > > bess-chairs@tools.ietf.org; softwire-
> > > > > > > > > > chairs@tools.ietf.org
> > > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com; Loa Andersson
> > > > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > > >
> > > > > > > > > > By the way, I think it would be better to allow the
> > > > > > > > > > BESS and Softwire WG co-chairs to be involved.
> > > > > > > > > >
> > > > > > > > > > Best regards,
> > > > > > > > > > Xiaohu
> > > > > > > > > >
> > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > From: Xuxiaohu
> > > > > > > > > > > Sent: Thursday, February 12, 2015 9:47 AM
> > > > > > > > > > > To: 'Black, David'; adrian@olddog.co.uk;
> > > > > > > > > > > rcallon@juniper.net; 'draft-ietf-l3vpn-end-
> > system@tools.ietf.org'
> > > > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com; 'Loa Andersson'
> > > > > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > > > >
> > > > > > > > > > > (cced to the authors of the end-system draft)
> > > > > > > > > > >
> > > > > > > > > > > Hi all,
> > > > > > > > > > >
> > > > > > > > > > > I thinks there must be some avoidable mistaken IANA
> > > > > > > > > > > action
> > > > request.
> > > > > > > > > > > The IANA Considerations of the latest version of the
> > > > > > > > > > > end-system draft
> > > > > > > > > > > (https://tools.ietf.org/html/draft-ietf-l3vpn-end-sy
> > > > > > > > > > > stem
> > > > > > > > > > > -0
> > > > > > > > > > > 4#pa
> > > > > > > > > > > ge
> > > > > > > > > > > -2
> > > > > > > > > > > 1)
> > > > > > > > > > > which
> > > > > > > > > > was
> > > > > > > > > > > published on October 2, 2014 clearly states that "
> > > > > > > > > > > This document has no IANA actions." Furthermore, the
> > > > > > > > > > > -03 version
> > > > > > > > > > > (https://tools.ietf.org/html/draft-ietf-l3vpn-end-sy
> > > > > > > > > > > stem
> > > > > > > > > > > -0
> > > > > > > > > > > 3) which was published on September 18, 2014 and all
> > > > > > > > > > > the previous versions didn't mention the BGP tunnel
> > > > > > > > > > > type matter at all. On the contrary, the BGP tunnel
> > > > > > > > > > > type for MPLS-in-UDP has been mentioned since the
> > > > > > > > > > > 00 version of the MPLS-in-UDP draft
> > > > > > > > > > > (https://tools.ietf.org/html/draft-xu-mpls-in-udp-00
> > > > > > > > > > > #pag
> > > > > > > > > > > e-
> > > > > > > > > > > 4) which was published April 28, 2012. However,
> > > > > > > > > > > According to the WG consensus during the WG adoption
> > > > > > > > > > > poll period, that section about "Signaling for
> > > > > > > > > > > Encapsulation
> in
> > > UDP"
> > > > > > > > > > > was removed and accordingly be specified in a
> > > > > > > > > > > separate draft
> > > > > > > > > > > (https://tools.ietf.org/html/draft-xu-softwire-encap
> > > > > > > > > > > s-ud
> > > > > > > > > > > p-
> > > > > > > > > > > 00) which was published on February 12, 2013.
> > > > > > > > > > >
> > > > > > > > > > > Since the WG consensus during the adoption poll of
> > > > > > > > > > > the MPLS-in-UDP draft is to specify the signaling
> > > > > > > > > > > for encapsulation in UDP in a separate draft and all
> > > > > > > > > > > the specifications for the BGP signaling for GRE,
> > > > > > > > > > > IPsec and etc were all defined in separate drafts
> > > > > > > > > > > belonging to the Softwire WG, I do believe we should
> > > > > > > > > > > define
> > > > > > > > > > the
> > > > > > > > > > > signaling for UDP tunnel in a separate draft
> > > > > > > > > > > belonging to the Softwire
> > > > > > > WG.
> > > > > > > > > > >
> > > > > > > > > > > Since authors of the end-system draft believe the
> > > > > > > > > > > BGP tunnel type for MPLS-in-UDP is necessary and the
> > > > > > > > > > > MPLS-in-UDP draft is going to be published soon, the
> > > > > > > > > > > normative way is to move forward
> > > > > > > > > > > draft-xu-softwire-encaps-udp as
> > > > quickly as possible, IMHO.
> > > > > > > > > > >
> > > > > > > > > > > Best regards,
> > > > > > > > > > > Xiaohu
> > > > > > > > > > >
> > > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > > From: Black, David [mailto:david.black@emc.com]
> > > > > > > > > > > > Sent: Wednesday, February 11, 2015 9:58 PM
> > > > > > > > > > > > To: adrian@olddog.co.uk; Xuxiaohu;
> > > > > > > > > > > > rcallon@juniper.net
> > > > > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com; Black, David
> > > > > > > > > > > > Subject: RE: draft-ietf-l3vpn-end-system and
> > > > > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > > > > >
> > > > > > > > > > > > Adrian,
> > > > > > > > > > > >
> > > > > > > > > > > > Ok, that's roughly what I expected - between IANA
> > > > > > > > > > > > and the RFC Editor, the l3vpn-end-system draft
> > > > > > > > > > > > will record IANA's
> > > > actions
> > > > > > here.
> > > > > > > > > > > >
> > > > > > > > > > > > I had included you as the (ir)responsible AD for
> > > > > > > > > > > > the l3vpn-end-system draft, and indeed what is
> > > > > > > > > > > > transpiring is a version of "ADs can make many
> > > > > > > > > > > things happen."
> > > > > > > > > > > >
> > > > > > > > > > > > The good news is that we don't need another draft
> > > > > > > > > > > > to allocate that BGP tunnel type code point, which
> > > > > > > > > > > > was where this whole thread started, so chalk this
> > > > > > > > > > > > up as a small victory in the never-ending battle
> > > > > > > > > > > > to reduce IESG
> > > > > > > > > > > workload ;-).
> > > > > > > > > > > >
> > > > > > > > > > > > Alvaro - welcome, and congratulations on your new r=
ole!
> > > > > > > > > > > >
> > > > > > > > > > > > Thanks,
> > > > > > > > > > > > --David
> > > > > > > > > > > >
> > > > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > > > From: Adrian Farrel [mailto:adrian@olddog.co.uk]
> > > > > > > > > > > > > Sent: Wednesday, February 11, 2015 4:53 AM
> > > > > > > > > > > > > To: Black, David; 'Xuxiaohu';
> > > > > > > > > > > > > rcallon@juniper.net
> > > > > > > > > > > > > Cc: Alvaro Retana; akatlas@gmail.com
> > > > > > > > > > > > > Subject: draft-ietf-l3vpn-end-system and
> > > > > > > > > > > > > draft-ietf-mpls-in-udp
> > > > > > > > > > > > >
> > > > > > > > > > > > > Hi and sorry,
> > > > > > > > > > > > >
> > > > > > > > > > > > > I should have looked more deeply *before*
> > > > > > > > > > > > > sending my previous
> > > > > > > email.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Here is the resolution to IANA's issue with
> > > > > > > > > > > > > draft-ietf-l3vpn-end-system that I proposed and
> > > > > > > > > > > > > they
> > > accepted.
> > > > > > > > > > > > >
> > > > > > > > > > > > > We're just waiting for the authors of
> > > > > > > > > > > > > draft-ietf-l3vpn-end-system to do something.
> > > > > > > > > > > > >
> > > > > > > > > > > > > A
> > > > > > > > > > > > >
> > > > > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > > > > From: Pearl Liang via RT
> > > > > > > > > > > > > > [mailto:iana-issues@iana.org]
> > > > > > > > > > > > > > Sent: 09 December 2014 17:40
> > > > > > > > > > > > > > Cc: thomas.morin@orange.com;
> > > > > > > > > > > > > > adrian@olddog.co.uk; bess@ietf.org;
> > > > > > > > > > > > > > draft-ietf-
> > > > > > > > > > > > > > l3vpn-end-system.all@tools.ietf.org
> > > > > > > > > > > > > > Subject: [IANA #798045] IANA's comments on
> > > > > > > > > > > > > > draft-ietf-l3vpn-end-system
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Hi Adrian,
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > This makes it clear whether or not that
> > > > > > > > > > > > > > assignment needs to be updated when this draft
> > > > > > > > > > > > > > is approved for publication as
> > > > RFC:
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > [[[
> > > > > > > > > > > > > > I think that might be valuable. So the IANA
> > > > > > > > > > > > > > section should
> > > > read...
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >    IANA has previously made an allocation from
> > > > > > > > > > > > > > the "BGP Tunnel
> > > > > > > > > > > > Encapsulation
> > > > > > > > > > > > > >    Attribute Tunnel Types" registry that reads:
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >    Value  | Name                      |
> Reference
> > > > > > > > > > > > > >
> > > > --------+---------------------------+-----------------------------
> > > > --------+---------------------------+--
> > > > > > > > > > > > > >        13  | MPLS in UDP Encapsulation |
> > > > > > > > > > > > > > [draft-ietf-l3vpn-end-system]
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >    IANA is requested to change the reference
> > > > > > > > > > > > > > to point to the RFC
> > > > > > > > > number
> > > > > > > > > > > > > >    of this document when it is published.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > ]]]
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > The current text  "This document has no IANA ac=
tions."
> > > > > > > > > > > > > > provides no instructions and incorrectly tell
> > > > > > > > > > > > > > people there is no actions
> > > > > > > > > requested.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Thanks
> > > > > > > > > > > > > > ~pl
> > > > > > > > > > > > > >
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > On Tue Dec 09 13:20:57 2014,
> > > > > > > > > > > > > > adrian@olddog.co.uk
> wrote:
> > > > > > > > > > > > > > > Hi,
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Replying to myself and keeping the same IANA
> > > > > > > > > > > > > > > tracking
> > > > > number.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > > IESG/Authors/WG Chairs:
> > > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > > IANA has reviewed
> draft-ietf-l3vpn-end-system-04.
> > > > > > > > > > > > > > > > > Authors should
> > > > > > > > > > > > > > > review
> > > > > > > > > > > > > > > > > the comments and/or questions below.
> > > > > > > > > > > > > > > > > Please report any
> > > > > > > > > > > > > > > inaccuracies and
> > > > > > > > > > > > > > > > > respond to any questions as soon as possi=
ble.
> > > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > > IANA's reviewer has the following
> > > comments/questions:
> > > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > > IANA has a question about the IANA
> > > > > > > > > > > > > > > > > Considerations section of this
> > > > > > > > > > > > > > > document.
> > > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > > Previously, an early assignment has been
> > > > > > > > > > > > > > > > > made to support this
> > > > > > > > > > > > > > > draft. The
> > > > > > > > > > > > > > > > > original request for an assignment is bel=
ow:
> > > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > >> <begin request=3D""> Contact Name:
> > > > > > > > > > > > > > > > >> Thomas Morin
> > > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > > >> Contact Email:
> > > > > > > > > > > > > > > > >> thomas.morin@orange.com
> > > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > > >> Type of Assignment:
> > > > > > > > > > > > > > > > >> Assignement of a BGP parameter in a
> > > > > > > > > > > > > > > > >> FCFS
> registry.
> > > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > > >> Registry:
> > > > > > > > > > > > > > > > >> BGP Tunnel Encapsulation Attribute
> > > > > > > > > > > > > > > > >> Tunnel Types
> > > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > > >> See:
> > > > > > > > > > > > > > > > >> https://www.iana.org/assignments/bgp-pa
> > > > > > > > > > > > > > > > >> rame
> > > > > > > > > > > > > > > > >> te
> > > > > > > > > > > > > > > > >> rs
> > > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > > >> Description:
> > > > > > > > > > > > > > > > >> Needed for draft-ietf-l3vpn-end-system,
> > > > > > > > > > > > > > > > >> to allow the use of an MPLS-over-UDP
> > > > > > > > > > > > > > > > >> encapsulation as specified in
> > > > > > > > > > > > > > > > >> draft-ietf-mpls-in-
> > > > > > > > > > > > > > > udp .
> > > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > > >> No value has been proposed yet, next
> > > > > > > > > > > > > > > > >> available value
> > > > > > > > > > > > > > > > >> 13 would be
> > > > > > > > > > > > > > > fine.
> > > > > > > > > > > > > > > > >>
> > > > > > > > > > > > > > > > >> Additional Info:
> > > > > > > > > > > > > > > > >> draft-ietf-l3vpn-end-system </end>
> > > > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > > IANA Question --> The IANA
> > > > > > > > > > > > > > > > > Considerations section said "This
> > > > > > > > > > > > > > > document has
> > > > > > > > > > > > > > > > > no IANA actions."  and, as a result, the
> > > > > > > > > > > > > > > > > assignment made through
> > > > > > > > > > > > > > > the request
> > > > > > > > > > > > > > > > > above would not be made permanent. Is
> > > > > > > > > > > > > > > > > this the author's intent? If
> > > > > > > > > > > > > > > not,
> > > > > > > > > > > > > > > could
> > > > > > > > > > > > > > > > > the draft be revised to indicate that
> > > > > > > > > > > > > > > > > the assignment made based on
> > > > > > > > > > > > > > > the
> > > > > > > > > > > > > > > request
> > > > > > > > > > > > > > > > > above be changed from an initial
> > > > > > > > > > > > > > > > > assignment to a permanent
> > > > > > > > > > > > > > > assignment.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > How do you mean?
> > > > > > > > > > > > > > > The registry is FCFS for which *any*
> > > > > > > > > > > > > > > document is
> > > > sufficient.
> > > > > > > > > > > > > > > The assignment has been made and is as
> > > > > > > > > > > > > > > permanent as any FCFS assignment ever is.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > > > Please note that IANA cannot reserve
> > > > > > > > > > > > > > > > > specific
> values.
> > > > > > > > > > > > > > > > > However,
> > > > > > > > > > > > > > > early
> > > > > > > > > > > > > > > > > allocation is available for some types
> > > > > > > > > > > > > > > > > of
> > > > registrations.
> > > > > > > > > > > > > > > > > For more
> > > > > > > > > > > > > > > information,
> > > > > > > > > > > > > > > > > please see RFC 7120.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Yes, but this is a FCFS registry to which
> > > > > > > > > > > > > > > 7120 does not apply, and nor does
> > > > > > > > > > > > > > > "reservation of
> values".
> > > > > > > > > > > > > > > With FCFS the value is assigned when
> > > > > > > > > > > > > > > requested and that's
> > > > it.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Now, it is a different question whether this
> > > > > > > > > > > > > > > document should ask for the registry to be
> > > > > > > > > > > > > > > updated to point to the consequent RFC
> > > > > > > > > > > > > > > instead of
> the I-
> > D.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > I think that might be valuable. So the IANA
> > > > > > > > > > > > > > > section should
> > > > > > read...
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > >    IANA has previously made an allocation
> > > > > > > > > > > > > > > from the "BGP Tunnel Encapsulation
> > > > > > > > > > > > > > >    Attribute Tunnel Types" registry that read=
s:
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > >    Value  | Name                      |
> Reference
> > > > > > > > > > > > > > >    --------+---------------------------
> > > > > > > > > > > > > > > +-------------------------------
> > > > > > > > > > > > > > >        13  | MPLS in UDP Encapsulation |
> > > > > > > > > > > > > > > [draft-ietf-l3vpn-end-system]
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > >    IANA is requested to change the reference
> > > > > > > > > > > > > > > to point to the RFC number
> > > > > > > > > > > > > > >    of this document when it is published.
> > > > > > > > > > > > > > >
> > > > > > > > > > > > > > > Cheers,
> > > > > > > > > > > > > > > Adrian
> > > > > > > > > > > > >
> > > > > > > > >
> > > > > > >
> > > > > > > _______________________________________________
> > > > > > > BESS mailing list
> > > > > > > BESS@ietf.org
> > > > > > > https://www.ietf.org/mailman/listinfo/bess
> > > > >
> > > > > _______________________________________________
> > > > > BESS mailing list
> > > > > BESS@ietf.org
> > > > > https://www.ietf.org/mailman/listinfo/bess
> > > >
> > > > _______________________________________________
> > > > BESS mailing list
> > > > BESS@ietf.org
> > > > https://www.ietf.org/mailman/listinfo/bess


From nobody Tue Mar  3 19:01:24 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B591A1A8FD2 for <bess@ietfa.amsl.com>; Tue,  3 Mar 2015 19:01:22 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.211
X-Spam-Level: 
X-Spam-Status: No, score=-4.211 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id nMtkxRO6YFDO for <bess@ietfa.amsl.com>; Tue,  3 Mar 2015 19:01:20 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 931531A8F50 for <bess@ietf.org>; Tue,  3 Mar 2015 19:01:18 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml404-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BPV99585; Wed, 04 Mar 2015 03:01:16 +0000 (GMT)
Received: from NKGEML401-HUB.china.huawei.com (10.98.56.32) by lhreml404-hub.china.huawei.com (10.201.5.218) with Microsoft SMTP Server (TLS) id 14.3.158.1; Wed, 4 Mar 2015 03:01:16 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.146]) by nkgeml401-hub.china.huawei.com ([10.98.56.32]) with mapi id 14.03.0158.001; Wed, 4 Mar 2015 11:01:05 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: New Version Notification for draft-hao-bess-inter-nvo3-vpn-01.txt
Thread-Index: AQHQViFU1l15lhjgP02mvwhncjq2a50LlqIz
Date: Wed, 4 Mar 2015 03:01:04 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F847129@nkgeml501-mbs.china.huawei.com>
References: <20150304021703.25618.62175.idtracker@ietfa.amsl.com>
In-Reply-To: <20150304021703.25618.62175.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: [10.135.23.94]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/mS_MBFGRP0VA8TIho5WGnGnPwR0>
Cc: Shahram Davari <davari@broadcom.com>, Robert Raszuk <robert@raszuk.net>, Andrew Qu <andrew.qu@mediatek.com>, Lucy yong <lucy.yong@huawei.com>, Osama Zia <osamaz@microsoft.com>, Luyuan Fang <lufang@microsoft.com>
Subject: Re: [bess] New Version Notification for draft-hao-bess-inter-nvo3-vpn-01.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Mar 2015 03:01:22 -0000

Hi Folks,
A new version of this draft was posted, some editorial changes have been ma=
de.=20
Appreciate further suggestions and comments.

Thanks,
weiguo
________________________________________
From: internet-drafts@ietf.org [internet-drafts@ietf.org]
Sent: Wednesday, March 04, 2015 10:17
To: Lucy yong; Shahram Davari; Andrew Qu; Luyuan Fang; Haoweiguo; Lucy yong=
; Luyuan Fang; Osama Zia; Robert Raszuk; Osama Zia; Shahram Davari; Andrew =
Qu; Robert Raszuk; Haoweiguo
Subject: New Version Notification for draft-hao-bess-inter-nvo3-vpn-01.txt

A new version of I-D, draft-hao-bess-inter-nvo3-vpn-01.txt
has been successfully submitted by Weiguo Hao and posted to the
IETF repository.

Name:           draft-hao-bess-inter-nvo3-vpn
Revision:       01
Title:          Inter-AS Option B between NVO3 and BGP/MPLS IP VPN network
Document date:  2015-03-04
Group:          Individual Submission
Pages:          15
URL:            http://www.ietf.org/internet-drafts/draft-hao-bess-inter-nv=
o3-vpn-01.txt
Status:         https://datatracker.ietf.org/doc/draft-hao-bess-inter-nvo3-=
vpn/
Htmlized:       http://tools.ietf.org/html/draft-hao-bess-inter-nvo3-vpn-01
Diff:           http://www.ietf.org/rfcdiff?url2=3Ddraft-hao-bess-inter-nvo=
3-vpn-01

Abstract:
   This draft describes the solution of inter-as option-B connection
   between NVO3 network and MPLS/IP VPN network. The ASBR located in
   NVO3 network is called ASBR-d, the control plane and data plane
   procedures at ASBR-d are specified in this document, they are
   different from traditional option-B ASBR defined in [RFC 4364].




Please note that it may take a couple of minutes from the time of submissio=
n
until the htmlized version and diff are available at tools.ietf.org.

The IETF Secretariat=


From nobody Wed Mar  4 05:03:42 2015
Return-Path: <adrian@olddog.co.uk>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 036591A1A2D for <bess@ietfa.amsl.com>; Wed,  4 Mar 2015 05:03:41 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.6
X-Spam-Level: 
X-Spam-Status: No, score=-102.6 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id vYj1Sy_r0YSA for <bess@ietfa.amsl.com>; Wed,  4 Mar 2015 05:03:39 -0800 (PST)
Received: from asmtp4.iomartmail.com (asmtp4.iomartmail.com [62.128.201.175]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1A1581A1A2E for <bess@ietf.org>; Wed,  4 Mar 2015 05:03:38 -0800 (PST)
Received: from asmtp4.iomartmail.com (localhost.localdomain [127.0.0.1]) by asmtp4.iomartmail.com (8.13.8/8.13.8) with ESMTP id t24D3XTL020229; Wed, 4 Mar 2015 13:03:34 GMT
Received: from 950129200 (089144228235.atnat0037.highway.a1.net [89.144.228.235]) (authenticated bits=0) by asmtp4.iomartmail.com (8.13.8/8.13.8) with ESMTP id t24D3WP3020184 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=NO); Wed, 4 Mar 2015 13:03:33 GMT
From: "Adrian Farrel" <adrian@olddog.co.uk>
To: "'Xuxiaohu'" <xuxiaohu@huawei.com>, <bess@ietf.org>
Date: Wed, 4 Mar 2015 13:03:35 -0000
Message-ID: <011801d0567b$a0751c20$e15f5460$@olddog.co.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AdBWe4tQXyylJI22Tu61qG0sNlc9+Q==
Content-Language: en-gb
X-TM-AS-MML: disable
X-TM-AS-Product-Ver: IMSS-7.1.0.1576-7.5.0.1018-21372.004
X-TM-AS-Result: No--13.738-10.0-31-10
X-imss-scan-details: No--13.738-10.0-31-10
X-TMASE-MatchedRID: hls5oAVArl/W/bDrA6VrLZpWgCLYjjT9HzdbEmjUgfX4JyR+b5tvoP4Z AUsty2ENlW7I0ZoshYqCyuemRuUOinKgzS9kU8wEma6DzXaohvNhBfGxmdHCgnRBu8vg+6OQd3E t50Fx2rl2GnvEaeOrd8YgBOlcOjyzr9Xz9yCPONJDU5wIS9P5t5KB2VHVF2Dya0TOsL14A2noGj iUsQw3aanwi6yCm50HyHHU6IPtG2j+lL9qSl1Fb+lvTv6LG/B5kT7cMJfe6JuHwGEm+CpYGfJsl +USu3BFq2El/kehX69YW7q5cTFeugzyMxeMEX6wOX/V8P8ail3Yr6U3ZlQkdsRB0bsfrpPIfiAq rjYtFiSWnozkD0h7dvpd87jRSFkVT88iH6c8/c4ns4/X8OQKY37cGd19dSFd
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/69CJIwxCDpAIijkZ2IJJUU0Swj8>
Subject: Re: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE: Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: adrian@olddog.co.uk
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Mar 2015 13:03:41 -0000

> > So you want an I-D that describes how to use the code point to achieve a
> > variety of things. That is potentially useful, but I haven't read that text
in 
> >your I-D.
> 
> I can update the text accordingly soon. Of course, it would be great if you
could
> propose any text.

Thank you for the offer, but I don't personally have any desire to see this
written, so I will not be suggesting text.

> > But please understand that the code point is already allocated (and I
shouldn't
> > be surprised to hear it is coded and deployed). So it is way too late to say
that
> > the wrong document asked for the code point.
> 
> How about proceeding draft-xu-bess-encaps-udp quickly and then making it to
> take over the allocated code point? After all, the draft is very short and
more
> importantly the IETF community has reached a consensus on the necessity of
> allocating a code point for UDP, it seems not a problem to proceed it quickly,
> IMHO.

I don't understand at all.
The code point has been allocated. We're done. We don't need any more work. It's
done.

Adrian


From nobody Wed Mar  4 05:54:06 2015
Return-Path: <bclaise@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A1FBE1A1A00 for <bess@ietfa.amsl.com>; Wed,  4 Mar 2015 05:54:04 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -13.91
X-Spam-Level: 
X-Spam-Status: No, score=-13.91 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qm3LlunsP8U9 for <bess@ietfa.amsl.com>; Wed,  4 Mar 2015 05:54:03 -0800 (PST)
Received: from aer-iport-3.cisco.com (aer-iport-3.cisco.com [173.38.203.53]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 52F011A19F8 for <bess@ietf.org>; Wed,  4 Mar 2015 05:54:02 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=7771; q=dns/txt; s=iport; t=1425477242; x=1426686842; h=message-id:date:from:mime-version:to:subject:references: in-reply-to; bh=8UXLK1F4yNU9PSE0NO/jDlWKYGH9V4QWZloXkGkXNjY=; b=OYRhLv5Z0RmxXG7D+otnFtYwhnKzhIOud9qiWp3ta97t0kus0HOrMxSV Q0EPrHr659U0J9FgEYmtpDzBWfwZLSjKLkLQPkQLtMmVwlaCE5VyjhG33 67XnbkrIjSOv9C412wWiIjsNpp7PPttvXr+0+GEu+OlGCSOFur0/0xsSi E=;
X-Files: Attached Message Part : 124
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0CsBAAhDfdU/xbLJq1QCoNUWsETAQmFcQKBcwEBAQEBAXyEDwEBAQQBAQFrChEcAwECCgMTDwkDAgECARUmAggTBgIBAQULiBsN1xQBAQEBAQEBAwEBAQEBAQEBGosShAwGCgIBLREYBoQlBZFjgS5RhWmBU4UejHcjg289MYJDAQEB
X-IronPort-AV: E=Sophos;i="5.09,687,1418083200";  d="scan'208,217";a="364791010"
Received: from aer-iport-nat.cisco.com (HELO aer-core-2.cisco.com) ([173.38.203.22]) by aer-iport-3.cisco.com with ESMTP; 04 Mar 2015 13:54:00 +0000
Received: from [10.60.67.85] (ams-bclaise-8914.cisco.com [10.60.67.85]) by aer-core-2.cisco.com (8.14.5/8.14.5) with ESMTP id t24DrxLf015787 for <bess@ietf.org>; Wed, 4 Mar 2015 13:53:59 GMT
Message-ID: <54F70E77.1070502@cisco.com>
Date: Wed, 04 Mar 2015 14:53:59 +0100
From: Benoit Claise <bclaise@cisco.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:31.0) Gecko/20100101 Thunderbird/31.4.0
MIME-Version: 1.0
To: bess@ietf.org
References: <54F70B64.4010608@cisco.com>
In-Reply-To: <54F70B64.4010608@cisco.com>
X-Forwarded-Message-Id: <54F70B64.4010608@cisco.com>
Content-Type: multipart/mixed; boundary="------------010304000302060105000705"
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/D4XKA8KLYwVBqHN-sFn6XHE0bWQ>
Subject: [bess] Fwd: [L3sm] Some background on the L3VPN Service YANG Model work
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Mar 2015 13:54:04 -0000

This is a multi-part message in MIME format.
--------------010304000302060105000705
Content-Type: multipart/alternative;
 boundary="------------090300050305060908000608"


--------------090300050305060908000608
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit

FYI. Sorry for the potential duplicates.

Regards, Benoit


-------- Forwarded Message --------
Subject: 	[L3sm] Some background on the L3VPN Service YANG Model work
Date: 	Wed, 4 Mar 2015 14:40:52 +0100
From: 	Benoit Claise <bclaise@cisco.com>
To: 	l3sm@ietf.org



Dear all,

Let me give you some background on this new piece of work: The Layer 
Three Virtual Private Network Service Model (L3SM)

Three months ago, Adrian Farrel and I set up a design team to create a 
L3VPN _service _YANG model.
It needs to be clearly understood that this L3VPN _service _model is not 
an L3VPN configuration model. That is, it does not provide details for 
configuring network elements or protocols. Instead it contains the 
characteristics of the service, as discussed between the operators and 
their customers.  Therefore, that design team was composed of operators 
only.

That design team created a first draft: 
http://datatracker.ietf.org/doc/draft-l3vpn-service-yang/
Now that the draft has been posted, the work and discussion should 
continue in a public forum, i.e. this l3sm@ietf.org mailing list.

Next step? we're busy trying to create a new short-lived WG, focusing 
solely on this L3VPN service YANG model.
See the proposal at https://datatracker.ietf.org/doc/charter-ietf-l3sm/
If this WG is created (This proposal is on the IESG telechat this week 
for external review approval), this would be a good test for the IETF 
community: are we ready to standardize a first service YANG model? 
Personally, I believe we are.
IMO, the advantage/driver for this work is explained in the charter 
proposal:

    The deliverable from this working group will provide information to evaluate the
    set of YANG models that have already been developed or are under development,
    and will help identify any missing models or details.  The deliverable can be
    viewed as driving requirements for protocol configuration model so that the
    service parameters can be mapped into inputs used by the protocol models.

Don't hesitate to forward this email.
     List address: l3sm@ietf.org
     Archive: http://www.ietf.org/mail-archive/web/l3sm/
     To subscribe: https://www.ietf.org/mailman/listinfo/l3sm

Regards, Benoit






--------------090300050305060908000608
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: 8bit

<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=windows-1252">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    FYI. Sorry for the potential duplicates.<br>
    <br>
    Regards, Benoit<br>
    <div class="moz-forward-container"><br>
      <br>
      -------- Forwarded Message --------
      <table class="moz-email-headers-table" border="0" cellpadding="0"
        cellspacing="0">
        <tbody>
          <tr>
            <th align="RIGHT" nowrap="nowrap" valign="BASELINE">Subject:
            </th>
            <td>[L3sm] Some background on the L3VPN Service YANG Model
              work</td>
          </tr>
          <tr>
            <th align="RIGHT" nowrap="nowrap" valign="BASELINE">Date: </th>
            <td>Wed, 4 Mar 2015 14:40:52 +0100</td>
          </tr>
          <tr>
            <th align="RIGHT" nowrap="nowrap" valign="BASELINE">From: </th>
            <td>Benoit Claise <a class="moz-txt-link-rfc2396E" href="mailto:bclaise@cisco.com">&lt;bclaise@cisco.com&gt;</a></td>
          </tr>
          <tr>
            <th align="RIGHT" nowrap="nowrap" valign="BASELINE">To: </th>
            <td><a class="moz-txt-link-abbreviated" href="mailto:l3sm@ietf.org">l3sm@ietf.org</a></td>
          </tr>
        </tbody>
      </table>
      <br>
      <br>
      <meta http-equiv="Content-Type" content="text/html;
        charset=windows-1252">
      Dear all,<br>
      <br>
      Let me give you some background on this new piece of work: The
      Layer Three Virtual Private Network Service Model (L3SM)<br>
      <br>
      Three months ago, Adrian Farrel and I set up a design team to
      create a L3VPN <u>service </u>YANG model.<br>
      It needs to be clearly understood that this L3VPN <u>service </u>model

      is not an L3VPN configuration model. That is, it does not provide
      details for configuring network elements or protocols. Instead it
      contains the characteristics of the service, as discussed between
      the operators and their customers.  Therefore, that design team
      was composed of operators only.<br>
      <br>
      That design team created a first draft: <a moz-do-not-send="true"
        class="moz-txt-link-freetext"
        href="http://datatracker.ietf.org/doc/draft-l3vpn-service-yang/">http://datatracker.ietf.org/doc/draft-l3vpn-service-yang/</a>
      <br>
      Now that the draft has been posted, the work and discussion should
      continue in a public forum, i.e. this <a moz-do-not-send="true"
        class="moz-txt-link-abbreviated" href="mailto:l3sm@ietf.org">l3sm@ietf.org</a>
      mailing list.<br>
      <br>
      Next step? we're busy trying to create a new short-lived WG,
      focusing solely on this L3VPN service YANG model.<br>
      See the proposal at <a moz-do-not-send="true"
        class="moz-txt-link-freetext"
        href="https://datatracker.ietf.org/doc/charter-ietf-l3sm/">https://datatracker.ietf.org/doc/charter-ietf-l3sm/</a><br>
      If this WG is created (This proposal is on the IESG telechat this
      week for external review approval), this would be a good test for
      the IETF community: are we ready to standardize a first service
      YANG model? Personally, I believe we are. <br>
      IMO, the advantage/driver for this work is explained in the
      charter proposal:<br>
      <blockquote>
        <pre>The deliverable from this working group will provide information to evaluate the
set of YANG models that have already been developed or are under development,
and will help identify any missing models or details.  The deliverable can be
viewed as driving requirements for protocol configuration model so that the
service parameters can be mapped into inputs used by the protocol models.</pre>
      </blockquote>
      Don't hesitate to forward this email. <br>
          List address: <a moz-do-not-send="true"
        class="moz-txt-link-abbreviated" href="mailto:l3sm@ietf.org">l3sm@ietf.org</a>
      <br>
          Archive: <a moz-do-not-send="true"
        class="moz-txt-link-freetext"
        href="http://www.ietf.org/mail-archive/web/l3sm/">http://www.ietf.org/mail-archive/web/l3sm/</a>
      <br>
          To subscribe: <a moz-do-not-send="true"
        class="moz-txt-link-freetext"
        href="https://www.ietf.org/mailman/listinfo/l3sm">https://www.ietf.org/mailman/listinfo/l3sm</a><br>
      <br>
      Regards, Benoit<br>
      <br>
      <br>
      <br>
      <br>
    </div>
    <br>
  </body>
</html>

--------------090300050305060908000608--

--------------010304000302060105000705
Content-Type: text/plain; charset=windows-1252;
 name="Attached Message Part"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
 filename="Attached Message Part"

_______________________________________________
L3sm mailing list
L3sm@ietf.org
https://www.ietf.org/mailman/listinfo/l3sm


--------------010304000302060105000705--


From nobody Wed Mar  4 06:07:15 2015
Return-Path: <iesg-secretary@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7BB411A1A00; Wed,  4 Mar 2015 06:07:13 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.9
X-Spam-Level: 
X-Spam-Status: No, score=-101.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id i-l-edTYjM3N; Wed,  4 Mar 2015 06:07:12 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 0E8EE1A1A2D; Wed,  4 Mar 2015 06:07:12 -0800 (PST)
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
From: The IESG <iesg-secretary@ietf.org>
To: "IETF-Announce" <ietf-announce@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Sender: <iesg-secretary@ietf.org>
Message-ID: <20150304140712.9915.63565.idtracker@ietfa.amsl.com>
Date: Wed, 04 Mar 2015 06:07:12 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/E35q5VLRAhABMETyNeH__vO4axU>
Cc: bess@ietf.org
Subject: [bess] Last Call: <draft-ietf-bess-mvpn-bidir-03.txt> (MVPN: Using Bidirectional P-Tunnels) to Proposed Standard
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Reply-To: ietf@ietf.org
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Mar 2015 14:07:13 -0000

The IESG has received a request from the BGP Enabled Services WG (bess)
to consider the following document:
- 'MVPN: Using Bidirectional P-Tunnels'
  <draft-ietf-bess-mvpn-bidir-03.txt> as Proposed Standard

The IESG plans to make a decision in the next few weeks, and solicits
final comments on this action. Please send substantive comments to the
ietf@ietf.org mailing lists by 2015-03-18. Exceptionally, comments may be
sent to iesg@ietf.org instead. In either case, please retain the
beginning of the Subject line to allow automated sorting.

Abstract

   A set of prior RFCs specify procedures for supporting multicast in
   BGP/MPLS IP VPNs.  These procedures allow customer multicast data to
   travel across a service provider's backbone network through a set of
   multicast tunnels.  The tunnels are advertised in certain BGP
   multicast "auto-discovery" routes, by means of a BGP attribute known
   as the "Provider Multicast Service Interface (PMSI) Tunnel
   attribute".  Encodings have been defined that allow the PMSI Tunnel
   attribute to identify bidirectional (multipoint-to-multipoint)
   multicast distribution trees.  However, the prior RFCs do not provide
   all the necessary procedures for using bidirectional tunnels to
   support multicast VPNs.  This document updates RFCs 6513, 6514 and
   6625 by specifying those procedures.  In particular, it specifies the
   procedures for assigning customer multicast flows (unidirectional or
   bidirectional) to specific bidirectional tunnels in the provider
   backbone, for advertising such assignments, and for determining which
   flows have been assigned to which tunnels.




The file can be obtained via
http://datatracker.ietf.org/doc/draft-ietf-bess-mvpn-bidir/

IESG discussion can be tracked via
http://datatracker.ietf.org/doc/draft-ietf-bess-mvpn-bidir/ballot/


The following IPR Declarations may be related to this I-D:

   http://datatracker.ietf.org/ipr/2405/




From nobody Wed Mar  4 06:45:07 2015
Return-Path: <shares@ndzh.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A4CED1A1B51 for <bess@ietfa.amsl.com>; Wed,  4 Mar 2015 06:45:06 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -99.054
X-Spam-Level: 
X-Spam-Status: No, score=-99.054 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DOS_OUTLOOK_TO_MX=2.845, HTML_MESSAGE=0.001, USER_IN_WHITELIST=-100] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HUPxvv2NSs60 for <bess@ietfa.amsl.com>; Wed,  4 Mar 2015 06:45:05 -0800 (PST)
Received: from hickoryhill-consulting.com (hhc-web3.hickoryhill-consulting.com [64.9.205.143]) by ietfa.amsl.com (Postfix) with ESMTP id F35251A07BE for <bess@ietf.org>; Wed,  4 Mar 2015 06:45:04 -0800 (PST)
X-Default-Received-SPF: pass (skip=forwardok (res=PASS)) x-ip-name=74.43.47.92; 
From: "Susan Hares" <shares@ndzh.com>
To: "'Benoit Claise'" <bclaise@cisco.com>, <bess@ietf.org>
References: <54F70B64.4010608@cisco.com> <54F70E77.1070502@cisco.com>
In-Reply-To: <54F70E77.1070502@cisco.com>
Date: Wed, 4 Mar 2015 09:45:00 -0500
Message-ID: <051e01d05689$cb178950$61469bf0$@ndzh.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_NextPart_000_051F_01D0565F.E24292C0"
X-Mailer: Microsoft Outlook 14.0
Content-Language: en-us
Thread-Index: AQFTDL69ZbbHprq1RUN2ME2I7AvFHwJFJWZinfTQzIA=
X-Authenticated-User: skh@ndzh.com 
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/YVb2_X16e2cMGPWNYhq4f9TGUgU>
Subject: Re: [bess] Fwd: [L3sm] Some background on the L3VPN Service YANG Model	work
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 04 Mar 2015 14:45:06 -0000

This is a multipart message in MIME format.

------=_NextPart_000_051F_01D0565F.E24292C0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit

Benoit:

 

Is the L3VPN service model a protocol dependent model (BGP Services) or a
protocol independent modules? 

 

Sue 

 

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Benoit Claise
Sent: Wednesday, March 04, 2015 8:54 AM
To: bess@ietf.org
Subject: [bess] Fwd: [L3sm] Some background on the L3VPN Service YANG Model
work

 

FYI. Sorry for the potential duplicates.

Regards, Benoit



-------- Forwarded Message -------- 


Subject: 

[L3sm] Some background on the L3VPN Service YANG Model work


Date: 

Wed, 4 Mar 2015 14:40:52 +0100


From: 

Benoit Claise  <mailto:bclaise@cisco.com> <bclaise@cisco.com>


To: 

l3sm@ietf.org



Dear all,

Let me give you some background on this new piece of work: The Layer Three
Virtual Private Network Service Model (L3SM)

Three months ago, Adrian Farrel and I set up a design team to create a L3VPN
service YANG model.
It needs to be clearly understood that this L3VPN service model is not an
L3VPN configuration model. That is, it does not provide details for
configuring network elements or protocols. Instead it contains the
characteristics of the service, as discussed between the operators and their
customers.  Therefore, that design team was composed of operators only.

That design team created a first draft:
http://datatracker.ietf.org/doc/draft-l3vpn-service-yang/ 
Now that the draft has been posted, the work and discussion should continue
in a public forum, i.e. this l3sm@ietf.org mailing list.

Next step? we're busy trying to create a new short-lived WG, focusing solely
on this L3VPN service YANG model.
See the proposal at https://datatracker.ietf.org/doc/charter-ietf-l3sm/
If this WG is created (This proposal is on the IESG telechat this week for
external review approval), this would be a good test for the IETF community:
are we ready to standardize a first service YANG model? Personally, I
believe we are. 
IMO, the advantage/driver for this work is explained in the charter
proposal:

The deliverable from this working group will provide information to evaluate
the
set of YANG models that have already been developed or are under
development,
and will help identify any missing models or details.  The deliverable can
be
viewed as driving requirements for protocol configuration model so that the
service parameters can be mapped into inputs used by the protocol models.

Don't hesitate to forward this email. 
    List address: l3sm@ietf.org 
    Archive: http://www.ietf.org/mail-archive/web/l3sm/ 
    To subscribe: https://www.ietf.org/mailman/listinfo/l3sm

Regards, Benoit





 


------=_NextPart_000_051F_01D0565F.E24292C0
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-microsoft-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"Microsoft Word 14 =
(filtered medium)"><style><!--
/* Font Definitions */
@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:Consolas;
	panose-1:2 11 6 9 2 2 4 3 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";
	color:black;}
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";
	color:black;}
span.HTMLPreformattedChar
	{mso-style-name:"HTML Preformatted Char";
	mso-style-priority:99;
	mso-style-link:"HTML Preformatted";
	font-family:Consolas;
	color:black;}
span.EmailStyle19
	{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 bgcolor=3Dwhite =
lang=3DEN-US link=3Dblue vlink=3Dpurple><div class=3DWordSection1><p =
class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Benoit:<o:p></o:p></span></p><p class=3DMsoNormal><span =
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-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Is the L3VPN service model a protocol dependent model (BGP Services) =
or a protocol independent modules? <o:p></o:p></span></p><p =
class=3DMsoNormal><span =
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-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Sue <o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'><o:p>&nbsp;</o:p></span></p><div><div =
style=3D'border:none;border-top:solid #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";color:windowt=
ext'>From:</span></b><span =
style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif";color:windowt=
ext'> BESS [mailto:bess-bounces@ietf.org] <b>On Behalf Of </b>Benoit =
Claise<br><b>Sent:</b> Wednesday, March 04, 2015 8:54 AM<br><b>To:</b> =
bess@ietf.org<br><b>Subject:</b> [bess] Fwd: [L3sm] Some background on =
the L3VPN Service YANG Model work<o:p></o:p></span></p></div></div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>FYI. Sorry =
for the potential duplicates.<br><br>Regards, =
Benoit<o:p></o:p></p><div><p class=3DMsoNormal><br><br>-------- =
Forwarded Message -------- <o:p></o:p></p><table class=3DMsoNormalTable =
border=3D0 cellspacing=3D0 cellpadding=3D0><tr><td nowrap valign=3Dtop =
style=3D'padding:0in 0in 0in 0in'><p class=3DMsoNormal align=3Dright =
style=3D'text-align:right'><b>Subject: <o:p></o:p></b></p></td><td =
style=3D'padding:0in 0in 0in 0in'><p class=3DMsoNormal>[L3sm] Some =
background on the L3VPN Service YANG Model =
work<o:p></o:p></p></td></tr><tr><td nowrap valign=3Dtop =
style=3D'padding:0in 0in 0in 0in'><p class=3DMsoNormal align=3Dright =
style=3D'text-align:right'><b>Date: <o:p></o:p></b></p></td><td =
style=3D'padding:0in 0in 0in 0in'><p class=3DMsoNormal>Wed, 4 Mar 2015 =
14:40:52 +0100<o:p></o:p></p></td></tr><tr><td nowrap valign=3Dtop =
style=3D'padding:0in 0in 0in 0in'><p class=3DMsoNormal align=3Dright =
style=3D'text-align:right'><b>From: <o:p></o:p></b></p></td><td =
style=3D'padding:0in 0in 0in 0in'><p class=3DMsoNormal>Benoit Claise <a =
href=3D"mailto:bclaise@cisco.com">&lt;bclaise@cisco.com&gt;</a><o:p></o:p=
></p></td></tr><tr><td nowrap valign=3Dtop style=3D'padding:0in 0in 0in =
0in'><p class=3DMsoNormal align=3Dright =
style=3D'text-align:right'><b>To: <o:p></o:p></b></p></td><td =
style=3D'padding:0in 0in 0in 0in'><p class=3DMsoNormal><a =
href=3D"mailto:l3sm@ietf.org">l3sm@ietf.org</a><o:p></o:p></p></td></tr><=
/table><p class=3DMsoNormal><br><br>Dear all,<br><br>Let me give you =
some background on this new piece of work: The Layer Three Virtual =
Private Network Service Model (L3SM)<br><br>Three months ago, Adrian =
Farrel and I set up a design team to create a L3VPN <u>service </u>YANG =
model.<br>It needs to be clearly understood that this L3VPN <u>service =
</u>model is not an L3VPN configuration model. That is, it does not =
provide details for configuring network elements or protocols. Instead =
it contains the characteristics of the service, as discussed between the =
operators and their customers.&nbsp; Therefore, that design team was =
composed of operators only.<br><br>That design team created a first =
draft: <a =
href=3D"http://datatracker.ietf.org/doc/draft-l3vpn-service-yang/">http:/=
/datatracker.ietf.org/doc/draft-l3vpn-service-yang/</a> <br>Now that the =
draft has been posted, the work and discussion should continue in a =
public forum, i.e. this <a =
href=3D"mailto:l3sm@ietf.org">l3sm@ietf.org</a> mailing =
list.<br><br>Next step? we're busy trying to create a new short-lived =
WG, focusing solely on this L3VPN service YANG model.<br>See the =
proposal at <a =
href=3D"https://datatracker.ietf.org/doc/charter-ietf-l3sm/">https://data=
tracker.ietf.org/doc/charter-ietf-l3sm/</a><br>If this WG is created =
(This proposal is on the IESG telechat this week for external review =
approval), this would be a good test for the IETF community: are we =
ready to standardize a first service YANG model? Personally, I believe =
we are. <br>IMO, the advantage/driver for this work is explained in the =
charter proposal:<o:p></o:p></p><pre>The deliverable from this working =
group will provide information to evaluate the<o:p></o:p></pre><pre>set =
of YANG models that have already been developed or are under =
development,<o:p></o:p></pre><pre>and will help identify any missing =
models or details.&nbsp; The deliverable can =
be<o:p></o:p></pre><pre>viewed as driving requirements for protocol =
configuration model so that the<o:p></o:p></pre><pre>service parameters =
can be mapped into inputs used by the protocol =
models.<o:p></o:p></pre><p class=3DMsoNormal =
style=3D'margin-bottom:12.0pt'>Don't hesitate to forward this email. =
<br>&nbsp;&nbsp;&nbsp; List address: <a =
href=3D"mailto:l3sm@ietf.org">l3sm@ietf.org</a> <br>&nbsp;&nbsp;&nbsp; =
Archive: <a =
href=3D"http://www.ietf.org/mail-archive/web/l3sm/">http://www.ietf.org/m=
ail-archive/web/l3sm/</a> <br>&nbsp;&nbsp;&nbsp; To subscribe: <a =
href=3D"https://www.ietf.org/mailman/listinfo/l3sm">https://www.ietf.org/=
mailman/listinfo/l3sm</a><br><br>Regards, =
Benoit<br><br><br><br><o:p></o:p></p></div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div></body></html>
------=_NextPart_000_051F_01D0565F.E24292C0--


From nobody Wed Mar  4 17:38:58 2015
Return-Path: <xuxiaohu@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 064911ACCDC for <bess@ietfa.amsl.com>; Wed,  4 Mar 2015 17:38:57 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.211
X-Spam-Level: 
X-Spam-Status: No, score=-4.211 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NOid3Hbi1-T1 for <bess@ietfa.amsl.com>; Wed,  4 Mar 2015 17:38:55 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 7B7641A899E for <bess@ietf.org>; Wed,  4 Mar 2015 17:38:55 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BTH18157; Thu, 05 Mar 2015 01:38:54 +0000 (GMT)
Received: from NKGEML402-HUB.china.huawei.com (10.98.56.33) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Thu, 5 Mar 2015 01:38:52 +0000
Received: from NKGEML512-MBS.china.huawei.com ([169.254.8.115]) by nkgeml402-hub.china.huawei.com ([10.98.56.33]) with mapi id 14.03.0158.001; Thu, 5 Mar 2015 09:38:46 +0800
From: Xuxiaohu <xuxiaohu@huawei.com>
To: "adrian@olddog.co.uk" <adrian@olddog.co.uk>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE: Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
Thread-Index: AdBWe4tQXyylJI22Tu61qG0sNlc9+QAZRwsA
Date: Thu, 5 Mar 2015 01:38:46 +0000
Message-ID: <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE0830FC5A@NKGEML512-MBS.china.huawei.com>
References: <011801d0567b$a0751c20$e15f5460$@olddog.co.uk>
In-Reply-To: <011801d0567b$a0751c20$e15f5460$@olddog.co.uk>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.99.55]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/I8-_IpO3V9fk7ENTK_6jOpUHpdE>
Subject: Re: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE: Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Mar 2015 01:38:57 -0000

> -----Original Message-----
> From: Adrian Farrel [mailto:adrian@olddog.co.uk]
> Sent: Wednesday, March 04, 2015 9:04 PM
> To: Xuxiaohu; bess@ietf.org
> Subject: RE: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//=
RE:
> Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
>=20
> > > So you want an I-D that describes how to use the code point to
> > > achieve a variety of things. That is potentially useful, but I
> > > haven't read that text
> in
> > >your I-D.
> >
> > I can update the text accordingly soon. Of course, it would be great
> > if you
> could
> > propose any text.
>=20
> Thank you for the offer, but I don't personally have any desire to see th=
is written,
> so I will not be suggesting text.
>=20
> > > But please understand that the code point is already allocated (and
> > > I
> shouldn't
> > > be surprised to hear it is coded and deployed). So it is way too
> > > late to say
> that
> > > the wrong document asked for the code point.
> >
> > How about proceeding draft-xu-bess-encaps-udp quickly and then making
> > it to take over the allocated code point? After all, the draft is very
> > short and
> more
> > importantly the IETF community has reached a consensus on the
> > necessity of allocating a code point for UDP, it seems not a problem
> > to proceed it quickly, IMHO.
>=20
> I don't understand at all.
> The code point has been allocated. We're done. We don't need any more wor=
k.
> It's done.

If you admit the fact that the wrong doc asked for the code point, I'm a bi=
t confused why you still insist on that wrong action? Just because there is=
 a strict IETF rule that the code point reference is not allowed to be corr=
ected in any case or any other reasons? Sorry for such a direct question. I=
 just want to learn the IETF code point related rules in passing.

Xiaohu

> Adrian


From nobody Wed Mar  4 18:03:37 2015
Return-Path: <spencerdawkins.ietf@gmail.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 826B21A8A0E; Wed,  4 Mar 2015 18:03:34 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.898
X-Spam-Level: 
X-Spam-Status: No, score=-1.898 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_ADSP_CUSTOM_MED=0.001, FREEMAIL_FROM=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1L3dVlCVbH7z; Wed,  4 Mar 2015 18:03:33 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id EBCFA1A8A09; Wed,  4 Mar 2015 18:03:32 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: "Spencer Dawkins" <spencerdawkins.ietf@gmail.com>
To: "The IESG" <iesg@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150305020332.7713.4720.idtracker@ietfa.amsl.com>
Date: Wed, 04 Mar 2015 18:03:32 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/xtXPTyV4clWtaQ9paWEcBvDrj4Y>
Cc: draft-ietf-bess-orf-covering-prefixes.all@ietf.org, bess-chairs@ietf.org, martin.vigoureux@alcatel-lucent.com, bess@ietf.org
Subject: [bess] Spencer Dawkins' No Objection on draft-ietf-bess-orf-covering-prefixes-04: (with COMMENT)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Mar 2015 02:03:34 -0000

Spencer Dawkins has entered the following ballot position for
draft-ietf-bess-orf-covering-prefixes-04: No Objection

When responding, please keep the subject line intact and reply to all
email addresses included in the To and CC lines. (Feel free to cut this
introductory paragraph, however.)


Please refer to http://www.ietf.org/iesg/statement/discuss-criteria.html
for more information about IESG DISCUSS and COMMENT positions.


The document, along with other ballot positions, can be found here:
http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/



----------------------------------------------------------------------
COMMENT:
----------------------------------------------------------------------

I found myself wishing that the Abstract and/or Introduction included a
sentence describing what a "covering prefix" was. "ORF" is nicely
explained in the Introduction, but all I get on "covering prefix" is that
they're applicable in Virtual Hub-and-Spoke VPNs and BGP/MPLS Ethernet
VPN networks.

I actually know most of the terms defined in the Terminology section, but
"covering prefix" isn't explained there, either. 

The first clue I could find (searching for "cover") is on page 8, in this
sentence.

   However, because Loc-RIB also
   contains a more specific covering route (3:192.0.2.0/89),
   2:192.0.2.0/88 does not match.  

I can guess what a covering prefix is using context clues, but I'm
guessing.



From nobody Wed Mar  4 18:03:39 2015
Return-Path: <spencerdawkins.ietf@gmail.com>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id A62891A8A16; Wed,  4 Mar 2015 18:03:34 -0800 (PST)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 826B21A8A0E; Wed,  4 Mar 2015 18:03:34 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.898
X-Spam-Level: 
X-Spam-Status: No, score=-1.898 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_ADSP_CUSTOM_MED=0.001, FREEMAIL_FROM=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1L3dVlCVbH7z; Wed,  4 Mar 2015 18:03:33 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id EBCFA1A8A09; Wed,  4 Mar 2015 18:03:32 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: "Spencer Dawkins" <spencerdawkins.ietf@gmail.com>
To: "The IESG" <iesg@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150305020332.7713.4720.idtracker@ietfa.amsl.com>
Date: Wed, 04 Mar 2015 18:03:32 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/xtXPTyV4clWtaQ9paWEcBvDrj4Y>
Cc: draft-ietf-bess-orf-covering-prefixes.all@ietf.org, bess-chairs@ietf.org, martin.vigoureux@alcatel-lucent.com, bess@ietf.org
Subject: [bess] Spencer Dawkins' No Objection on draft-ietf-bess-orf-covering-prefixes-04: (with COMMENT)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Mar 2015 02:03:34 -0000

Spencer Dawkins has entered the following ballot position for
draft-ietf-bess-orf-covering-prefixes-04: No Objection

When responding, please keep the subject line intact and reply to all
email addresses included in the To and CC lines. (Feel free to cut this
introductory paragraph, however.)


Please refer to http://www.ietf.org/iesg/statement/discuss-criteria.html
for more information about IESG DISCUSS and COMMENT positions.


The document, along with other ballot positions, can be found here:
http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/



----------------------------------------------------------------------
COMMENT:
----------------------------------------------------------------------

I found myself wishing that the Abstract and/or Introduction included a
sentence describing what a "covering prefix" was. "ORF" is nicely
explained in the Introduction, but all I get on "covering prefix" is that
they're applicable in Virtual Hub-and-Spoke VPNs and BGP/MPLS Ethernet
VPN networks.

I actually know most of the terms defined in the Terminology section, but
"covering prefix" isn't explained there, either. 

The first clue I could find (searching for "cover") is on page 8, in this
sentence.

   However, because Loc-RIB also
   contains a more specific covering route (3:192.0.2.0/89),
   2:192.0.2.0/88 does not match.  

I can guess what a covering prefix is using context clues, but I'm
guessing.



From nobody Thu Mar  5 04:31:27 2015
Return-Path: <jari.arkko@piuha.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5F6D81A88D4; Thu,  5 Mar 2015 04:31:23 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.91
X-Spam-Level: 
X-Spam-Status: No, score=-1.91 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tglwCUEUfxmx; Thu,  5 Mar 2015 04:31:22 -0800 (PST)
Received: from p130.piuha.net (p130.piuha.net [IPv6:2a00:1d50:2::130]) by ietfa.amsl.com (Postfix) with ESMTP id 91D901A0204; Thu,  5 Mar 2015 04:31:21 -0800 (PST)
Received: from localhost (localhost [127.0.0.1]) by p130.piuha.net (Postfix) with ESMTP id E1F842CEC1; Thu,  5 Mar 2015 14:31:20 +0200 (EET) (envelope-from jari.arkko@piuha.net)
X-Virus-Scanned: amavisd-new at piuha.net
Received: from p130.piuha.net ([127.0.0.1]) by localhost (p130.piuha.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id S2zG-OTzg4we; Thu,  5 Mar 2015 14:31:20 +0200 (EET)
Received: from [127.0.0.1] (p130.piuha.net [IPv6:2a00:1d50:2::130]) by p130.piuha.net (Postfix) with ESMTP id 20B4C2CD0E; Thu,  5 Mar 2015 14:31:20 +0200 (EET) (envelope-from jari.arkko@piuha.net)
Content-Type: multipart/signed; boundary="Apple-Mail=_80D925B2-64D0-489C-976B-4E714AA7F5AD"; protocol="application/pgp-signature"; micalg=pgp-sha512
Mime-Version: 1.0 (Mac OS X Mail 7.3 \(1878.6\))
From: Jari Arkko <jari.arkko@piuha.net>
In-Reply-To: <CE03DB3D7B45C245BCA0D243277949363A1106@MX104CL02.corp.emc.com>
Date: Thu, 5 Mar 2015 14:31:20 +0200
Message-Id: <AD351293-6833-4689-AE60-40FABE67BF82@piuha.net>
References: <CE03DB3D7B45C245BCA0D243277949363A1106@MX104CL02.corp.emc.com>
To: "Black, David" <david.black@emc.com>
X-Mailer: Apple Mail (2.1878.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/GQ7GRPPlQv-wFigC9iP3YYoE-AU>
Cc: "ietf@ietf.org" <ietf@ietf.org>, "rbonica@juniper.net" <rbonica@juniper.net>, "General Area Review Team \(gen-art@ietf.org\)" <gen-art@ietf.org>, "luay.jalil@verizon.com" <luay.jalil@verizon.com>, "Lucy yong \(lucy.yong@huawei.com\)" <lucy.yong@huawei.com>, "keyupate@cisco.com" <keyupate@cisco.com>, "hj2387@att.com" <hj2387@att.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Gen-ART review of draft-ietf-bess-orf-covering-prefixes-04
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Mar 2015 12:31:23 -0000

--Apple-Mail=_80D925B2-64D0-489C-976B-4E714AA7F5AD
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=windows-1252

Thanks for your review, David.

I think Adrian has suggested some edits to be made. Thanks.

For what it is worth, I found the draft relatively hard to read and I =
knew much of the terminology already. But I was able to understand what =
the draft specified. But I think it would be useful make some additional =
edits along the lines of Adrian=92s suggestions or perhaps beyond. One =
thing that I often find useful is to explain the semantics of a concept =
or field as early as possible. There were a few cases in this document =
where you talked about something for a while but it was only clear later =
what its meaning was. Example: there=92s a lot of rules about the minlen =
field early in the draft, but it is only on page six that the spec =
actually says what it does. Explaining what it is for would have made =
understanding easier.

In any case, these are just comments from my perspective. I have placed =
a No-Obj position for this document on tonight=92s IESG telechat.

Jari


--Apple-Mail=_80D925B2-64D0-489C-976B-4E714AA7F5AD
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
	filename=signature.asc
Content-Type: application/pgp-signature;
	name=signature.asc
Content-Description: Message signed with OpenPGP using GPGMail

-----BEGIN PGP SIGNATURE-----
Comment: GPGTools - https://gpgtools.org

iQIcBAEBCgAGBQJU+EyYAAoJEM80gCTQU46qx9gP+wYamKHCJCGVD5F5YTBZs58X
NOWt1dL39waP65VFNUrQuMskWVw2awU6MPkEd3Yq0Q4+/4avA8veMgPHH0ugm1kw
vIrmsZ3Fh/G7SICF+G7pUGwJ+lbPrqEbh5TdGgRZnJsBlWx+bF5xKRK4dCp0JM+Y
sbrF35cmjYdfSmJOh4vp4rN5pK+4U7E3JmhVBCDKiuUb1gJWzHjJ4qKqnOyBfdat
az8ljYp46c7ALny0Wh7IyKmqKUCWjkxfWm0tSF51pYy9XfdK7zf0JGMweyiCKiK4
X29KOPDfyasUfzU7tEx1TJ+3uuSIoh+S893QnLmlIf04pCnPZeM+NcZe8356yzH5
wVRtZhNk8jkNs0erOXngvmhq9S9MQnRJYAJpwRK1e3plh+TV6Y5GaD2nB1E0YRHe
dG26vIR4su7PYp58LPOuW56tx+Sz1Pbz5xkf86qrvfUBmdWrYNdBpxTeIXeneEQt
4wCA2DDC6+y1qxY/B5bv+lwD10s7LBkUE1nedxPaLOjQu1m86AN044Mo0CvFXpm0
8PUIDPiSwBdPWRP7QqFGPEOcEjYzeJhPBJ//P33462kXnPIURqufkO2mjViloBi9
yhjLTAC3sWKfEiNP820/06oJXijDBrroVVvKv3OEhBgVUp0sII1gmbvFGf8SHt4g
XHiOuedHzwzNBMD3T6Js
=G+qE
-----END PGP SIGNATURE-----

--Apple-Mail=_80D925B2-64D0-489C-976B-4E714AA7F5AD--


From nobody Thu Mar  5 04:32:07 2015
Return-Path: <jari.arkko@piuha.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F32BC1B2B2F; Thu,  5 Mar 2015 04:32:04 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2Wf55Y7VK7qV; Thu,  5 Mar 2015 04:32:03 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id BED7F1A8783; Thu,  5 Mar 2015 04:32:03 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
From: "Jari Arkko" <jari.arkko@piuha.net>
To: "The IESG" <iesg@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150305123203.19544.32057.idtracker@ietfa.amsl.com>
Date: Thu, 05 Mar 2015 04:32:03 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/_OlrXW2fPfJc3xxikBG9IupdCHI>
Cc: draft-ietf-bess-orf-covering-prefixes.all@ietf.org, david Black <david.black@emc.com>, bess-chairs@ietf.org, martin.vigoureux@alcatel-lucent.com, bess@ietf.org
Subject: [bess] Jari Arkko's No Objection on draft-ietf-bess-orf-covering-prefixes-04: (with COMMENT)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Mar 2015 12:32:05 -0000

Jari Arkko has entered the following ballot position for
draft-ietf-bess-orf-covering-prefixes-04: No Objection

When responding, please keep the subject line intact and reply to all
email addresses included in the To and CC lines. (Feel free to cut this
introductory paragraph, however.)


Please refer to http://www.ietf.org/iesg/statement/discuss-criteria.html
for more information about IESG DISCUSS and COMMENT positions.


The document, along with other ballot positions, can be found here:
http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/



----------------------------------------------------------------------
COMMENT:
----------------------------------------------------------------------

The Gen-ART review pointed out that the draft was difficult to read. 

For what it is worth, I found the draft relatively hard to read and I
knew much of the terminology already. But I was able to understand what
the draft specified. But I think it would be useful make some additional
edits along the lines of Adrian’s suggestions or perhaps beyond. One
thing that I often find useful is to explain the semantics of a concept
or field as early as possible. There were a few cases in this document
where you talked about something for a while but it was only clear later
what its meaning was. Example: there’s a lot of rules about the minlen
field early in the draft, but it is only on page six that the spec
actually says what it does. Explaining what it is for would have made
understanding easier.

In any case, these are just comments from my perspective. This is a
No-Obj position.



From nobody Thu Mar  5 04:32:08 2015
Return-Path: <jari.arkko@piuha.net>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 206131B2C65; Thu,  5 Mar 2015 04:32:05 -0800 (PST)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F32BC1B2B2F; Thu,  5 Mar 2015 04:32:04 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2Wf55Y7VK7qV; Thu,  5 Mar 2015 04:32:03 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id BED7F1A8783; Thu,  5 Mar 2015 04:32:03 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
From: "Jari Arkko" <jari.arkko@piuha.net>
To: "The IESG" <iesg@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150305123203.19544.32057.idtracker@ietfa.amsl.com>
Date: Thu, 05 Mar 2015 04:32:03 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/_OlrXW2fPfJc3xxikBG9IupdCHI>
Cc: draft-ietf-bess-orf-covering-prefixes.all@ietf.org, david Black <david.black@emc.com>, bess-chairs@ietf.org, martin.vigoureux@alcatel-lucent.com, bess@ietf.org
Subject: [bess] Jari Arkko's No Objection on draft-ietf-bess-orf-covering-prefixes-04: (with COMMENT)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Mar 2015 12:32:05 -0000

Jari Arkko has entered the following ballot position for
draft-ietf-bess-orf-covering-prefixes-04: No Objection

When responding, please keep the subject line intact and reply to all
email addresses included in the To and CC lines. (Feel free to cut this
introductory paragraph, however.)


Please refer to http://www.ietf.org/iesg/statement/discuss-criteria.html
for more information about IESG DISCUSS and COMMENT positions.


The document, along with other ballot positions, can be found here:
http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/



----------------------------------------------------------------------
COMMENT:
----------------------------------------------------------------------

The Gen-ART review pointed out that the draft was difficult to read. 

For what it is worth, I found the draft relatively hard to read and I
knew much of the terminology already. But I was able to understand what
the draft specified. But I think it would be useful make some additional
edits along the lines of Adrian’s suggestions or perhaps beyond. One
thing that I often find useful is to explain the semantics of a concept
or field as early as possible. There were a few cases in this document
where you talked about something for a while but it was only clear later
what its meaning was. Example: there’s a lot of rules about the minlen
field early in the draft, but it is only on page six that the spec
actually says what it does. Explaining what it is for would have made
understanding easier.

In any case, these are just comments from my perspective. This is a
No-Obj position.



From nobody Thu Mar  5 05:45:06 2015
Return-Path: <tmmorin.orange@gmail.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A76EE1B2C9D for <bess@ietfa.amsl.com>; Thu,  5 Mar 2015 05:45:04 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hkcXpX3txpqm for <bess@ietfa.amsl.com>; Thu,  5 Mar 2015 05:45:00 -0800 (PST)
Received: from mail-wg0-x236.google.com (mail-wg0-x236.google.com [IPv6:2a00:1450:400c:c00::236]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id AAF0C1A038C for <bess@ietf.org>; Thu,  5 Mar 2015 05:44:57 -0800 (PST)
Received: by wghl18 with SMTP id l18so6215784wgh.5 for <bess@ietf.org>; Thu, 05 Mar 2015 05:44:56 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=sender:message-id:date:from:organization:user-agent:mime-version:to :subject:references:in-reply-to:content-type :content-transfer-encoding; bh=G+UufEVnEekc0flLQ/Gz1+8Wt+q+0+vR6j8qNRn8Ol4=; b=otJX3zwDV3ylxIRfdhBn/0FNs/9Zmfu1w76+c04h3hcydQ3dmUXhl5XnIbBgOZxf2b 58q/5kzN34AFMgZXnU5muRmiqRiRP7BezfznEJ2iwwBkL/kgTpOfow7DOMZabORORpnw wjoTfJiC775sWCfzLaNAMOJwpp3/1qzH93UKUVeh8pY28d/d9p6t3uVbHak/uKg8/f6c RZg3tqDi0h9QZOIx24H4KucrDWVTly1edqgtz6/mSqAx+frrIIDm713tmjFboh9iGrvL /3imswaJM3yqMHXP60zlbIpsfMEtPSgTcY+Q/xreDgo8BW8WHwOIxsB1R3Lxh7OD7bkj BJbA==
X-Received: by 10.180.149.206 with SMTP id uc14mr65884256wib.57.1425563096499;  Thu, 05 Mar 2015 05:44:56 -0800 (PST)
Received: from [127.0.0.1] (ARennes-652-1-16-41.w2-10.abo.wanadoo.fr. [2.10.135.41]) by mx.google.com with ESMTPSA id dn1sm8275284wid.11.2015.03.05.05.44.55 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Thu, 05 Mar 2015 05:44:55 -0800 (PST)
Sender: Thomas Morin <tmmorin.orange@gmail.com>
Message-ID: <54F85DD6.9080500@orange.com>
Date: Thu, 05 Mar 2015 14:44:54 +0100
From: Thomas Morin <thomas.morin@orange.com>
Organization: Orange
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: bess@ietf.org, draft-sajassi-bess-evpn-etree@tools.ietf.org
References: <54CF8393.9080303@orange.com>
In-Reply-To: <54CF8393.9080303@orange.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/yodFHyRYpqBwKbRuuPOol8cWvcw>
Subject: Re: [bess] Poll for adoption: draft-sajassi-bess-evpn-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Mar 2015 13:45:04 -0000

Hello working group,

This poll is now closed.
We consider that strong-enough support to adopt the document has not 
been expressed at this time.

Thank you,

-Thomas/Martin


2015-02-02, Thomas Morin :
> Hello working group,
>
> This email starts a two-week poll on adopting
> draft-sajassi-bess-evpn-etree-00 [1] as a working group item.
>
> Please send comments to the list and state if you support adoption or
> not (in the later case, please also state the reasons).
>
> This poll runs until **February 16th**.
>
>
> *Coincidentally*, we are also polling for knowledge of any IPR that
> applies to this draft, to ensure that IPR has been disclosed in
> compliance with IETF IPR rules (see RFCs 3979, 4879, 3669
> and 5378 for more details).
>
> ==> *If* you are listed as a document author or contributor please
> respond to this email and indicate whether or not you are aware of any 
> relevant IPR.
>
> The draft will not be adopted until a response has been received from
> each author and contributor.
>
> If you are not listed as an author or contributor, then please 
> explicitly respond only if you are aware of any IPR that has not yet 
> been disclosed in conformance with IETF rules.
>
> Thank you,
>
> Martin & Thomas
> bess chairs
>
> [1] https://tools.ietf.org/html/draft-sajassi-bess-evpn-etree
>
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess


From nobody Thu Mar  5 10:08:20 2015
Return-Path: <adrian@olddog.co.uk>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5D60C1A3BA4 for <bess@ietfa.amsl.com>; Thu,  5 Mar 2015 10:08:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.6
X-Spam-Level: 
X-Spam-Status: No, score=-102.6 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9K2UwUBCtAMQ for <bess@ietfa.amsl.com>; Thu,  5 Mar 2015 10:08:17 -0800 (PST)
Received: from asmtp1.iomartmail.com (asmtp1.iomartmail.com [62.128.201.248]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BE1561A6EDE for <bess@ietf.org>; Thu,  5 Mar 2015 10:08:10 -0800 (PST)
Received: from asmtp1.iomartmail.com (localhost.localdomain [127.0.0.1]) by asmtp1.iomartmail.com (8.13.8/8.13.8) with ESMTP id t25I85fn013236; Thu, 5 Mar 2015 18:08:05 GMT
Received: from 950129200 (089144217035.atnat0026.highway.a1.net [89.144.217.35]) (authenticated bits=0) by asmtp1.iomartmail.com (8.13.8/8.13.8) with ESMTP id t25I81mA013222 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=NO); Thu, 5 Mar 2015 18:08:02 GMT
From: "Adrian Farrel" <adrian@olddog.co.uk>
To: "'Xuxiaohu'" <xuxiaohu@huawei.com>, <bess@ietf.org>
References: <011801d0567b$a0751c20$e15f5460$@olddog.co.uk> <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE0830FC5A@NKGEML512-MBS.china.huawei.com>
In-Reply-To: <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE0830FC5A@NKGEML512-MBS.china.huawei.com>
Date: Thu, 5 Mar 2015 18:08:04 -0000
Message-ID: <006701d0576f$54a6ac60$fdf40520$@olddog.co.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQE283CfN5IGaWfBMRW+Mxaet841MAFU4JJnnjZMfrA=
Content-Language: en-gb
X-TM-AS-MML: disable
X-TM-AS-Product-Ver: IMSS-7.1.0.1576-7.5.0.1018-21378.001
X-TM-AS-Result: No--14.919-10.0-31-10
X-imss-scan-details: No--14.919-10.0-31-10
X-TMASE-MatchedRID: 1GZI+iG+MtfW/bDrA6VrLcrT39PoDNtWG08M2I9s0LreTWVpUu3B1+aN 0ZrIlpD2HJg6MNOhyoEE6Oq8X2Wm0j7u9X0tTsZjYD9XTRdaMO3SL+EVfOJR05MxNpDOG+h68my X5RK7cEWrYSX+R6Ffr1hburlxMV66p/RCY+priR6Q+gWwzffozgbB3zBRj24zfeHPnu31iHCJsH V20BY/2ECKryZ7b+657oJdzHygnP/z4nYRVqkbX+IfK/Jd5eHmnvBHr/aFnM7WCKzDYkqjyXFD0 I0jlEDdtSCkRsJOQ9fInBeseqdTki9FtW7XfHuebBu6+EIezdzSokR1Kq7cHR0wpmbAmLhIkwE2 YUcQMETi8zVgXoAltlwtzewu2M635MIx11wv+CM7AFczfjr/7NK27f7zJL2ZiyIcOjUiIqqVCtw SShe5OQbZLhu3X9XIrconoP401LI=
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/DTMfDjOv8bq6OQpXCJaWOFXRecw>
Subject: Re: [bess] Request for WG adoption of draft-xu-bess-encaps-udp//RE: Why transform draft-xu-softwire-encaps-udp to draft-xu-bess-encaps-udp
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: adrian@olddog.co.uk
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Mar 2015 18:08:19 -0000

This is my last email on this subject.

> > > How about proceeding draft-xu-bess-encaps-udp quickly and then making
> > > it to take over the allocated code point? After all, the draft is very
> > > short and more importantly the IETF community has reached a
> > > consensus on the necessity of allocating a code point for UDP, it seems
> > > not a problem to proceed it quickly, IMHO.
> >
> > I don't understand at all.
> > The code point has been allocated. We're done. We don't need any more work.
> > It's done.
> 
> If you admit the fact that the wrong doc asked for the code point, I'm a bit
> confused why you still insist on that wrong action? Just because there is a
strict
> IETF rule that the code point reference is not allowed to be corrected in any
case
> or any other reasons? Sorry for such a direct question. I just want to learn
the
> IETF code point related rules in passing.

If.

But I don't admit any such thing.

The registry is First Come First Served.
The code point was requested according to that policy.
A document was provided as a reference (and that document provides further
reference to [I-D.ietf-mpls-in-udp])

The first objective is to allocate the code point - done
The second objective is to give a reference to a document that describes the use
of the code point - done

You are proposing a new I-D that provides more description and use cases. Sure.
Knock yourself out. I'm sure people will be happy to discuss the content of your
document.

Adrian




From nobody Thu Mar  5 10:18:26 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6A3BB1A6F32; Thu,  5 Mar 2015 10:18:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RJSEnxg1Qiyt; Thu,  5 Mar 2015 10:18:18 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 65A5B1A6F04; Thu,  5 Mar 2015 10:18:18 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150305181818.30638.62772.idtracker@ietfa.amsl.com>
Date: Thu, 05 Mar 2015 10:18:18 -0800
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/2LdsHMRb8KMzp8VvB_SyJJyYZqA>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-04.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Mar 2015 18:18:19 -0000

IESG state changed to Approved-announcement to be sent::Point Raised - writeup needed from IESG Evaluation
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Thu Mar  5 10:18:27 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 872BE1A6F33; Thu,  5 Mar 2015 10:18:19 -0800 (PST)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6A3BB1A6F32; Thu,  5 Mar 2015 10:18:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RJSEnxg1Qiyt; Thu,  5 Mar 2015 10:18:18 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 65A5B1A6F04; Thu,  5 Mar 2015 10:18:18 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150305181818.30638.62772.idtracker@ietfa.amsl.com>
Date: Thu, 05 Mar 2015 10:18:18 -0800
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/2LdsHMRb8KMzp8VvB_SyJJyYZqA>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-04.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 05 Mar 2015 18:18:19 -0000

IESG state changed to Approved-announcement to be sent::Point Raised - writeup needed from IESG Evaluation
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Fri Mar  6 03:53:29 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 19DEC1ACDB1; Fri,  6 Mar 2015 03:53:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qr12YAmyhfYr; Fri,  6 Mar 2015 03:53:19 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 72D9E1ACDA2; Fri,  6 Mar 2015 03:53:19 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150306115319.14397.76333.idtracker@ietfa.amsl.com>
Date: Fri, 06 Mar 2015 03:53:19 -0800
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/KC88YFN9AtIh76-y_UszX2TDMCc>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-04.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 11:53:25 -0000

IESG state changed to Approved-announcement to be sent::Revised I-D Needed from Approved-announcement to be sent::Point Raised - writeup needed
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Fri Mar  6 03:53:30 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 3DC531ACDB3; Fri,  6 Mar 2015 03:53:25 -0800 (PST)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 19DEC1ACDB1; Fri,  6 Mar 2015 03:53:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qr12YAmyhfYr; Fri,  6 Mar 2015 03:53:19 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 72D9E1ACDA2; Fri,  6 Mar 2015 03:53:19 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150306115319.14397.76333.idtracker@ietfa.amsl.com>
Date: Fri, 06 Mar 2015 03:53:19 -0800
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/KC88YFN9AtIh76-y_UszX2TDMCc>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-04.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 11:53:25 -0000

IESG state changed to Approved-announcement to be sent::Revised I-D Needed from Approved-announcement to be sent::Point Raised - writeup needed
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Fri Mar  6 04:45:52 2015
Return-Path: <adrian@olddog.co.uk>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id AF89F1ACDAD; Fri,  6 Mar 2015 03:54:03 -0800 (PST)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8AE141ACD7D for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Fri,  6 Mar 2015 03:54:03 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.9
X-Spam-Level: 
X-Spam-Status: No, score=-101.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Co91ZEULFmzD for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Fri,  6 Mar 2015 03:54:02 -0800 (PST)
Received: from gamay.tools.ietf.org (gamay.tools.ietf.org [IPv6:2607:f170:8000:1500::de]) (using TLSv1.2 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id F2AD01ACDBC for <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>; Fri,  6 Mar 2015 03:53:53 -0800 (PST)
Received: from asmtp1.iomartmail.com ([62.128.201.248]:57879) by gamay.tools.ietf.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.80) (envelope-from <adrian@olddog.co.uk>) id 1YTqpD-0001bb-HK for draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org; Fri, 06 Mar 2015 06:53:52 -0500
Received: from asmtp1.iomartmail.com (localhost.localdomain [127.0.0.1]) by asmtp1.iomartmail.com (8.13.8/8.13.8) with ESMTP id t26BrWpZ001712; Fri, 6 Mar 2015 11:53:33 GMT
Received: from 950129200 (089144209037.atnat0018.highway.a1.net [89.144.209.37]) (authenticated bits=0) by asmtp1.iomartmail.com (8.13.8/8.13.8) with ESMTP id t26BrVAY001695 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=NO); Fri, 6 Mar 2015 11:53:32 GMT
From: "Adrian Farrel" <adrian@olddog.co.uk>
To: <draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org>
Date: Fri, 6 Mar 2015 11:53:33 -0000
Message-ID: <01e301d05804$2cdf2780$869d7680$@olddog.co.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AdBYBBtfj9hH1/g7Qv+Yst/R6SP/nQ==
Content-Language: en-gb
X-TM-AS-MML: disable
X-TM-AS-Product-Ver: IMSS-7.1.0.1576-7.5.0.1018-21380.003
X-TM-AS-Result: No--0.581-10.0-31-10
X-imss-scan-details: No--0.581-10.0-31-10
X-TMASE-MatchedRID: Hr9axt0KMZywIZP5FYus+691/YHX0i1lNRv92EahEcHSYAzZ6KmqWj0J SifQ1MZ5eE0XHEXdyyf/pHIx7Ve7fA9rzSdQQVxtvHKClHGjjr0Q2zDF7O7/jgpcq1+LTpsUo8W MkQWv6iV95l0nVeyiuBQF+BLVItD4C24oEZ6SpSmcfuxsiY4QFE6YUbhsq/RWUpTFUqh0CoGlAI IwbdwiQFoHDSOxdKqIJCXcKjhEfBY7CDV+Jr+uvip386qp8dLaO2Pngq1MoQXzXnV7SUpsKOQ/b JqlQQ/copZ51QxAj3fx+wL7E8WTG+w2n3R9PgSOsBTJSD2iAW0=
X-SA-Exim-Connect-IP: 62.128.201.248
X-SA-Exim-Rcpt-To: draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org
X-SA-Exim-Mail-From: adrian@olddog.co.uk
X-SA-Exim-Version: 4.2.1 (built Mon, 26 Dec 2011 16:57:07 +0000)
X-SA-Exim-Scanned: Yes (on gamay.tools.ietf.org)
Resent-To: draft-ietf-bess-orf-covering-prefixes.all@ietf.org
Resent-Message-Id: <20150306115353.F2AD01ACDBC@ietfa.amsl.com>
Resent-Date: Fri,  6 Mar 2015 03:53:53 -0800 (PST)
Resent-From: adrian@olddog.co.uk
Archived-At: <http://mailarchive.ietf.org/arch/msg/draft-ietf-bess-orf-covering-prefixes.all@tools/KZYudBE72obyW5oVahdWA2g-WEU>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/NCqo1AbmLKzI6ezW-4PkWp1B9cs>
X-Mailman-Approved-At: Fri, 06 Mar 2015 04:45:50 -0800
Cc: db3546@att.com
Subject: [bess] Progress with draft-ietf-bess-orf-covering-prefixes
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: adrian@olddog.co.uk
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 11:54:03 -0000

Hi,

draft-ietf-bess-orf-covering-prefixes was on yesterday's IESG telechat and was
approved for publication as an RFC.

However, there are some Comments in datatracker that I would like you to try to
address in a new revision. You can see the Comments at
https://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/ballot/

I think all of the issues are contained in my Comment and in that from Brian
Haberman (all the other Comments are duplicates).

This looks like an easy editorial revision, so please try to post it before the
pre-Dallas I-D cut-off at the end of the weekend. If you miss that, please send
me a copy and I will get it posted.

Thanks,
Adrian


From nobody Fri Mar  6 08:43:31 2015
Return-Path: <internet-drafts@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B2F3F1A0334; Fri,  6 Mar 2015 08:43:27 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id lOFYcmg63Par; Fri,  6 Mar 2015 08:43:26 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id E4D571A0107; Fri,  6 Mar 2015 08:43:11 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <i-d-announce@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150306164311.31034.80302.idtracker@ietfa.amsl.com>
Date: Fri, 06 Mar 2015 08:43:11 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/KcFm3x1ki8BACwpTLu9IYlUAWi8>
Cc: bess@ietf.org
Subject: [bess] I-D Action: draft-ietf-bess-orf-covering-prefixes-05.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 16:43:28 -0000

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

        Title           : Covering Prefixes Outbound Route Filter for BGP-4
        Authors         : Huajin Jeng
                          Luay Jalil
                          Ron Bonica
                          Keyur Patel
                          Lucy Yong
	Filename        : draft-ietf-bess-orf-covering-prefixes-05.txt
	Pages           : 20
	Date            : 2015-03-06

Abstract:
   This document defines a new Outbound Route Filter (ORF) type, called
   the "Covering Prefixes ORF (CP-ORF)".  CP-ORF is applicable in
   Virtual Hub-and-Spoke VPNs.  It also is applicable in BGP/MPLS
   Ethernet VPN (EVPN) networks.



The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-bess-orf-covering-prefixes-05

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=draft-ietf-bess-orf-covering-prefixes-05


Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

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


From nobody Fri Mar  6 08:43:39 2015
Return-Path: <internet-drafts@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7511B1ACF24; Fri,  6 Mar 2015 08:43:31 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zJTGw4xwOult; Fri,  6 Mar 2015 08:43:29 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 0E2451A1AA3; Fri,  6 Mar 2015 08:43:12 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>, <adrian@olddog.co.uk>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150306164312.31034.58993.idtracker@ietfa.amsl.com>
Date: Fri, 06 Mar 2015 08:43:12 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/jEQ1F5EoNhuvijKSbgICFt1DtQI>
Subject: [bess] New Version Notification - draft-ietf-bess-orf-covering-prefixes-05.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 16:43:31 -0000

A new version (-05) has been submitted for draft-ietf-bess-orf-covering-prefixes:
http://www.ietf.org/internet-drafts/draft-ietf-bess-orf-covering-prefixes-05.txt

Sub state has been changed to AD Followup from Revised ID Needed


The IETF datatracker page for this Internet-Draft is:
https://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/

Diff from previous version:
http://www.ietf.org/rfcdiff?url2=draft-ietf-bess-orf-covering-prefixes-05

Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

IETF Secretariat.


From nobody Fri Mar  6 08:43:40 2015
Return-Path: <internet-drafts@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 9AB2F1A1AA3; Fri,  6 Mar 2015 08:43:31 -0800 (PST)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7511B1ACF24; Fri,  6 Mar 2015 08:43:31 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zJTGw4xwOult; Fri,  6 Mar 2015 08:43:29 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 0E2451A1AA3; Fri,  6 Mar 2015 08:43:12 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>, <adrian@olddog.co.uk>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150306164312.31034.58993.idtracker@ietfa.amsl.com>
Date: Fri, 06 Mar 2015 08:43:12 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/jEQ1F5EoNhuvijKSbgICFt1DtQI>
Subject: [bess] New Version Notification - draft-ietf-bess-orf-covering-prefixes-05.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 16:43:31 -0000

A new version (-05) has been submitted for draft-ietf-bess-orf-covering-prefixes:
http://www.ietf.org/internet-drafts/draft-ietf-bess-orf-covering-prefixes-05.txt

Sub state has been changed to AD Followup from Revised ID Needed


The IETF datatracker page for this Internet-Draft is:
https://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/

Diff from previous version:
http://www.ietf.org/rfcdiff?url2=draft-ietf-bess-orf-covering-prefixes-05

Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

IETF Secretariat.


From nobody Fri Mar  6 08:55:57 2015
Return-Path: <internet-drafts@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 01FE11AD0A3; Fri,  6 Mar 2015 08:55:46 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id znfUX-VNdtQw; Fri,  6 Mar 2015 08:55:37 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id F21841ACF17; Fri,  6 Mar 2015 08:55:04 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <i-d-announce@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150306165504.31026.53946.idtracker@ietfa.amsl.com>
Date: Fri, 06 Mar 2015 08:55:04 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/9I0r089aYLdrPskGbLHgtLwVAWc>
Cc: bess@ietf.org
Subject: [bess] I-D Action: draft-ietf-bess-orf-covering-prefixes-06.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 16:55:46 -0000

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

        Title           : Covering Prefixes Outbound Route Filter for BGP-4
        Authors         : Huajin Jeng
                          Luay Jalil
                          Ron Bonica
                          Keyur Patel
                          Lucy Yong
	Filename        : draft-ietf-bess-orf-covering-prefixes-06.txt
	Pages           : 20
	Date            : 2015-03-06

Abstract:
   This document defines a new Outbound Route Filter (ORF) type, called
   the "Covering Prefixes ORF (CP-ORF)".  CP-ORF is applicable in
   Virtual Hub-and-Spoke VPNs.  It also is applicable in BGP/MPLS
   Ethernet VPN (EVPN) networks.



The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-bess-orf-covering-prefixes-06

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=draft-ietf-bess-orf-covering-prefixes-06


Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

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


From nobody Fri Mar  6 08:55:58 2015
Return-Path: <internet-drafts@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2CA691AD0A3; Fri,  6 Mar 2015 08:55:50 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RTYTbJF36xWG; Fri,  6 Mar 2015 08:55:46 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 2C8E01AD04E; Fri,  6 Mar 2015 08:55:05 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>, <adrian@olddog.co.uk>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150306165505.31026.88229.idtracker@ietfa.amsl.com>
Date: Fri, 06 Mar 2015 08:55:05 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/bO7KAP9YoI4zqdgSS6mPsgPj6Qo>
Subject: [bess] New Version Notification - draft-ietf-bess-orf-covering-prefixes-06.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 16:55:50 -0000

A new version (-06) has been submitted for draft-ietf-bess-orf-covering-prefixes:
http://www.ietf.org/internet-drafts/draft-ietf-bess-orf-covering-prefixes-06.txt


The IETF datatracker page for this Internet-Draft is:
https://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/

Diff from previous version:
http://www.ietf.org/rfcdiff?url2=draft-ietf-bess-orf-covering-prefixes-06

Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

IETF Secretariat.


From nobody Fri Mar  6 08:56:00 2015
Return-Path: <internet-drafts@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 5B0221AD0A5; Fri,  6 Mar 2015 08:55:50 -0800 (PST)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2CA691AD0A3; Fri,  6 Mar 2015 08:55:50 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RTYTbJF36xWG; Fri,  6 Mar 2015 08:55:46 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 2C8E01AD04E; Fri,  6 Mar 2015 08:55:05 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>, <adrian@olddog.co.uk>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150306165505.31026.88229.idtracker@ietfa.amsl.com>
Date: Fri, 06 Mar 2015 08:55:05 -0800
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/bO7KAP9YoI4zqdgSS6mPsgPj6Qo>
Subject: [bess] New Version Notification - draft-ietf-bess-orf-covering-prefixes-06.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 16:55:50 -0000

A new version (-06) has been submitted for draft-ietf-bess-orf-covering-prefixes:
http://www.ietf.org/internet-drafts/draft-ietf-bess-orf-covering-prefixes-06.txt


The IETF datatracker page for this Internet-Draft is:
https://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/

Diff from previous version:
http://www.ietf.org/rfcdiff?url2=draft-ietf-bess-orf-covering-prefixes-06

Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

IETF Secretariat.


From nobody Fri Mar  6 14:12:55 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 847F11A8033; Fri,  6 Mar 2015 14:12:54 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uZx20U2XY3DG; Fri,  6 Mar 2015 14:12:53 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 813C61A6F2B; Fri,  6 Mar 2015 14:12:53 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150306221253.1103.3805.idtracker@ietfa.amsl.com>
Date: Fri, 06 Mar 2015 14:12:53 -0800
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/QbVc4qOT_tIwDd5fp_Hbcjs20k4>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 22:12:54 -0000

IESG state changed to Approved-announcement to be sent from Approved-announcement to be sent::AD Followup
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Fri Mar  6 14:12:57 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id A6EBC1A6F2B; Fri,  6 Mar 2015 14:12:54 -0800 (PST)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 847F11A8033; Fri,  6 Mar 2015 14:12:54 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uZx20U2XY3DG; Fri,  6 Mar 2015 14:12:53 -0800 (PST)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 813C61A6F2B; Fri,  6 Mar 2015 14:12:53 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p2
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150306221253.1103.3805.idtracker@ietfa.amsl.com>
Date: Fri, 06 Mar 2015 14:12:53 -0800
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/QbVc4qOT_tIwDd5fp_Hbcjs20k4>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 22:12:54 -0000

IESG state changed to Approved-announcement to be sent from Approved-announcement to be sent::AD Followup
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Sun Mar  8 19:32:57 2015
Return-Path: <kishoret@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E0A321A0636; Sun,  8 Mar 2015 19:32:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.902
X-Spam-Level: 
X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9bgp_06s9miL; Sun,  8 Mar 2015 19:32:54 -0700 (PDT)
Received: from na01-by2-obe.outbound.protection.outlook.com (mail-by2on0135.outbound.protection.outlook.com [207.46.100.135]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 378211A017C; Sun,  8 Mar 2015 19:32:54 -0700 (PDT)
Received: from DM2PR05MB686.namprd05.prod.outlook.com (10.141.176.147) by BY2PR05MB063.namprd05.prod.outlook.com (10.242.34.151) with Microsoft SMTP Server (TLS) id 15.1.99.14; Mon, 9 Mar 2015 02:32:51 +0000
Received: from DM2PR05MB686.namprd05.prod.outlook.com ([10.141.176.147]) by DM2PR05MB686.namprd05.prod.outlook.com ([10.141.176.147]) with mapi id 15.01.0106.007; Mon, 9 Mar 2015 02:32:51 +0000
From: Kishore Tiruveedhula <kishoret@juniper.net>
To: "bess@ietf.org" <bess@ietf.org>, "rtg-yang-coord@ietf.org" <rtg-yang-coord@ietf.org>
Thread-Topic: New Version Notification for draft-tsingh-bess-pbb-evpn-yang-cfg-00.txt
Thread-Index: AQHQWFVKSMnmKqnOt0Gzw8rKYK5nwJ0TLvKA
Date: Mon, 9 Mar 2015 02:32:51 +0000
Message-ID: <D1227CDF.233EE%kishoret@juniper.net>
References: <20150306213400.17780.84752.idtracker@ietfa.amsl.com>
In-Reply-To: <20150306213400.17780.84752.idtracker@ietfa.amsl.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [66.129.241.11]
authentication-results: ietf.org; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BY2PR05MB063;
x-forefront-antispam-report: BMV:0; SFV:NSPM; SFS:(10019020)(6009001)(164054003)(479174004)(377424004)(24454002)(53754006)(377454003)(51704005)(15975445007)(2501003)(92566002)(102836002)(77096005)(2950100001)(2900100001)(2656002)(1720100001)(19580405001)(62966003)(87936001)(77156002)(230783001)(66066001)(86362001)(50986999)(54356999)(99286002)(76176999)(40100003)(36756003)(122556002)(19580395003)(106116001)(83506001)(46102003); DIR:OUT; SFP:1102; SCL:1; SRVR:BY2PR05MB063; H:DM2PR05MB686.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BY2PR05MB063AAE77DBD5FE5CDC3B938981B0@BY2PR05MB063.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5001009)(5005006); SRVR:BY2PR05MB063; BCL:0; PCL:0; RULEID:; SRVR:BY2PR05MB063; 
x-forefront-prvs: 05102978A2
Content-Type: text/plain; charset="us-ascii"
Content-ID: <FD0E05EEC7D2514CA0BD4A67E35C452E@namprd05.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: juniper.net
X-MS-Exchange-CrossTenant-originalarrivaltime: 09 Mar 2015 02:32:51.1470 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BY2PR05MB063
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/tatdXBtjYpLXIzUnDhhEzCys9Ls>
Cc: Tapraj Singh <tsingh@juniper.net>, Ali Sajassi <sajassi@cisco.com>, Luay Jalil <luay.jalil@verizon.com>, Deepak Kumar <dekumar@cisco.com>
Subject: Re: [bess] New Version Notification for draft-tsingh-bess-pbb-evpn-yang-cfg-00.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Mar 2015 02:32:56 -0000

Hi all,
   We submitted the new draft PBB EVPN Yang model and would like to
solicit comments.


Thanks,
Kishore

On 3/6/15, 4:34 PM, "internet-drafts@ietf.org" <internet-drafts@ietf.org>
wrote:

>
>A new version of I-D, draft-tsingh-bess-pbb-evpn-yang-cfg-00.txt
>has been successfully submitted by Kishore Tiruveedhula and posted to the
>IETF repository.
>
>Name:		draft-tsingh-bess-pbb-evpn-yang-cfg
>Revision:	00
>Title:		YANG Data Model for PBB EVPN protocol
>Document date:	2015-03-06
>Group:		Individual Submission
>Pages:		14
>URL:           =20
>http://www.ietf.org/internet-drafts/draft-tsingh-bess-pbb-evpn-yang-cfg-00
>.txt
>Status:        =20
>https://datatracker.ietf.org/doc/draft-tsingh-bess-pbb-evpn-yang-cfg/
>Htmlized:      =20
>http://tools.ietf.org/html/draft-tsingh-bess-pbb-evpn-yang-cfg-00
>
>
>Abstract:
>   This document defines a YANG data model that can be used to configure
>   and manage PBB EVPN.
>
>
>                 =20
>       =20
>
>
>Please note that it may take a couple of minutes from the time of
>submission
>until the htmlized version and diff are available at tools.ietf.org.
>
>The IETF Secretariat
>


From nobody Mon Mar  9 01:57:10 2015
Return-Path: <adrian@olddog.co.uk>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id A76B71A86E4; Fri,  6 Mar 2015 14:16:02 -0800 (PST)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 84D2D1A8033 for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Fri,  6 Mar 2015 14:16:02 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.9
X-Spam-Level: 
X-Spam-Status: No, score=-101.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, USER_IN_WHITELIST=-100] autolearn=unavailable
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Sg5GHVzcjDYo for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Fri,  6 Mar 2015 14:16:02 -0800 (PST)
Received: from gamay.tools.ietf.org (gamay.tools.ietf.org [IPv6:2607:f170:8000:1500::de]) (using TLSv1.2 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E7E0B1A0334 for <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>; Fri,  6 Mar 2015 14:16:01 -0800 (PST)
Received: from asmtp4.iomartmail.com ([62.128.201.175]:38219) by gamay.tools.ietf.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.80) (envelope-from <adrian@olddog.co.uk>) id 1YU0XI-0000J9-7r for draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org; Fri, 06 Mar 2015 17:16:00 -0500
Received: from asmtp4.iomartmail.com (localhost.localdomain [127.0.0.1]) by asmtp4.iomartmail.com (8.13.8/8.13.8) with ESMTP id t26MFhSI023961; Fri, 6 Mar 2015 22:15:43 GMT
Received: from 950129200 (089144213255.atnat0022.highway.a1.net [89.144.213.255]) (authenticated bits=0) by asmtp4.iomartmail.com (8.13.8/8.13.8) with ESMTP id t26MFfjt023951 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=NO); Fri, 6 Mar 2015 22:15:42 GMT
From: "Adrian Farrel" <adrian@olddog.co.uk>
To: <iesg@ietf.org>
Date: Fri, 6 Mar 2015 22:15:43 -0000
Message-ID: <038d01d0585b$17184880$4548d980$@olddog.co.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AdBYWxTk+a52fYhFSiuLQIcnmUQjcA==
Content-Language: en-gb
X-TM-AS-MML: disable
X-TM-AS-Product-Ver: IMSS-7.1.0.1576-7.5.0.1018-21382.003
X-TM-AS-Result: No--3.279-10.0-31-10
X-imss-scan-details: No--3.279-10.0-31-10
X-TMASE-MatchedRID: Ta1p0tj4C5HUbBocSvqNd5dc7I2df+msmyqQJWNsukm638ZUY6gSd5sX 98LUY2W7I8qZxoWaGPUfZdczzDm/uk1+zyfzlN7y/sToY2qzpx7tnih9yxn/S0GB75ZTJT+Tjoc zmuoPCq0TrT9WDR+aTEYrL+9XI82kv6S858q65QiyFMISptJMECVXed9cfRfF5D9smqVBD9yiln nVDECPd/H7AvsTxZMb7DafdH0+BI6wFMlIPaIBbQ==
X-SA-Exim-Connect-IP: 62.128.201.175
X-SA-Exim-Rcpt-To: draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org
X-SA-Exim-Mail-From: adrian@olddog.co.uk
X-SA-Exim-Version: 4.2.1 (built Mon, 26 Dec 2011 16:57:07 +0000)
X-SA-Exim-Scanned: Yes (on gamay.tools.ietf.org)
Resent-To: draft-ietf-bess-orf-covering-prefixes.all@ietf.org
Resent-Message-Id: <20150306221601.E7E0B1A0334@ietfa.amsl.com>
Resent-Date: Fri,  6 Mar 2015 14:16:01 -0800 (PST)
Resent-From: adrian@olddog.co.uk
Archived-At: <http://mailarchive.ietf.org/arch/msg/draft-ietf-bess-orf-covering-prefixes.all@tools/xUu0Xc-lPXtNg6DVdrxHaz60d_o>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/MOc1upiPFTINMLgsF0s8z-avukM>
X-Mailman-Approved-At: Mon, 09 Mar 2015 01:57:08 -0700
Cc: draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org
Subject: [bess] Approved draft-ietf-bess-orf-covering-prefixes
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: adrian@olddog.co.uk
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 06 Mar 2015 22:16:02 -0000

Hi,

This I-D has been updated to fix comments raised in IESG review.

No notes are required.

The document is approved.

Secretariat (Bcc) please advance the document.

Thanks,
Adrian


From nobody Mon Mar  9 04:11:08 2015
Return-Path: <tmmorin.orange@gmail.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1BDEA1A87EA for <bess@ietfa.amsl.com>; Mon,  9 Mar 2015 04:11:07 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.176
X-Spam-Level: 
X-Spam-Status: No, score=-1.176 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, MIME_HTML_ONLY=0.723, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ra2j-2wdOZH3 for <bess@ietfa.amsl.com>; Mon,  9 Mar 2015 04:11:05 -0700 (PDT)
Received: from mail-wi0-x22d.google.com (mail-wi0-x22d.google.com [IPv6:2a00:1450:400c:c05::22d]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9333D1A87AD for <bess@ietf.org>; Mon,  9 Mar 2015 04:11:05 -0700 (PDT)
Received: by wibbs8 with SMTP id bs8so19268027wib.4 for <bess@ietf.org>; Mon, 09 Mar 2015 04:11:04 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=sender:message-id:date:from:organization:user-agent:mime-version:to :subject:content-type:content-transfer-encoding; bh=DSafgsNf9lsX1AZaUtuFYLN9/LZHv1OHMw1SrqocSY8=; b=GtxNKsRO4MiaTQkpEC6OxEx83KfOqA2Z5+EL6VglDKUsYwvEbfpQ9IxHyU+4blur+6 OqtH+D964DUKvUIjIJPo4Feb9Rof6vSi7zzyWFG6twky/5NskQ5icGPAXVkgTABXrcIO r32eeTZhhi0gBuCwd2mSPCzG/sy/6pPwYeFgnn+9AMDy7VBR81SjD4ba4miMyLhhT6/7 DCpIyLPfvOk0fkj9JxAOAWm1nR1Jwes0fwgFvC9lDSzKZWLFhJrLW9ThgSqgxtrdqd0a 7OKh/f5+K5Uc6NNT3A7F1VMv7+Cto9z5f13JLwdvA1g9bLHXfHmrLqxOxCgzKurPmdxX QYLw==
X-Received: by 10.194.121.10 with SMTP id lg10mr57900143wjb.71.1425899464440;  Mon, 09 Mar 2015 04:11:04 -0700 (PDT)
Received: from [127.0.0.1] (ARennes-652-1-16-41.w2-10.abo.wanadoo.fr. [2.10.135.41]) by mx.google.com with ESMTPSA id e2sm27775096wjy.46.2015.03.09.04.11.03 for <bess@ietf.org> (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 09 Mar 2015 04:11:03 -0700 (PDT)
Sender: Thomas Morin <tmmorin.orange@gmail.com>
Message-ID: <54FD7FC5.5060200@orange.com>
Date: Mon, 09 Mar 2015 12:11:01 +0100
From: Thomas Morin <thomas.morin@orange.com>
Organization: Orange
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: "bess@ietf.org" <bess@ietf.org>
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/9d7n78e9CjSL9Lr5MMy6_mJJCxo>
Subject: [bess] Draft agenda for Dallas
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Mar 2015 11:11:07 -0000

<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=utf-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    Hi working group,<br>
    <br>
    Here is the draft agenda for our meeting in Dallas, scheduled
    Wednesday, March 25, 9:00-11:30.<br>
    <br>
    Note well that this agenda is still subject to change.<br>
    Feel free to comment...<br>
    <br>
    Best,<br>
    <br>
    -Thomas/Martin<br>
    <br>
    <a class="moz-txt-link-freetext" href="http://www.ietf.org/proceedings/92/agenda/agenda-92-bess">http://www.ietf.org/proceedings/92/agenda/agenda-92-bess</a><br>
    <pre>Administrivia and WG Status
chairs, 10 min

draft-xu-bess-l3vpn-prefix-orf-00 
draft-xu-bess-virtual-subnet-rib-reduction-00 
Xiaohu X., 20 min

draft-morin-bess-mvpn-fast-failover-01 
Robert K., 5 min

draft-dolganow-bess-mvpn-expl-track-00
Eric R., 15 min

draft-kebler-bess-sa-pref-00 
Robert K., 5 min

draft-ietf-pta-flags-00 
Eric R., 20 min

draft-tsingh-bess-pbb-evpn-yang-cfg-00 
Kishore T., 15 min

draft-mohanty-l2vpn-evpn-df-election-01 
Satya M., 15 min
</pre>
    <br>
  </body>
</html>


From nobody Mon Mar  9 04:21:10 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 799C01A8842 for <bess@ietfa.amsl.com>; Mon,  9 Mar 2015 04:21:09 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.21
X-Spam-Level: 
X-Spam-Status: No, score=-4.21 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WLyszz7vGeJC for <bess@ietfa.amsl.com>; Mon,  9 Mar 2015 04:21:07 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 445A11A8841 for <bess@ietf.org>; Mon,  9 Mar 2015 04:21:07 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml401-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQA71067; Mon, 09 Mar 2015 11:21:05 +0000 (GMT)
Received: from NKGEML403-HUB.china.huawei.com (10.98.56.34) by lhreml401-hub.china.huawei.com (10.201.5.240) with Microsoft SMTP Server (TLS) id 14.3.158.1; Mon, 9 Mar 2015 11:21:03 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.146]) by nkgeml403-hub.china.huawei.com ([10.98.56.34]) with mapi id 14.03.0158.001; Mon, 9 Mar 2015 19:21:00 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: Thomas Morin <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Draft agenda for Dallas
Thread-Index: AQHQWlnE9+q1eNhXHk2MiyE4rrSAbZ0T/7ht
Date: Mon, 9 Mar 2015 11:20:59 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F847A08@nkgeml501-mbs.china.huawei.com>
References: <54FD7FC5.5060200@orange.com>
In-Reply-To: <54FD7FC5.5060200@orange.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.135.23.94]
Content-Type: multipart/alternative; boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F847A08nkgeml501mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/JRisL8TWL-ci9M9Y96Xb5scJpQ4>
Subject: Re: [bess] Draft agenda for Dallas
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Mar 2015 11:21:09 -0000

--_000_DD5FC8DE455C3348B94340C0AB5517334F847A08nkgeml501mbschi_
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi Thomas,

Sorry for late requesting. I would like to request two slots for the follow=
ing two drafts:

1. draft-hao-bess-inter-nvo3-vpn-01

  Presenter: Lucy Yong

  Timer: 10 min

2.draft-hao-bess-evpn-df-handshaking-00

  Presenter: Weiguo Hao

  Timer: 10 min



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org] on behalf of Thomas Morin [thomas.morin@=
orange.com]
Sent: Monday, March 09, 2015 19:11
To: bess@ietf.org
Subject: [bess] Draft agenda for Dallas

Hi working group,

Here is the draft agenda for our meeting in Dallas, scheduled Wednesday, Ma=
rch 25, 9:00-11:30.

Note well that this agenda is still subject to change.
Feel free to comment...

Best,

-Thomas/Martin

http://www.ietf.org/proceedings/92/agenda/agenda-92-bess

Administrivia and WG Status
chairs, 10 min

draft-xu-bess-l3vpn-prefix-orf-00
draft-xu-bess-virtual-subnet-rib-reduction-00
Xiaohu X., 20 min

draft-morin-bess-mvpn-fast-failover-01
Robert K., 5 min

draft-dolganow-bess-mvpn-expl-track-00
Eric R., 15 min

draft-kebler-bess-sa-pref-00
Robert K., 5 min

draft-ietf-pta-flags-00
Eric R., 20 min

draft-tsingh-bess-pbb-evpn-yang-cfg-00
Kishore T., 15 min

draft-mohanty-l2vpn-evpn-df-election-01
Satya M., 15 min



--_000_DD5FC8DE455C3348B94340C0AB5517334F847A08nkgeml501mbschi_
Content-Type: text/html; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Diso-8859-=
1">
<style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
</head>
<body bgcolor=3D"#ffffff" fPStyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3" face=3D"Calibri">Hi Thomas,</font></p>
<p><font size=3D"3" face=3D"Calibri">Sorry for late requesting. I would lik=
e to request two slots for the following two drafts:</font></p>
<p><font size=3D"3" face=3D"Calibri">1. <span style=3D"FONT-FAMILY: 'Courie=
r New'; FONT-SIZE: 12pt; mso-fareast-font-family: SimSun; mso-ansi-language=
: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA" lang=3D"EN-=
US">
d</span><span style=3D"FONT-FAMILY: 'Courier New'; FONT-SIZE: 12pt; mso-far=
east-font-family: SimSun; mso-ansi-language: EN-US; mso-fareast-language: E=
N-US; mso-bidi-language: AR-SA" lang=3D"EN-US">raft-</span><span style=3D"F=
ONT-FAMILY: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: SimSun=
; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language:=
 AR-SA" lang=3D"EN-US">hao</span><span style=3D"FONT-FAMILY: 'Courier New';=
 FONT-SIZE: 12pt; mso-fareast-font-family: SimSun; mso-ansi-language: EN-US=
; mso-fareast-language: EN-US; mso-bidi-language: AR-SA" lang=3D"EN-US">-</=
span><span style=3D"FONT-FAMILY: 'Courier New'; FONT-SIZE: 12pt; mso-fareas=
t-font-family: &#23435;&#20307;; mso-ansi-language: EN-US; mso-fareast-lang=
uage: ZH-CN; mso-bidi-language: AR-SA; mso-fareast-theme-font: minor-fareas=
t" lang=3D"EN-US">bess</span><span style=3D"FONT-FAMILY: 'Courier New'; FON=
T-SIZE: 12pt; mso-fareast-font-family: SimSun; mso-ansi-language: EN-US; ms=
o-fareast-language: EN-US; mso-bidi-language: AR-SA" lang=3D"EN-US">-</span=
><span style=3D"FONT-FAMILY: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-fo=
nt-family: SimSun; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; m=
so-bidi-language: AR-SA" lang=3D"EN-US">inter</span><span style=3D"FONT-FAM=
ILY: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: SimSun; mso-a=
nsi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"=
 lang=3D"EN-US">-</span><span style=3D"FONT-FAMILY: 'Courier New'; FONT-SIZ=
E: 12pt; mso-fareast-font-family: SimSun; mso-ansi-language: EN-US; mso-far=
east-language: ZH-CN; mso-bidi-language: AR-SA" lang=3D"EN-US">nvo3-vpn-0</=
span><span style=3D"FONT-FAMILY: 'Courier New'; FONT-SIZE: 12pt; mso-fareas=
t-font-family: &#23435;&#20307;; mso-ansi-language: EN-US; mso-fareast-lang=
uage: ZH-CN; mso-bidi-language: AR-SA; mso-fareast-theme-font: minor-fareas=
t" lang=3D"EN-US">1</span></font></p>
<p><font size=3D"3" face=3D"Calibri"><span style=3D"FONT-FAMILY: 'Courier N=
ew'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-l=
anguage: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-=
fareast-theme-font: minor-fareast" lang=3D"EN-US">&nbsp; Presenter:
 Lucy Yong</span></font></p>
<p><font size=3D"3" face=3D"Calibri"><span style=3D"FONT-FAMILY: 'Courier N=
ew'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-l=
anguage: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-=
fareast-theme-font: minor-fareast" lang=3D"EN-US">&nbsp; Timer:
 10 min</span></font></p>
<p><font size=3D"3" face=3D"Calibri"><span style=3D"FONT-FAMILY: 'Courier N=
ew'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-l=
anguage: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-=
fareast-theme-font: minor-fareast" lang=3D"EN-US">2.<span style=3D"FONT-FAM=
ILY: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: SimSun; mso-a=
nsi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA"=
 lang=3D"EN-US">draft-hao-</span><span style=3D"FONT-FAMILY: 'Courier New';=
 FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-langu=
age: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-fare=
ast-theme-font: minor-fareast" lang=3D"EN-US">bess</span><span style=3D"FON=
T-FAMILY: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: SimSun; =
mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: A=
R-SA" lang=3D"EN-US">-</span><span style=3D"FONT-FAMILY: 'Courier New'; FON=
T-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-language:=
 EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-fareast-=
theme-font: minor-fareast" lang=3D"EN-US">evpn-df</span><span style=3D"FONT=
-FAMILY: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: SimSun; m=
so-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR=
-SA" lang=3D"EN-US">-</span><span style=3D"FONT-FAMILY: 'Courier New'; FONT=
-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-language: =
EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-fareast-t=
heme-font: minor-fareast" lang=3D"EN-US">handshaking-</span><span style=3D"=
FONT-FAMILY: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: SimSu=
n; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language=
: AR-SA" lang=3D"EN-US">0</span><span style=3D"FONT-FAMILY: 'Courier New'; =
FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-langua=
ge: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-farea=
st-theme-font: minor-fareast" lang=3D"EN-US">0</span></span></font></p>
<p><font size=3D"3" face=3D"Calibri"><span style=3D"FONT-FAMILY: 'Courier N=
ew'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-l=
anguage: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-=
fareast-theme-font: minor-fareast" lang=3D"EN-US"><span style=3D"FONT-FAMIL=
Y: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307=
;; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language=
: AR-SA; mso-fareast-theme-font: minor-fareast" lang=3D"EN-US">&nbsp;
 Presenter: Weiguo Hao</span></span></font></p>
<p><font size=3D"3" face=3D"Calibri"><span style=3D"FONT-FAMILY: 'Courier N=
ew'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-l=
anguage: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-=
fareast-theme-font: minor-fareast" lang=3D"EN-US"><span style=3D"FONT-FAMIL=
Y: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307=
;; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language=
: AR-SA; mso-fareast-theme-font: minor-fareast" lang=3D"EN-US">&nbsp;
 Timer: 10 min</span></span></font></p>
<p><font size=3D"3" face=3D"Calibri"><span style=3D"FONT-FAMILY: 'Courier N=
ew'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-l=
anguage: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-=
fareast-theme-font: minor-fareast" lang=3D"EN-US"><span style=3D"FONT-FAMIL=
Y: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307=
;; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language=
: AR-SA; mso-fareast-theme-font: minor-fareast" lang=3D"EN-US"></span></spa=
n></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri"><span style=3D"FONT-FAMILY: 'Courier N=
ew'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-l=
anguage: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-=
fareast-theme-font: minor-fareast" lang=3D"EN-US"><span style=3D"FONT-FAMIL=
Y: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307=
;; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language=
: AR-SA; mso-fareast-theme-font: minor-fareast" lang=3D"EN-US">Thanks,</spa=
n></span></font></p>
<p><font size=3D"3" face=3D"Calibri"><span style=3D"FONT-FAMILY: 'Courier N=
ew'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307;; mso-ansi-l=
anguage: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language: AR-SA; mso-=
fareast-theme-font: minor-fareast" lang=3D"EN-US"><span style=3D"FONT-FAMIL=
Y: 'Courier New'; FONT-SIZE: 12pt; mso-fareast-font-family: &#23435;&#20307=
;; mso-ansi-language: EN-US; mso-fareast-language: ZH-CN; mso-bidi-language=
: AR-SA; mso-fareast-theme-font: minor-fareast" lang=3D"EN-US">weiguo</span=
></span></font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF636469"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> BESS [bess-bounces@ietf.org] on beha=
lf of Thomas Morin [thomas.morin@orange.com]<br>
<b>Sent:</b> Monday, March 09, 2015 19:11<br>
<b>To:</b> bess@ietf.org<br>
<b>Subject:</b> [bess] Draft agenda for Dallas<br>
</font><br>
</div>
<div></div>
<div>Hi working group,<br>
<br>
Here is the draft agenda for our meeting in Dallas, scheduled Wednesday, Ma=
rch 25, 9:00-11:30.<br>
<br>
Note well that this agenda is still subject to change.<br>
Feel free to comment...<br>
<br>
Best,<br>
<br>
-Thomas/Martin<br>
<br>
<a class=3D"moz-txt-link-freetext" href=3D"http://www.ietf.org/proceedings/=
92/agenda/agenda-92-bess" target=3D"_blank">http://www.ietf.org/proceedings=
/92/agenda/agenda-92-bess</a><br>
<pre>Administrivia and WG Status
chairs, 10 min

draft-xu-bess-l3vpn-prefix-orf-00=20
draft-xu-bess-virtual-subnet-rib-reduction-00=20
Xiaohu X., 20 min

draft-morin-bess-mvpn-fast-failover-01=20
Robert K., 5 min

draft-dolganow-bess-mvpn-expl-track-00
Eric R., 15 min

draft-kebler-bess-sa-pref-00=20
Robert K., 5 min

draft-ietf-pta-flags-00=20
Eric R., 20 min

draft-tsingh-bess-pbb-evpn-yang-cfg-00=20
Kishore T., 15 min

draft-mohanty-l2vpn-evpn-df-election-01=20
Satya M., 15 min
</pre>
<br>
</div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F847A08nkgeml501mbschi_--


From nobody Mon Mar  9 09:09:10 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B6ABA1A8722 for <bess@ietfa.amsl.com>; Mon,  9 Mar 2015 09:09:08 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id b9Lr7cDImhne for <bess@ietfa.amsl.com>; Mon,  9 Mar 2015 09:09:06 -0700 (PDT)
Received: from alln-iport-6.cisco.com (alln-iport-6.cisco.com [173.37.142.93]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8E43E1A014B for <bess@ietf.org>; Mon,  9 Mar 2015 09:09:06 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=4325; q=dns/txt; s=iport; t=1425917346; x=1427126946; h=from:to:subject:date:message-id:in-reply-to:mime-version; bh=x/GDJIPeGRrmDzG0/F2Jxo//KLZwYZFD3w/6Rj+mZPY=; b=h7jljMkd+MseElPQ1IoNun4kHv4+8MYuGCBofL3nIzK5DjTgr9ZUdKLL 5pcAMAEW2Jzo89S9iCRwzXNGOr0S2OxTVL5SvBgelk78JTUWMbUnhnPIm X9zNepPRb0Yjv9eLMRXFznR442IhS7ubSkZEmbLiWjbm0mSQUaeNFDb0C 8=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0ANBQA1xf1U/4QNJK1cgkNDUloEwmeFcAKBJ00BAQEBAQF8hA8BAgSBCwEIBA0DAQIoORQJCAIEARIJiCYNwDoBAQEHAQEBAR6LF4RdGIQtBZAPiVOTdCODbm+BRH8BAQE
X-IronPort-AV: E=Sophos;i="5.11,368,1422921600";  d="scan'208,217";a="130264375"
Received: from alln-core-10.cisco.com ([173.36.13.132]) by alln-iport-6.cisco.com with ESMTP; 09 Mar 2015 16:09:05 +0000
Received: from xhc-aln-x06.cisco.com (xhc-aln-x06.cisco.com [173.36.12.80]) by alln-core-10.cisco.com (8.14.5/8.14.5) with ESMTP id t29G95HS010492 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 9 Mar 2015 16:09:05 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.92]) by xhc-aln-x06.cisco.com ([173.36.12.80]) with mapi id 14.03.0195.001; Mon, 9 Mar 2015 11:09:05 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: Thomas Morin <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Draft agenda for Dallas
Thread-Index: AQHQWlnEvY4mjLNoXUafUuMKUuhAAJ0UQUAA
Date: Mon, 9 Mar 2015 16:08:42 +0000
Message-ID: <D12311EF.128559%sajassi@cisco.com>
In-Reply-To: <54FD7FC5.5060200@orange.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.128.2.56]
Content-Type: multipart/alternative; boundary="_000_D12311EF128559sajassiciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/Q44MHpia37Ze3cYz6Nox6gqP2r0>
Subject: Re: [bess] Draft agenda for Dallas
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Mar 2015 16:09:08 -0000

--_000_D12311EF128559sajassiciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable


Sorry for this late request but I=92d like to ask for the following two slo=
ts:

1) draft-ietf-bess-evpn-overlay, presenter: Ali Sajassi, Time: 7min
2) draft-sajassi-bess-evpn-etree, presenter: Ali Sajassi, Time: 7min

Cheers,
Ali

From: Thomas Morin <thomas.morin@orange.com<mailto:thomas.morin@orange.com>=
>
Organization: Orange
Date: Monday, March 9, 2015 at 3:11 AM
To: "bess@ietf.org<mailto:bess@ietf.org>" <bess@ietf.org<mailto:bess@ietf.o=
rg>>
Subject: [bess] Draft agenda for Dallas

Hi working group,

Here is the draft agenda for our meeting in Dallas, scheduled Wednesday, Ma=
rch 25, 9:00-11:30.

Note well that this agenda is still subject to change.
Feel free to comment...

Best,

-Thomas/Martin

http://www.ietf.org/proceedings/92/agenda/agenda-92-bess

Administrivia and WG Status
chairs, 10 min

draft-xu-bess-l3vpn-prefix-orf-00
draft-xu-bess-virtual-subnet-rib-reduction-00
Xiaohu X., 20 min

draft-morin-bess-mvpn-fast-failover-01
Robert K., 5 min

draft-dolganow-bess-mvpn-expl-track-00
Eric R., 15 min

draft-kebler-bess-sa-pref-00
Robert K., 5 min

draft-ietf-pta-flags-00
Eric R., 20 min

draft-tsingh-bess-pbb-evpn-yang-cfg-00
Kishore T., 15 min

draft-mohanty-l2vpn-evpn-df-election-01
Satya M., 15 min



--_000_D12311EF128559sajassiciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <F69EF88C01FD9A4F9753CACDC9F8A5D7@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div><br>
</div>
<div>Sorry for this late request but I=92d like to ask for the following tw=
o slots:</div>
<div><br>
</div>
<div>1)&nbsp;draft-ietf-bess-evpn-overlay, presenter: Ali Sajassi, Time: 7m=
in</div>
<div>2) draft-sajassi-bess-evpn-etree, presenter: Ali Sajassi, Time: 7min</=
div>
<div><br>
</div>
<div>Cheers,</div>
<div>Ali</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Thomas Morin &lt;<a href=3D"m=
ailto:thomas.morin@orange.com">thomas.morin@orange.com</a>&gt;<br>
<span style=3D"font-weight:bold">Organization: </span>Orange<br>
<span style=3D"font-weight:bold">Date: </span>Monday, March 9, 2015 at 3:11=
 AM<br>
<span style=3D"font-weight:bold">To: </span>&quot;<a href=3D"mailto:bess@ie=
tf.org">bess@ietf.org</a>&quot; &lt;<a href=3D"mailto:bess@ietf.org">bess@i=
etf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>[bess] Draft agenda for Da=
llas<br>
</div>
<div><br>
</div>
<div>
<div text=3D"#000000" bgcolor=3D"#FFFFFF">Hi working group,<br>
<br>
Here is the draft agenda for our meeting in Dallas, scheduled Wednesday, Ma=
rch 25, 9:00-11:30.<br>
<br>
Note well that this agenda is still subject to change.<br>
Feel free to comment...<br>
<br>
Best,<br>
<br>
-Thomas/Martin<br>
<br>
<a class=3D"moz-txt-link-freetext" href=3D"http://www.ietf.org/proceedings/=
92/agenda/agenda-92-bess">http://www.ietf.org/proceedings/92/agenda/agenda-=
92-bess</a><br>
<pre>Administrivia and WG Status
chairs, 10 min

draft-xu-bess-l3vpn-prefix-orf-00=20
draft-xu-bess-virtual-subnet-rib-reduction-00=20
Xiaohu X., 20 min

draft-morin-bess-mvpn-fast-failover-01=20
Robert K., 5 min

draft-dolganow-bess-mvpn-expl-track-00
Eric R., 15 min

draft-kebler-bess-sa-pref-00=20
Robert K., 5 min

draft-ietf-pta-flags-00=20
Eric R., 20 min

draft-tsingh-bess-pbb-evpn-yang-cfg-00=20
Kishore T., 15 min

draft-mohanty-l2vpn-evpn-df-election-01=20
Satya M., 15 min
</pre>
<br>
</div>
</div>
</span>
</body>
</html>

--_000_D12311EF128559sajassiciscocom_--


From nobody Mon Mar  9 12:15:58 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 710941ABD3C; Mon,  9 Mar 2015 12:15:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pd3JID5u6pcf; Mon,  9 Mar 2015 12:15:47 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 378391A92B3; Mon,  9 Mar 2015 12:15:47 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p3
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150309191547.20070.54256.idtracker@ietfa.amsl.com>
Date: Mon, 09 Mar 2015 12:15:47 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/epF24_UeDUmLnyQbu8Kk7fSEGEI>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Mar 2015 19:15:54 -0000

IESG has approved the document and state has been changed to Approved-announcement sent
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Mon Mar  9 12:15:59 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 969281ABD8F; Mon,  9 Mar 2015 12:15:54 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 710941ABD3C; Mon,  9 Mar 2015 12:15:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pd3JID5u6pcf; Mon,  9 Mar 2015 12:15:47 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 378391A92B3; Mon,  9 Mar 2015 12:15:47 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p3
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150309191547.20070.54256.idtracker@ietfa.amsl.com>
Date: Mon, 09 Mar 2015 12:15:47 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/epF24_UeDUmLnyQbu8Kk7fSEGEI>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Mar 2015 19:15:54 -0000

IESG has approved the document and state has been changed to Approved-announcement sent
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Mon Mar  9 12:16:05 2015
Return-Path: <iesg-secretary@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 085911A92B0; Mon,  9 Mar 2015 12:16:00 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.9
X-Spam-Level: 
X-Spam-Status: No, score=-101.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, USER_IN_WHITELIST=-100] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 68MxJ7ehVMAL; Mon,  9 Mar 2015 12:15:58 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 59FB31A92F1; Mon,  9 Mar 2015 12:15:47 -0700 (PDT)
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
From: The IESG <iesg-secretary@ietf.org>
To: "IETF-Announce" <ietf-announce@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p3
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150309191547.20070.15099.idtracker@ietfa.amsl.com>
Date: Mon, 09 Mar 2015 12:15:47 -0700
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/NYY6ePrQXef-oLQQktHqdAOEaS8>
Cc: bess chair <bess-chairs@tools.ietf.org>, bess mailing list <bess@ietf.org>, RFC Editor <rfc-editor@rfc-editor.org>
Subject: [bess] Protocol Action: 'Covering Prefixes Outbound Route Filter for BGP-4' to Proposed Standard (draft-ietf-bess-orf-covering-prefixes-06.txt)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Mar 2015 19:16:00 -0000

The IESG has approved the following document:
- 'Covering Prefixes Outbound Route Filter for BGP-4'
  (draft-ietf-bess-orf-covering-prefixes-06.txt) as Proposed Standard

This document is the product of the BGP Enabled Services Working Group.

The IESG contact persons are Alia Atlas and Adrian Farrel.

A URL of this Internet Draft is:
http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/




Technical Summary

  A BGP speaker can send Outbound Route Filters (ORF) to a peer.  The
  peer uses ORFs to filter routing updates that it sends to the BGP
  speaker.  Using ORF, a BGP speaker can realize a "route pull"
  paradigm, in which the BGP speaker, on demand, pulls certain routes
  from the peer.

  This document defines a new Outbound Route Filter (ORF) type, called
  the "Covering Prefixes ORF (CP-ORF)".  CP-ORF is applicable in
  Virtual Hub-and-Spoke VPNs.  It also is applicable in BGP/MPLS
  Ethernet VPN (EVPN) networks.

Working Group Summary

  There has been no controversy relating to the technical aspect of
  this Document. A late IPR disclosure was made against this document
  (see question 8). The WG did not oppose to the progress of the
  document.

Document Quality

  The WG Chairs are aware of one implementation plan of the
  technologies described in this document. It should be noted that the
  contexts (Virtual Hub-and-Spoke VPNs and EVPNs) in which this
  Document applies are fairly recent thus explaining that
  implementations are not yet available
  
Personnel

  Martin Vigoureux is the Document Shepherd
  Adrian Farrel is the responsible AD


From nobody Mon Mar  9 12:57:20 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A65791A8982; Mon,  9 Mar 2015 12:57:17 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5HRyho4TpvdQ; Mon,  9 Mar 2015 12:57:15 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id EF6131AC3F4; Mon,  9 Mar 2015 12:57:11 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p3
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150309195711.10831.75874.idtracker@ietfa.amsl.com>
Date: Mon, 09 Mar 2015 12:57:11 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/O9aPfdI-D5JdL57itIfL3RiFaR4>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Mar 2015 19:57:18 -0000

IESG state changed to RFC Ed Queue from Approved-announcement sent
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Mon Mar  9 12:57:21 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 507E81A8F48; Mon,  9 Mar 2015 12:57:18 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A65791A8982; Mon,  9 Mar 2015 12:57:17 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5HRyho4TpvdQ; Mon,  9 Mar 2015 12:57:15 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id EF6131AC3F4; Mon,  9 Mar 2015 12:57:11 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p3
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150309195711.10831.75874.idtracker@ietfa.amsl.com>
Date: Mon, 09 Mar 2015 12:57:11 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/O9aPfdI-D5JdL57itIfL3RiFaR4>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Mar 2015 19:57:18 -0000

IESG state changed to RFC Ed Queue from Approved-announcement sent
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Mon Mar  9 14:20:23 2015
Return-Path: <internet-drafts@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E66651ACD7E; Mon,  9 Mar 2015 14:20:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3NukFhyOZfRY; Mon,  9 Mar 2015 14:20:19 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 790E61A0126; Mon,  9 Mar 2015 14:20:19 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <i-d-announce@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p3
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150309212019.6241.29843.idtracker@ietfa.amsl.com>
Date: Mon, 09 Mar 2015 14:20:19 -0700
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/5J-8xl7mwivswxwylW2zmvvEcuQ>
Cc: bess@ietf.org
Subject: [bess] I-D Action: draft-ietf-bess-evpn-prefix-advertisement-01.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 09 Mar 2015 21:20:21 -0000

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

        Title           : IP Prefix Advertisement in EVPN
        Authors         : Jorge Rabadan
                          Wim Henderickx
                          Senad Palislamovic
                          Florin Balus
                          Aldrin Isaac
	Filename        : draft-ietf-bess-evpn-prefix-advertisement-01.txt
	Pages           : 24
	Date            : 2015-03-09

Abstract:
   EVPN provides a flexible control plane that allows intra-subnet
   connectivity in an IP/MPLS and/or an NVO-based network. In NVO
   networks, there is also a need for a dynamic and efficient inter-
   subnet connectivity across Tenant Systems and End Devices that can be
   physical or virtual and may not support their own routing protocols.
   This document defines a new EVPN route type for the advertisement of
   IP Prefixes and explains some use-case examples where this new route-
   type is used.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-bess-evpn-prefix-advertisement/

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-bess-evpn-prefix-advertisement-01

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=draft-ietf-bess-evpn-prefix-advertisement-01


Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

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


From nobody Mon Mar  9 17:37:21 2015
Return-Path: <satyamoh@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BE4611ACE73 for <bess@ietfa.amsl.com>; Mon,  9 Mar 2015 17:37:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id nqWF5jY_wvRZ for <bess@ietfa.amsl.com>; Mon,  9 Mar 2015 17:37:16 -0700 (PDT)
Received: from rcdn-iport-4.cisco.com (rcdn-iport-4.cisco.com [173.37.86.75]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id EA27B1ACE58 for <bess@ietf.org>; Mon,  9 Mar 2015 17:37:15 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=5848; q=dns/txt; s=iport; t=1425947837; x=1427157437; h=from:to:subject:date:message-id:references:in-reply-to: mime-version; bh=uYxCDERFKUu8ozUhNL9hL9R2ayKHIuqvDwG7kCjtHEs=; b=LNwbniXtPeAiTzicco0bw0omyUWkUpqmdWgkQxg4iKak812MMdk5L9/g 9Xv4W3x7fSbOYqGTQ3HxZDyOTC/MtF22ZeX8bR+zZpybyt/zm6DOy6GRm UedxcgbH7jPla+gyfspmYupzLbu4iGPKlG6qJGXCIhsCnSVZCLynIiRj+ s=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0AoBQADPP5U/4kNJK1cgkNDUloEwHWBfIVuAoEqTQEBAQEBAXyEEAEBBHcSAgEIPwcyFBECBBMJiCYNwgUBAQEBAQUBAQEBAQEcixeEaguELQWODYICg2SFb4EaOYJvjzIjggEdgVBvAYFDfwEBAQ
X-IronPort-AV: E=Sophos;i="5.11,371,1422921600";  d="scan'208,217";a="402283855"
Received: from alln-core-4.cisco.com ([173.36.13.137]) by rcdn-iport-4.cisco.com with ESMTP; 10 Mar 2015 00:37:16 +0000
Received: from xhc-rcd-x08.cisco.com (xhc-rcd-x08.cisco.com [173.37.183.82]) by alln-core-4.cisco.com (8.14.5/8.14.5) with ESMTP id t2A0bFoe026438 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Tue, 10 Mar 2015 00:37:15 GMT
Received: from xmb-aln-x07.cisco.com ([169.254.2.221]) by xhc-rcd-x08.cisco.com ([173.37.183.82]) with mapi id 14.03.0195.001; Mon, 9 Mar 2015 19:37:15 -0500
From: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: New Version Notification for draft-mohanty-bess-evpn-df-election-00.txt
Thread-Index: AQHQWrm33ylGCLL4VU6bG6WBseaAZp0UvbaA
Date: Tue, 10 Mar 2015 00:37:14 +0000
Message-ID: <D1238900.27EB7%satyamoh@cisco.com>
References: <20150309223804.4522.85336.idtracker@ietfa.amsl.com>
In-Reply-To: <20150309223804.4522.85336.idtracker@ietfa.amsl.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.89.12.96]
Content-Type: multipart/alternative; boundary="_000_D123890027EB7satyamohciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/MX9u-e8E98mTlp5DSV_ol91ByDk>
Subject: Re: [bess] New Version Notification for draft-mohanty-bess-evpn-df-election-00.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 10 Mar 2015 00:37:20 -0000

--_000_D123890027EB7satyamohciscocom_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

FYI, below is the updated draft, with the corrected name from L2VPN to BESS=
.
[was previously draft-mohanty-l2vpn-evpn-df-election-00]

Also, the actual hash function is clarified here.

Thanks,
--Satya

On 3/9/15 3:38 PM, "internet-drafts@ietf.org<mailto:internet-drafts@ietf.or=
g>" <internet-drafts@ietf.org<mailto:internet-drafts@ietf.org>> wrote:


A new version of I-D, draft-mohanty-bess-evpn-df-election-00.txt
has been successfully submitted by Satya Ranjan Mohanty and posted to the
IETF repository.

Name: draft-mohanty-bess-evpn-df-election
Revision: 00
Title: A new Designated Forwarder Election for the EVPN
Document date: 2015-03-07
Group: Individual Submission
Pages: 11
URL:            http://www.ietf.org/internet-drafts/draft-mohanty-bess-evpn=
-df-election-00.txt
Status:         https://datatracker.ietf.org/doc/draft-mohanty-bess-evpn-df=
-election/
Htmlized:       http://tools.ietf.org/html/draft-mohanty-bess-evpn-df-elect=
ion-00


Abstract:
   This document describes an improved EVPN Designated Forwarder
   Election (DF) algorithm which can be used to enhance operational
   experience in terms of convergence speed and robustness over a WAN
   deploying EVPN




Please note that it may take a couple of minutes from the time of submissio=
n
until the htmlized version and diff are available at tools.ietf.org.

The IETF Secretariat

--_000_D123890027EB7satyamohciscocom_
Content-Type: text/html; charset="us-ascii"
Content-ID: <FC54B1F51BC5384A9B9637341358088D@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; font-family: Calibri, sans-serif; font-size: 14=
px; color: rgb(0, 0, 0); ">
<div>FYI, below is the updated draft, with the corrected name from L2VPN to=
 BESS.</div>
<div>
<div>[was previously draft-mohanty-l2vpn-evpn-df-election-00]</div>
</div>
<div><br>
</div>
<div>Also, the actual hash function is clarified here.</div>
<div><br>
</div>
<div>Thanks,</div>
<div>--Satya</div>
<div><br>
</div>
<div>On 3/9/15 3:38 PM, &quot;<a href=3D"mailto:internet-drafts@ietf.org">i=
nternet-drafts@ietf.org</a>&quot; &lt;<a href=3D"mailto:internet-drafts@iet=
f.org">internet-drafts@ietf.org</a>&gt; wrote:</div>
<div><br>
</div>
<blockquote id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE" style=3D"BORDER-LEFT:=
 #b5c4df 5 solid; PADDING:0 0 0 5; MARGIN:0 0 0 5;">
<div><br>
</div>
<div>A new version of I-D, draft-mohanty-bess-evpn-df-election-00.txt</div>
<div>has been successfully submitted by Satya Ranjan Mohanty and posted to =
the</div>
<div>IETF repository.</div>
<div><br>
</div>
<div>Name:<span class=3D"Apple-tab-span" style=3D"white-space:pre"> </span>=
<span class=3D"Apple-tab-span" style=3D"white-space:pre"></span>draft-mohan=
ty-bess-evpn-df-election</div>
<div>Revision:<span class=3D"Apple-tab-span" style=3D"white-space:pre"> </s=
pan>00</div>
<div>Title:<span class=3D"Apple-tab-span" style=3D"white-space:pre"> </span=
><span class=3D"Apple-tab-span" style=3D"white-space:pre"></span>A new Desi=
gnated Forwarder Election for the EVPN</div>
<div>Document date:<span class=3D"Apple-tab-span" style=3D"white-space:pre"=
> </span>2015-03-07</div>
<div>Group:<span class=3D"Apple-tab-span" style=3D"white-space:pre"> </span=
><span class=3D"Apple-tab-span" style=3D"white-space:pre"></span>Individual=
 Submission</div>
<div>Pages:<span class=3D"Apple-tab-span" style=3D"white-space:pre"> </span=
><span class=3D"Apple-tab-span" style=3D"white-space:pre"></span>11</div>
<div>URL:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;<a href=3D"http://www.ietf.org/internet-drafts/draft-mohanty-bess-evp=
n-df-election-00.txt">http://www.ietf.org/internet-drafts/draft-mohanty-bes=
s-evpn-df-election-00.txt</a></div>
<div>Status:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <a href=3D"htt=
ps://datatracker.ietf.org/doc/draft-mohanty-bess-evpn-df-election/">
https://datatracker.ietf.org/doc/draft-mohanty-bess-evpn-df-election/</a></=
div>
<div>Htmlized:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <a href=3D"http://tools.=
ietf.org/html/draft-mohanty-bess-evpn-df-election-00">
http://tools.ietf.org/html/draft-mohanty-bess-evpn-df-election-00</a></div>
<div><br>
</div>
<div><br>
</div>
<div>Abstract:</div>
<div>&nbsp;&nbsp; This document describes an improved EVPN Designated Forwa=
rder</div>
<div>&nbsp;&nbsp; Election (DF) algorithm which can be used to enhance oper=
ational</div>
<div>&nbsp;&nbsp; experience in terms of convergence speed and robustness o=
ver a WAN</div>
<div>&nbsp;&nbsp; deploying EVPN</div>
<div><br>
</div>
<div>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</div>
<div><br>
</div>
<div><br>
</div>
<div>Please note that it may take a couple of minutes from the time of subm=
ission</div>
<div>until the htmlized version and diff are available at tools.ietf.org.</=
div>
<div><br>
</div>
<div>The IETF Secretariat</div>
</blockquote>
</body>
</html>

--_000_D123890027EB7satyamohciscocom_--


From nobody Tue Mar 10 16:05:42 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6450A1A90B4; Tue, 10 Mar 2015 16:05:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id izJbX3vJIW8Z; Tue, 10 Mar 2015 16:05:40 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id A545F1A90B9; Tue, 10 Mar 2015 16:05:31 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p3
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150310230531.15388.56427.idtracker@ietfa.amsl.com>
Date: Tue, 10 Mar 2015 16:05:31 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/AvyZmoSZyfolnTPsK6vBRjDnELA>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 10 Mar 2015 23:05:41 -0000

IANA action state changed to In Progress
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Tue Mar 10 16:05:44 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 83D0A1A90B5; Tue, 10 Mar 2015 16:05:41 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6450A1A90B4; Tue, 10 Mar 2015 16:05:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id izJbX3vJIW8Z; Tue, 10 Mar 2015 16:05:40 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id A545F1A90B9; Tue, 10 Mar 2015 16:05:31 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.0.p3
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150310230531.15388.56427.idtracker@ietfa.amsl.com>
Date: Tue, 10 Mar 2015 16:05:31 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/AvyZmoSZyfolnTPsK6vBRjDnELA>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 10 Mar 2015 23:05:41 -0000

IANA action state changed to In Progress
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Wed Mar 11 16:05:30 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1CE171A892E; Wed, 11 Mar 2015 16:05:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OrTbRcrTuA3A; Wed, 11 Mar 2015 16:05:27 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 3157E1A890F; Wed, 11 Mar 2015 16:05:27 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150311230527.2951.94859.idtracker@ietfa.amsl.com>
Date: Wed, 11 Mar 2015 16:05:27 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/Yq27Q5oZTTFta2srRxM4oa-lOqQ>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Mar 2015 23:05:28 -0000

IANA action state changed to Waiting on Authors
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Wed Mar 11 16:05:31 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 4AFAC1A8943; Wed, 11 Mar 2015 16:05:28 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1CE171A892E; Wed, 11 Mar 2015 16:05:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OrTbRcrTuA3A; Wed, 11 Mar 2015 16:05:27 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 3157E1A890F; Wed, 11 Mar 2015 16:05:27 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150311230527.2951.94859.idtracker@ietfa.amsl.com>
Date: Wed, 11 Mar 2015 16:05:27 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/Yq27Q5oZTTFta2srRxM4oa-lOqQ>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Mar 2015 23:05:28 -0000

IANA action state changed to Waiting on Authors
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Thu Mar 12 00:05:36 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 960FB1A0545; Thu, 12 Mar 2015 00:05:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id A6UhOzfZyuW6; Thu, 12 Mar 2015 00:05:34 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id A0DB71A19F0; Thu, 12 Mar 2015 00:05:30 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150312070530.30860.87725.idtracker@ietfa.amsl.com>
Date: Thu, 12 Mar 2015 00:05:30 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/RHkWQ38PI3iWbG7NzGX4BYkUEKY>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Mar 2015 07:05:35 -0000

IANA action state changed to In Progress
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Thu Mar 12 00:05:38 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id B17B61A19F0; Thu, 12 Mar 2015 00:05:35 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 960FB1A0545; Thu, 12 Mar 2015 00:05:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id A6UhOzfZyuW6; Thu, 12 Mar 2015 00:05:34 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id A0DB71A19F0; Thu, 12 Mar 2015 00:05:30 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150312070530.30860.87725.idtracker@ietfa.amsl.com>
Date: Thu, 12 Mar 2015 00:05:30 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/RHkWQ38PI3iWbG7NzGX4BYkUEKY>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Mar 2015 07:05:35 -0000

IANA action state changed to In Progress
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Thu Mar 12 00:05:42 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A3F001A19F0; Thu, 12 Mar 2015 00:05:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xgmT2-ls6P09; Thu, 12 Mar 2015 00:05:38 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 8781D1A1A0C; Thu, 12 Mar 2015 00:05:31 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150312070531.30860.42887.idtracker@ietfa.amsl.com>
Date: Thu, 12 Mar 2015 00:05:31 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/PCimVHX7757Kem-m7uoURvyBVoo>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Mar 2015 07:05:41 -0000

IANA action state changed to Waiting on RFC Editor
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Thu Mar 12 00:05:44 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id C00C71A1A1D; Thu, 12 Mar 2015 00:05:41 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A3F001A19F0; Thu, 12 Mar 2015 00:05:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xgmT2-ls6P09; Thu, 12 Mar 2015 00:05:38 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 8781D1A1A0C; Thu, 12 Mar 2015 00:05:31 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150312070531.30860.42887.idtracker@ietfa.amsl.com>
Date: Thu, 12 Mar 2015 00:05:31 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/PCimVHX7757Kem-m7uoURvyBVoo>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Mar 2015 07:05:42 -0000

IANA action state changed to Waiting on RFC Editor
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Thu Mar 12 07:55:12 2015
Return-Path: <ben@niven-jenkins.co.uk>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8224E1A70FE; Wed, 11 Mar 2015 11:34:02 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.5
X-Spam-Level: 
X-Spam-Status: No, score=-0.5 tagged_above=-999 required=5 tests=[BAYES_05=-0.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tMEje-J1zJVH; Wed, 11 Mar 2015 11:34:00 -0700 (PDT)
Received: from mailex.mailcore.me (mailex.mailcore.me [94.136.40.61]) by ietfa.amsl.com (Postfix) with ESMTP id 237A61A702A; Wed, 11 Mar 2015 11:33:59 -0700 (PDT)
Received: from host4.velocix.com ([81.134.152.4] helo=[172.18.0.166]) by smtp04.mailcore.me with esmtpa (Exim 4.80.1) (envelope-from <ben@niven-jenkins.co.uk>) id 1YVlSA-0002Iq-CU; Wed, 11 Mar 2015 18:33:58 +0000
From: Ben Niven-Jenkins <ben@niven-jenkins.co.uk>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Date: Wed, 11 Mar 2015 18:33:54 +0000
Message-Id: <580A03D0-3A2A-4D39-80C8-030D8C9C37DB@niven-jenkins.co.uk>
To: rtg-ads@tools.ietf.org
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
X-Mailer: Apple Mail (2.2070.6)
X-Mailcore-Auth: 9600544
X-Mailcore-Domain: 172912
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/5K8ZVuVwMXNQDTCy9kKICsf8xEs>
X-Mailman-Approved-At: Thu, 12 Mar 2015 07:55:09 -0700
Cc: draft-ietf-bess-mvpn-bidir.all@tools.ietf.org, rtg-dir@ietf.org, bess@ietf.org
Subject: [bess] RtgDir review: draft-ietf-bess-mvpn-bidir-03.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Mar 2015 18:34:02 -0000

Hello,

I have been selected as the Routing Directorate reviewer for this draft. =
The Routing Directorate seeks to review all routing or routing-related =
drafts as they pass through IETF last call and IESG review, and =
sometimes on special request. The purpose of the review is to provide =
assistance to the Routing ADs. For more information about the Routing =
Directorate, please see =
=E2=80=8Bhttp://trac.tools.ietf.org/area/rtg/trac/wiki/RtgDir

Although these comments are primarily for the use of the Routing ADs, it =
would be helpful if you could consider them along with any other IETF =
Last Call comments that you receive, and strive to resolve them through =
discussion or by updating the draft.

Document: draft-ietf-bess-mvpn-bidir-03.txt
Reviewer: Ben Niven-Jenkins
Review Date: 11th March 2015
IETF LC End Date: Not known=20
Intended Status: Standards Track


Summary:
This document is basically ready for publication, but has nits that =
should be considered prior to publication.

Comments:
The draft is well written, in places I needed to re-read sentences to =
ensure I had understood them but I think this is more to do with the =
nature of the content and that multicast VPNs require complex =
descriptions rather than a reflection on the quality of the document.

Major issues:
No major issues found.

Minor issues:
No minor issues found.

Nits:
Section 1.1: C-multicast flow:
Change
  If a customer uses the "Any Source Multicast" (ASM) model, the
  some or all of the customer's C-flows may be traveling along the

to

  If a customer uses the "Any Source Multicast" (ASM) model, then
  some or all of the customer's C-flows may be traveling along the

i.e. s/the/then in the first line.

Section 1.2.4:
s/If a bidirectional P-tunnels/If bidirectional P-tunnels/

s/The method used by a given VRF used is determined/The method used by a =
given VRF is determined/

Section 3.2.2:
s/The PEs are REQUIRED to originate these routes are/The PEs REQUIRED to =
originate these routes are/

s/This document assumes that the root node address of an MP2MP LSP an IP =
address/This document assumes that the root node address of an MP2MP LSP =
is an IP address/

Section 3.2.4 states:

 In order to be compliant with this specification, an implementation
 that provides bidirectional P-tunnels MUST support one or both of the
 two P-tunnel technologies mentioned in section Section 1.2.1.

Saying implementations "MUST support one or both" sounds a bit strange =
to me and something like "MUST support at least one" sounds better (but =
both ultimately mean the same thing)?

Also the document states:

 A PE that does not provide C-BIDIR support using the "partitioned set
 of PEs" method may be deemed compliant to this specification if it
 supports the Unpartitioned Method, using either MP2MP LSPs or BIDIR-
 PIM multicast distribute trees as P-tunnels.

"may be deemed" implies there are cases where it "may not be deemed" =
compliant but I'm not sure what those are.

Do you really mean to say "is" instead of "may be"?




Regards
Ben


From nobody Thu Mar 12 07:55:13 2015
Return-Path: <iana-shared@icann.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id A2EC71A88DA; Wed, 11 Mar 2015 15:31:59 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 89D481A88D5 for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Wed, 11 Mar 2015 15:31:59 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.879
X-Spam-Level: 
X-Spam-Status: No, score=-0.879 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, MISSING_HEADERS=1.021] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HwKS1Z7x3iqv for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Wed, 11 Mar 2015 15:31:58 -0700 (PDT)
Received: from zinfandel.tools.ietf.org (zinfandel.tools.ietf.org [IPv6:2001:1890:123a::1:2a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id ACE231A0115 for <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>; Wed, 11 Mar 2015 15:31:57 -0700 (PDT)
Received: from smtp01.icann.org ([192.0.33.81]:43961 helo=smtp1.lax.icann.org) by zinfandel.tools.ietf.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.82_1-5b7a7c0-XX) (envelope-from <iana-shared@icann.org>) id 1YVpAT-0001Rm-2S for draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org; Wed, 11 Mar 2015 15:31:57 -0700
Received: from request3.lax.icann.org (request1.lax.icann.org [10.32.11.221]) by smtp1.lax.icann.org (8.13.8/8.13.8) with ESMTP id t2BMVpWT014477 for <draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org>; Wed, 11 Mar 2015 22:31:51 GMT
Received: by request3.lax.icann.org (Postfix, from userid 48) id 60F2DC2040F; Wed, 11 Mar 2015 22:31:51 +0000 (UTC)
RT-Owner: amanda.baber
From: "Amanda Baber via RT" <drafts-approval@iana.org>
In-Reply-To: <20150309191547.20070.81805.idtracker@ietfa.amsl.com>
References: <RT-Ticket-812089@icann.org> <20150309191547.20070.81805.idtracker@ietfa.amsl.com>
Message-ID: <rt-4.2.9-32314-1426113111-1024.812089-7-0@icann.org>
X-RT-Loop-Prevention: IANA
X-RT-Ticket: IANA #812089
X-Managed-BY: RT 4.2.9 (http://www.bestpractical.com/rt/)
X-RT-Originator: amanda.baber@icann.org
Content-Type: text/plain; charset="utf-8"
X-RT-Original-Encoding: utf-8
Precedence: bulk
Date: Wed, 11 Mar 2015 22:31:51 +0000
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
X-SA-Exim-Connect-IP: 192.0.33.81
X-SA-Exim-Rcpt-To: draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org
X-SA-Exim-Mail-From: iana-shared@icann.org
X-SA-Exim-Version: 4.2.1 (built Mon, 26 Dec 2011 16:24:06 +0000)
X-SA-Exim-Scanned: Yes (on zinfandel.tools.ietf.org)
Resent-To: draft-ietf-bess-orf-covering-prefixes.all@ietf.org
Resent-Message-Id: <20150311223157.ACE231A0115@ietfa.amsl.com>
Resent-Date: Wed, 11 Mar 2015 15:31:57 -0700 (PDT)
Resent-From: iana-shared@icann.org
Archived-At: <http://mailarchive.ietf.org/arch/msg/draft-ietf-bess-orf-covering-prefixes.all@tools/Md6Vvhg1jUm-Be4K_pWQJSWUhPQ>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/IPTzDkmgYnzkN9MYp_TUd87UDDw>
X-Mailman-Approved-At: Thu, 12 Mar 2015 07:55:09 -0700
Cc: draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org
Subject: [bess] [IANA #812089] Protocol Action: 'Covering Prefixes Outbound Route Filter for BGP-4' to Proposed Standard (draft-ietf-bess-orf-covering-prefixes-06.txt)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Reply-To: drafts-approval@iana.org
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Mar 2015 22:31:59 -0000

Dear Authors:

ATTENTION: A RESPONSE TO THIS MESSAGE IS NEEDED 

We've completed the IANA Actions for the following RFC-to-be:

draft-ietf-bess-orf-covering-prefixes-06

ACTION 1:

IANA has updated the reference for the following BGP Outbound Route Filtering (ORF) Type:

65	CP-ORF	[RFC-ietf-bess-orf-covering-prefixes-03]

Please see
http://www.iana.org/assignments/bgp-parameters


ACTION 2:

IANA has updated the reference for the following Transitive Opaque Extended Community Sub-Type:

0x03 	CP-ORF	[RFC-ietf-bess-orf-covering-prefixes-03]

Please see
http://www.iana.org/assignments/bgp-extended-communities


Please let us know whether the above IANA Actions look OK. As soon as we receive your confirmation, we'll notify the RFC Editor that this document's IANA Actions are complete. (If this document has a team of authors, one reply on behalf of everyone will suffice.)

We'll update the reference when the RFC Editor notifies us that they've assigned a number.

Thanks,

Amanda Baber
IANA Request Specialist
ICANN


From nobody Thu Mar 12 07:55:15 2015
Return-Path: <iana-shared@icann.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id BE23A1A03D5; Thu, 12 Mar 2015 00:00:05 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 93FD81A03C7 for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Thu, 12 Mar 2015 00:00:05 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.879
X-Spam-Level: 
X-Spam-Status: No, score=-0.879 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, MISSING_HEADERS=1.021] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wlW9PUBblHNs for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Thu, 12 Mar 2015 00:00:04 -0700 (PDT)
Received: from zinfandel.tools.ietf.org (zinfandel.tools.ietf.org [IPv6:2001:1890:123a::1:2a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 6A62B1A03AA for <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>; Thu, 12 Mar 2015 00:00:04 -0700 (PDT)
Received: from smtp01.icann.org ([192.0.33.81]:56442 helo=smtp1.lax.icann.org) by zinfandel.tools.ietf.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.82_1-5b7a7c0-XX) (envelope-from <iana-shared@icann.org>) id 1YVx6A-00080I-J8 for draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org; Thu, 12 Mar 2015 00:00:04 -0700
Received: from request3.lax.icann.org (request1.lax.icann.org [10.32.11.221]) by smtp1.lax.icann.org (8.13.8/8.13.8) with ESMTP id t2C6xuh4003334 for <draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org>; Thu, 12 Mar 2015 06:59:56 GMT
Received: by request3.lax.icann.org (Postfix, from userid 48) id 2E36FC20367; Thu, 12 Mar 2015 06:59:56 +0000 (UTC)
RT-Owner: amanda.baber
From: "Amanda Baber via RT" <drafts-approval@iana.org>
In-Reply-To: <rt-4.2.9-22915-1426123938-999.812089-7-0@icann.org>
References: <RT-Ticket-812089@icann.org> <20150309191547.20070.81805.idtracker@ietfa.amsl.com> <rt-4.2.9-32314-1426113111-1024.812089-7-0@icann.org> <CO1PR05MB442F5ABD0D80EE4AD1DCC10AE060@CO1PR05MB442.namprd05.prod.outlook.com> <rt-4.2.9-22915-1426123938-999.812089-7-0@icann.org>
Message-ID: <rt-4.2.9-30691-1426143595-1427.812089-7-0@icann.org>
X-RT-Loop-Prevention: IANA
X-RT-Ticket: IANA #812089
X-Managed-BY: RT 4.2.9 (http://www.bestpractical.com/rt/)
X-RT-Originator: amanda.baber@icann.org
Content-Type: text/plain; charset="utf-8"
X-RT-Original-Encoding: utf-8
Precedence: bulk
Date: Thu, 12 Mar 2015 06:59:56 +0000
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
X-SA-Exim-Connect-IP: 192.0.33.81
X-SA-Exim-Rcpt-To: draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org
X-SA-Exim-Mail-From: iana-shared@icann.org
X-SA-Exim-Version: 4.2.1 (built Mon, 26 Dec 2011 16:24:06 +0000)
X-SA-Exim-Scanned: Yes (on zinfandel.tools.ietf.org)
Resent-To: draft-ietf-bess-orf-covering-prefixes.all@ietf.org
Resent-Message-Id: <20150312070004.6A62B1A03AA@ietfa.amsl.com>
Resent-Date: Thu, 12 Mar 2015 00:00:04 -0700 (PDT)
Resent-From: iana-shared@icann.org
Archived-At: <http://mailarchive.ietf.org/arch/msg/draft-ietf-bess-orf-covering-prefixes.all@tools/GwLLK-orSQPP1jUcQEjoq3pBkeM>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/91VAeB-E7khtmWNjvh1tuqw35Ag>
X-Mailman-Approved-At: Thu, 12 Mar 2015 07:55:09 -0700
Cc: draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org
Subject: [bess] [IANA #812089] Protocol Action: 'Covering Prefixes Outbound Route Filter for BGP-4' to Proposed Standard (draft-ietf-bess-orf-covering-prefixes-06.txt)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Reply-To: drafts-approval@iana.org
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Mar 2015 07:00:05 -0000

Hi,

It should definitely be pointing at version 06. Here are the corrected versions:

http://www.iana.org/assignments/bgp-parameters
http://www.iana.org/assignments/bgp-extended-communities

We'll replace the draft string when the RFC Editor notifies us that they've assigned a number. In the meantime, I'll let them know that the actions are complete. 

thanks,
Amanda 

On Thu Mar 12 01:32:18 2015, rbonica@juniper.net wrote:
> Hi Amanda,
> 
> I have two questions.
> 
> The latest version of draft-ietf-bess-orf-covering prefixes is 06.
> Your reference is RFC-ietf-bess-orf-covering-prefixes-03. Does that
> make a difference?
> 
> Also, I assume that this string will be replaced with an RFC number
> when one is assigned. Is that right?
> 
> Otherwise, it looks OK.
> 
> Ron
> 
> 
> > -----Original Message-----
> > From: Amanda Baber via RT [mailto:drafts-approval@iana.org]
> > Sent: Wednesday, March 11, 2015 6:32 PM
> > Cc: draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org
> > Subject: [IANA #812089] Protocol Action: 'Covering Prefixes Outbound
> > Route
> > Filter for BGP-4' to Proposed Standard (draft-ietf-bess-orf-covering-
> > prefixes-
> > 06.txt)
> >
> > Dear Authors:
> >
> > ATTENTION: A RESPONSE TO THIS MESSAGE IS NEEDED
> >
> > We've completed the IANA Actions for the following RFC-to-be:
> >
> > draft-ietf-bess-orf-covering-prefixes-06
> >
> > ACTION 1:
> >
> > IANA has updated the reference for the following BGP Outbound Route
> > Filtering (ORF) Type:
> >
> > 65    CP-ORF  [RFC-ietf-bess-orf-covering-prefixes-03]
> >
> > Please see
> > http://www.iana.org/assignments/bgp-parameters
> >
> >
> > ACTION 2:
> >
> > IANA has updated the reference for the following Transitive Opaque
> > Extended Community Sub-Type:
> >
> > 0x03  CP-ORF  [RFC-ietf-bess-orf-covering-prefixes-03]
> >
> > Please see
> > http://www.iana.org/assignments/bgp-extended-communities
> >
> >
> > Please let us know whether the above IANA Actions look OK. As soon as
> > we
> > receive your confirmation, we'll notify the RFC Editor that this
> > document's
> > IANA Actions are complete. (If this document has a team of authors,
> > one
> > reply on behalf of everyone will suffice.)
> >
> > We'll update the reference when the RFC Editor notifies us that
> > they've
> > assigned a number.
> >
> > Thanks,
> >
> > Amanda Baber
> > IANA Request Specialist
> > ICANN
> 




From nobody Thu Mar 12 07:55:16 2015
Return-Path: <rbonica@juniper.net>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 232411A8BC3; Thu, 12 Mar 2015 01:40:53 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 04C201A8F3B for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Thu, 12 Mar 2015 01:40:53 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.235
X-Spam-Level: 
X-Spam-Status: No, score=-101.235 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_SOFTFAIL=0.665, USER_IN_WHITELIST=-100] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yRDUM1eAeYH8 for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Thu, 12 Mar 2015 01:40:51 -0700 (PDT)
Received: from zinfandel.tools.ietf.org (zinfandel.tools.ietf.org [IPv6:2001:1890:123a::1:2a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 93FE51A870E for <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>; Thu, 12 Mar 2015 01:40:48 -0700 (PDT)
Received: from mail-bn1on0142.outbound.protection.outlook.com ([157.56.110.142]:1377 helo=na01-bn1-obe.outbound.protection.outlook.com) by zinfandel.tools.ietf.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA384:256) (Exim 4.82_1-5b7a7c0-XX) (envelope-from <rbonica@juniper.net>) id 1YVyff-0004yo-KJ for draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org; Thu, 12 Mar 2015 01:40:48 -0700
Received: from CO1PR05MB444.namprd05.prod.outlook.com (10.141.73.140) by CO1PR05MB539.namprd05.prod.outlook.com (10.141.73.26) with Microsoft SMTP Server (TLS) id 15.1.99.14; Thu, 12 Mar 2015 01:31:55 +0000
Received: from CO1PR05MB442.namprd05.prod.outlook.com (10.141.73.146) by CO1PR05MB444.namprd05.prod.outlook.com (10.141.73.140) with Microsoft SMTP Server (TLS) id 15.1.106.15; Thu, 12 Mar 2015 01:31:53 +0000
Received: from CO1PR05MB442.namprd05.prod.outlook.com ([169.254.13.61]) by CO1PR05MB442.namprd05.prod.outlook.com ([169.254.13.61]) with mapi id 15.01.0106.007; Thu, 12 Mar 2015 01:31:53 +0000
From: Ronald Bonica <rbonica@juniper.net>
To: "drafts-approval@iana.org" <drafts-approval@iana.org>
Thread-Topic: [IANA #812089] Protocol Action: 'Covering Prefixes Outbound Route Filter for BGP-4' to Proposed Standard (draft-ietf-bess-orf-covering-prefixes-06.txt)
Thread-Index: AQHQXEsyIT9A0cX/qE+e2amxi6NIS50YDtXQ
Date: Thu, 12 Mar 2015 01:31:52 +0000
Message-ID: <CO1PR05MB442F5ABD0D80EE4AD1DCC10AE060@CO1PR05MB442.namprd05.prod.outlook.com>
References: <RT-Ticket-812089@icann.org> <20150309191547.20070.81805.idtracker@ietfa.amsl.com> <rt-4.2.9-32314-1426113111-1024.812089-7-0@icann.org>
In-Reply-To: <rt-4.2.9-32314-1426113111-1024.812089-7-0@icann.org>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [66.129.241.13]
authentication-results: iana.org; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:; BCL:0; PCL:0; RULEID:; SRVR:CO1PR05MB444; UriScan:; BCL:0; PCL:0; RULEID:; SRVR:CO1PR05MB539; 
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(13464003)(51704005)(164054003)(377454003)(122556002)(76576001)(40100003)(86362001)(15975445007)(50986999)(74316001)(2656002)(230783001)(76176999)(110136001)(2351001)(54356999)(102836002)(46102003)(77156002)(33656002)(62966003)(19580395003)(92566002)(66066001)(1720100001)(106116001)(2900100001)(19580405001)(99286002)(87936001)(2501003)(2950100001); DIR:OUT; SFP:1102; SCL:1; SRVR:CO1PR05MB444; H:CO1PR05MB442.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <CO1PR05MB4447D82BB0F3C9851D9A8CAAE060@CO1PR05MB444.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002009)(5005006); SRVR:CO1PR05MB444; BCL:0; PCL:0; RULEID:; SRVR:CO1PR05MB444; 
x-forefront-prvs: 05134F8B4F
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-originalarrivaltime: 12 Mar 2015 01:31:52.2274 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: CO1PR05MB444
X-OriginatorOrg: juniper.net
X-Helo-Check-Failed: Verification failed for HELO na01-bn1-obe.outbound.protection.outlook.com
X-SA-Exim-Connect-IP: 157.56.110.142
X-SA-Exim-Rcpt-To: draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org
X-SA-Exim-Mail-From: rbonica@juniper.net
X-SA-Exim-Version: 4.2.1 (built Mon, 26 Dec 2011 16:24:06 +0000)
X-SA-Exim-Scanned: Yes (on zinfandel.tools.ietf.org)
Resent-To: draft-ietf-bess-orf-covering-prefixes.all@ietf.org
Resent-Message-Id: <20150312084051.93FE51A870E@ietfa.amsl.com>
Resent-Date: Thu, 12 Mar 2015 01:40:48 -0700 (PDT)
Resent-From: rbonica@juniper.net
Archived-At: <http://mailarchive.ietf.org/arch/msg/draft-ietf-bess-orf-covering-prefixes.all@tools/Vdgd3gY7ga30mNTDJTRNVo05Jp0>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/1H2ES-VujWrz97kFOzpUXkESLfw>
X-Mailman-Approved-At: Thu, 12 Mar 2015 07:55:09 -0700
Cc: "draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org" <draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org>
Subject: Re: [bess] [IANA #812089] Protocol Action: 'Covering Prefixes Outbound Route Filter for BGP-4' to Proposed Standard (draft-ietf-bess-orf-covering-prefixes-06.txt)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Mar 2015 08:40:53 -0000
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From nobody Thu Mar 12 13:06:10 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 47C631A86F5; Thu, 12 Mar 2015 13:06:08 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ULlL0uot7QYl; Thu, 12 Mar 2015 13:06:04 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 33BFD1A702F; Thu, 12 Mar 2015 13:06:04 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150312200604.25660.45226.idtracker@ietfa.amsl.com>
Date: Thu, 12 Mar 2015 13:06:04 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/RjzGEOXVxIRq1ON_YjYGH9BdSuk>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Mar 2015 20:06:08 -0000

IANA action state changed to RFC-Ed-Ack
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Thu Mar 12 13:06:11 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 6AAB71A702F; Thu, 12 Mar 2015 13:06:08 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 47C631A86F5; Thu, 12 Mar 2015 13:06:08 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, TVD_SPACE_RATIO=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ULlL0uot7QYl; Thu, 12 Mar 2015 13:06:04 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 33BFD1A702F; Thu, 12 Mar 2015 13:06:04 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>, <bess-chairs@ietf.org>, <martin.vigoureux@alcatel-lucent.com>, <bess@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150312200604.25660.45226.idtracker@ietfa.amsl.com>
Date: Thu, 12 Mar 2015 13:06:04 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/RjzGEOXVxIRq1ON_YjYGH9BdSuk>
Subject: [bess] ID Tracker State Update Notice: <draft-ietf-bess-orf-covering-prefixes-06.txt>
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Mar 2015 20:06:08 -0000

IANA action state changed to RFC-Ed-Ack
ID Tracker URL: http://datatracker.ietf.org/doc/draft-ietf-bess-orf-covering-prefixes/


From nobody Sun Mar 15 08:11:22 2015
Return-Path: <erosen@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 373AB1A1B24; Sun, 15 Mar 2015 08:11:21 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.902
X-Spam-Level: 
X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yiGERB2OoTrE; Sun, 15 Mar 2015 08:11:19 -0700 (PDT)
Received: from na01-by2-obe.outbound.protection.outlook.com (mail-by2on0146.outbound.protection.outlook.com [207.46.100.146]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A93381A1B22; Sun, 15 Mar 2015 08:11:19 -0700 (PDT)
Received: from [172.29.32.30] (66.129.241.13) by BN3PR0501MB1091.namprd05.prod.outlook.com (25.160.113.13) with Microsoft SMTP Server (TLS) id 15.1.112.19; Sun, 15 Mar 2015 15:11:16 +0000
Message-ID: <5505A10F.7000704@juniper.net>
Date: Sun, 15 Mar 2015 11:11:11 -0400
From: Eric C Rosen <erosen@juniper.net>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: Ben Niven-Jenkins <ben@niven-jenkins.co.uk>, <rtg-ads@tools.ietf.org>
References: <580A03D0-3A2A-4D39-80C8-030D8C9C37DB@niven-jenkins.co.uk>
In-Reply-To: <580A03D0-3A2A-4D39-80C8-030D8C9C37DB@niven-jenkins.co.uk>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 7bit
X-Originating-IP: [66.129.241.13]
X-ClientProxiedBy: DM2PR00CA0014.namprd00.prod.outlook.com (25.160.243.24) To BN3PR0501MB1091.namprd05.prod.outlook.com (25.160.113.13)
Authentication-Results: ietf.org; dkim=none (message not signed) header.d=none;
X-Microsoft-Antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BN3PR0501MB1091;
X-Forefront-Antispam-Report: BMV:1; SFV:NSPM; SFS:(10019020)(6049001)(6009001)(87266999)(65816999)(50466002)(86362001)(47776003)(558084003)(2950100001)(66066001)(23676002)(87976001)(122386002)(40100003)(92566002)(36756003)(59896002)(46102003)(62966003)(33656002)(77096005)(77156002)(230783001)(83506001)(54356999)(76176999)(50986999)(42186005); DIR:OUT; SFP:1102; SCL:1; SRVR:BN3PR0501MB1091; H:[172.29.32.30]; FPR:; SPF:None; MLV:sfv; LANG:en; 
X-Microsoft-Antispam-PRVS: <BN3PR0501MB109162E233704064641403BED4050@BN3PR0501MB1091.namprd05.prod.outlook.com>
X-Exchange-Antispam-Report-Test: UriScan:;
X-Exchange-Antispam-Report-CFA-Test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BN3PR0501MB1091; BCL:0; PCL:0; RULEID:; SRVR:BN3PR0501MB1091; 
X-Forefront-PRVS: 05168A3970
X-OriginatorOrg: juniper.net
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 15 Mar 2015 15:11:16.9435 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN3PR0501MB1091
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/uj4Ua-Bux1vXrvDMe0j5R7HWHn0>
Cc: draft-ietf-bess-mvpn-bidir.all@tools.ietf.org, rtg-dir@ietf.org, bess@ietf.org, erosen@juniper.net
Subject: Re: [bess] RtgDir review: draft-ietf-bess-mvpn-bidir-03.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Mar 2015 15:11:21 -0000

Ben,

Thanks for your review and comments.

I have incorporated your corrections into the XML, and they will be 
included in the -04 revision.

Eric



From nobody Mon Mar 16 03:26:22 2015
Return-Path: <rbonica@juniper.net>
X-Original-To: expand-draft-ietf-bess-orf-covering-prefixes.all@virtual.ietf.org
Delivered-To: bess@ietfa.amsl.com
Received: by ietfa.amsl.com (Postfix, from userid 65534) id 52D691A1B0C; Thu, 12 Mar 2015 11:37:24 -0700 (PDT)
X-Original-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Delivered-To: xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 345ED1A0074 for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Thu, 12 Mar 2015 11:37:24 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.235
X-Spam-Level: 
X-Spam-Status: No, score=-101.235 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_SOFTFAIL=0.665, USER_IN_WHITELIST=-100] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xoK_jXUrpHBv for <xfilter-draft-ietf-bess-orf-covering-prefixes.all@ietfa.amsl.com>; Thu, 12 Mar 2015 11:37:22 -0700 (PDT)
Received: from zinfandel.tools.ietf.org (zinfandel.tools.ietf.org [IPv6:2001:1890:123a::1:2a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BAD1E1A0211 for <draft-ietf-bess-orf-covering-prefixes.all@ietf.org>; Thu, 12 Mar 2015 11:37:22 -0700 (PDT)
Received: from mail-by2on0114.outbound.protection.outlook.com ([207.46.100.114]:13280 helo=na01-by2-obe.outbound.protection.outlook.com) by zinfandel.tools.ietf.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA384:256) (Exim 4.82_1-5b7a7c0-XX) (envelope-from <rbonica@juniper.net>) id 1YW7yz-0008B4-R0 for draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org; Thu, 12 Mar 2015 11:37:22 -0700
Received: from CO1PR05MB441.namprd05.prod.outlook.com (10.141.73.147) by CO1PR05MB489.namprd05.prod.outlook.com (10.141.71.144) with Microsoft SMTP Server (TLS) id 15.1.99.14; Thu, 12 Mar 2015 15:06:48 +0000
Received: from CO1PR05MB442.namprd05.prod.outlook.com (10.141.73.146) by CO1PR05MB441.namprd05.prod.outlook.com (10.141.73.147) with Microsoft SMTP Server (TLS) id 15.1.106.15; Thu, 12 Mar 2015 15:06:46 +0000
Received: from CO1PR05MB442.namprd05.prod.outlook.com ([169.254.13.61]) by CO1PR05MB442.namprd05.prod.outlook.com ([169.254.13.61]) with mapi id 15.01.0106.007; Thu, 12 Mar 2015 15:06:46 +0000
From: Ronald Bonica <rbonica@juniper.net>
To: "drafts-approval@iana.org" <drafts-approval@iana.org>
Thread-Topic: [IANA #812089] Protocol Action: 'Covering Prefixes Outbound Route Filter for BGP-4' to Proposed Standard (draft-ietf-bess-orf-covering-prefixes-06.txt)
Thread-Index: AQHQXEsyIT9A0cX/qE+e2amxi6NIS50YDtXQgABc7SuAAIftsA==
Date: Thu, 12 Mar 2015 15:06:45 +0000
Message-ID: <CO1PR05MB442C5456A1FCE599990C693AE060@CO1PR05MB442.namprd05.prod.outlook.com>
References: <RT-Ticket-812089@icann.org> <20150309191547.20070.81805.idtracker@ietfa.amsl.com> <rt-4.2.9-32314-1426113111-1024.812089-7-0@icann.org> <CO1PR05MB442F5ABD0D80EE4AD1DCC10AE060@CO1PR05MB442.namprd05.prod.outlook.com> <rt-4.2.9-22915-1426123938-999.812089-7-0@icann.org> <rt-4.2.9-30691-1426143595-1427.812089-7-0@icann.org>
In-Reply-To: <rt-4.2.9-30691-1426143595-1427.812089-7-0@icann.org>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [66.129.241.12]
authentication-results: iana.org; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:; BCL:0; PCL:0; RULEID:; SRVR:CO1PR05MB441; UriScan:; BCL:0; PCL:0; RULEID:; SRVR:CO1PR05MB489; 
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(377454003)(55674003)(13464003)(24454002)(164054003)(51704005)(86362001)(2501003)(46102003)(74316001)(77156002)(15975445007)(102836002)(92566002)(2900100001)(62966003)(2950100001)(93886004)(1720100001)(40100003)(122556002)(2351001)(66066001)(76576001)(110136001)(50986999)(54356999)(76176999)(87936001)(19580405001)(99286002)(230783001)(33656002)(106116001)(19580395003)(2656002); DIR:OUT; SFP:1102; SCL:1; SRVR:CO1PR05MB441; H:CO1PR05MB442.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <CO1PR05MB441731782E9BA35EB4EF369AE060@CO1PR05MB441.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002009)(5005006); SRVR:CO1PR05MB441; BCL:0; PCL:0; RULEID:; SRVR:CO1PR05MB441; 
x-forefront-prvs: 05134F8B4F
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-originalarrivaltime: 12 Mar 2015 15:06:45.9952 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: CO1PR05MB441
X-OriginatorOrg: juniper.net
X-Helo-Check-Failed: Verification failed for HELO na01-by2-obe.outbound.protection.outlook.com
X-SA-Exim-Connect-IP: 207.46.100.114
X-SA-Exim-Rcpt-To: draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org
X-SA-Exim-Mail-From: rbonica@juniper.net
X-SA-Exim-Version: 4.2.1 (built Mon, 26 Dec 2011 16:24:06 +0000)
X-SA-Exim-Scanned: Yes (on zinfandel.tools.ietf.org)
Resent-To: draft-ietf-bess-orf-covering-prefixes.all@ietf.org
Resent-Message-Id: <20150312183722.BAD1E1A0211@ietfa.amsl.com>
Resent-Date: Thu, 12 Mar 2015 11:37:22 -0700 (PDT)
Resent-From: rbonica@juniper.net
Archived-At: <http://mailarchive.ietf.org/arch/msg/draft-ietf-bess-orf-covering-prefixes.all@tools/5O_JWwDH8FnaXtiLAINSfooYQts>
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/rskNjkkxf-rwCu3mYb2KyP8Kndc>
X-Mailman-Approved-At: Mon, 16 Mar 2015 03:26:20 -0700
Cc: "draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org" <draft-ietf-bess-orf-covering-prefixes.all@tools.ietf.org>
Subject: Re: [bess] [IANA #812089] Protocol Action: 'Covering Prefixes Outbound Route Filter for BGP-4' to Proposed Standard (draft-ietf-bess-orf-covering-prefixes-06.txt)
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 12 Mar 2015 18:37:24 -0000
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From nobody Mon Mar 16 07:39:21 2015
Return-Path: <ben@niven-jenkins.co.uk>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5FDCE1A87C8; Mon, 16 Mar 2015 07:38:40 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2jkJCZ1OAf-k; Mon, 16 Mar 2015 07:38:38 -0700 (PDT)
Received: from mailex.mailcore.me (mailex.mailcore.me [94.136.40.61]) by ietfa.amsl.com (Postfix) with ESMTP id 907281A87B2; Mon, 16 Mar 2015 07:38:38 -0700 (PDT)
Received: from aputeaux-554-1-125-59.w92-154.abo.wanadoo.fr ([92.154.84.59] helo=[192.168.1.141]) by smtp04.mailcore.me with esmtpa (Exim 4.80.1) (envelope-from <ben@niven-jenkins.co.uk>) id 1YXWA8-0001Ue-VU; Mon, 16 Mar 2015 14:38:37 +0000
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: Ben Niven-Jenkins <ben@niven-jenkins.co.uk>
In-Reply-To: <5505A10F.7000704@juniper.net>
Date: Mon, 16 Mar 2015 14:38:33 +0000
Content-Transfer-Encoding: quoted-printable
Message-Id: <8A90F6AB-2499-4AC8-A3F2-3F69531559FC@niven-jenkins.co.uk>
References: <580A03D0-3A2A-4D39-80C8-030D8C9C37DB@niven-jenkins.co.uk> <5505A10F.7000704@juniper.net>
To: Eric C Rosen <erosen@juniper.net>
X-Mailer: Apple Mail (2.2070.6)
X-Mailcore-Auth: 9600544
X-Mailcore-Domain: 172912
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/9ZQs8yr7-GS4Nb_Eb-cHCBiRpIA>
X-Mailman-Approved-At: Mon, 16 Mar 2015 07:39:20 -0700
Cc: draft-ietf-bess-mvpn-bidir.all@tools.ietf.org, rtg-dir@ietf.org, bess@ietf.org, rtg-ads@tools.ietf.org
Subject: Re: [bess] RtgDir review: draft-ietf-bess-mvpn-bidir-03.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 16 Mar 2015 14:38:40 -0000

Thanks Eric. Ben

> On 15 Mar 2015, at 15:11, Eric C Rosen <erosen@juniper.net> wrote:
>=20
> Ben,
>=20
> Thanks for your review and comments.
>=20
> I have incorporated your corrections into the XML, and they will be =
included in the -04 revision.
>=20
> Eric
>=20
>=20


From nobody Tue Mar 17 07:42:12 2015
Return-Path: <tmmorin.orange@gmail.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 527DC1A1ADF for <bess@ietfa.amsl.com>; Tue, 17 Mar 2015 07:42:11 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id lGs_FjgBIttq for <bess@ietfa.amsl.com>; Tue, 17 Mar 2015 07:42:05 -0700 (PDT)
Received: from mail-wi0-x232.google.com (mail-wi0-x232.google.com [IPv6:2a00:1450:400c:c05::232]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id F1E061A1A51 for <bess@ietf.org>; Tue, 17 Mar 2015 07:42:04 -0700 (PDT)
Received: by wifj2 with SMTP id j2so12961819wif.1 for <bess@ietf.org>; Tue, 17 Mar 2015 07:42:03 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=sender:message-id:date:from:organization:user-agent:mime-version:to :subject:content-type:content-transfer-encoding; bh=vbhS40HFRYd65OtkAlQcn+7ZW/XquUarwhs0BzdCwqA=; b=FGDri6YR6vWVEvtaUC8iTMVwO00RINefEF0t9E6+bNNjvEt/OIt5QPxdiFE/svSl1f DmtzFR5YWfUNaZopvv7Yspm9VcUi4efKiruYLQA1EUtFE0Xs2muP0iZZ1AE5KOHsFoJF ZlqzJm8MZXBSeOk7woBPt3+aFzRutIdinJ4ZX+oLuASfvkqRNolxQ2+9LfXmM36qJ3wK Hl9UintUuJXkx5K7jBeb3+7FcpAicVgKQlkaLBnN8W54SxY0Fc4Pi6iOePFY84KTZ7n1 CHWuv6MkpdCEnhvTAGMw4cmVQqfGc5EgMV3GAyycX1O/Dw2HrbbUKrnJ2MKYmKongR9A /XxA==
X-Received: by 10.194.156.133 with SMTP id we5mr136012120wjb.37.1426603323682;  Tue, 17 Mar 2015 07:42:03 -0700 (PDT)
Received: from [127.0.0.1] (ARennes-652-1-121-114.w92-139.abo.wanadoo.fr. [92.139.24.114]) by mx.google.com with ESMTPSA id d9sm2956726wib.20.2015.03.17.07.42.02 for <bess@ietf.org> (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 17 Mar 2015 07:42:03 -0700 (PDT)
Sender: Thomas Morin <tmmorin.orange@gmail.com>
Message-ID: <55083D39.9000207@orange.com>
Date: Tue, 17 Mar 2015 15:42:01 +0100
From: Thomas Morin <thomas.morin@orange.com>
Organization: Orange
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: "bess@ietf.org" <bess@ietf.org>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/Zduo9BjYFqqAkQcEiO5Mlco70Qk>
Subject: [bess] Agenda for Dallas
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 17 Mar 2015 14:42:11 -0000

Hi everyone,

Here is our final agenda for our meeting next week in Dallas (Wednesday, 
March 25, 9:00-11:30).

Note well that all slots include the time for discussing the drafts.

Monday 8pm (Dallas local time) is the _deadline_ for presenters to send 
us their slides, keeping in mind that sending them _before is better_.

Thanks and see you in Dallas,

-Thomas/Martin

http://www.ietf.org/proceedings/92/agenda/agenda-92-bess
htmlized: http://tools.ietf.org/wg/bess/agenda?item=agenda-92-bess.html

WG Status, chairs
10 min

draft-xu-bess-l3vpn-prefix-orf
draft-xu-bess-virtual-subnet-rib-reduction
Xiaohu X., 20 min
(30 min)

draft-morin-bess-mvpn-fast-failover
Robert K., 5 min
(35 min)

draft-dolganow-bess-mvpn-expl-track
Eric R., 15 min
(50 min)

draft-kebler-bess-sa-pref
Robert K., 5 min
(55 min)

draft-ietf-bess-pta-flags
Eric R., 20 min
(75min)

draft-snr-bess-evpn-na-flags
draft-snr-bess-evpn-proxy-arp-nd
Jorge R., 10 min
(85 min)

draft-mohanty-bess-evpn-df-election
Satya M., 15 min
(100 min)

draft-hao-bess-evpn-df-handshaking
Weiguo H., 10 min
(110 min)

draft-tsingh-bess-pbb-evpn-yang-cfg
Kishore T., 15 min
(125 min)

draft-boutros-bess-evpn-vpws-service-edge-gateway
Patrice B., 10 min
(135 min)

draft-abhattacharya-bess-l2vpn-ipv6-remotepe
Mohamed B., 5 min
(140 min)

draft-hao-bess-inter-nvo3-vpn
Lucy Y., 10 min
(150 min)

--------------------------------------------------------------------

--------------------------------------------------------------------



From nobody Mon Mar 23 06:26:45 2015
Return-Path: <internet-drafts@ietf.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 09DDC1A8A79; Mon, 23 Mar 2015 06:26:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oYrSepo0wkG0; Mon, 23 Mar 2015 06:26:41 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 5E4941A8A60; Mon, 23 Mar 2015 06:26:40 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <i-d-announce@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.3
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150323132640.12873.14050.idtracker@ietfa.amsl.com>
Date: Mon, 23 Mar 2015 06:26:40 -0700
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/k2qtFaMdh8qAq9DAswn7wiYGs80>
Cc: bess@ietf.org
Subject: [bess] I-D Action: draft-ietf-bess-mvpn-bidir-04.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 23 Mar 2015 13:26:43 -0000

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

        Title           : MVPN: Using Bidirectional P-Tunnels
        Authors         : Eric C. Rosen
                          IJsbrand Wijnands
                          Yiqun Cai
                          Arjen Boers
	Filename        : draft-ietf-bess-mvpn-bidir-04.txt
	Pages           : 33
	Date            : 2015-03-23

Abstract:
   A set of prior RFCs specify procedures for supporting multicast in
   BGP/MPLS IP VPNs.  These procedures allow customer multicast data to
   travel across a service provider's backbone network through a set of
   multicast tunnels.  The tunnels are advertised in certain BGP
   multicast "auto-discovery" routes, by means of a BGP attribute known
   as the "Provider Multicast Service Interface (PMSI) Tunnel
   attribute".  Encodings have been defined that allow the PMSI Tunnel
   attribute to identify bidirectional (multipoint-to-multipoint)
   multicast distribution trees.  However, the prior RFCs do not provide
   all the necessary procedures for using bidirectional tunnels to
   support multicast VPNs.  This document updates RFCs 6513, 6514 and
   6625 by specifying those procedures.  In particular, it specifies the
   procedures for assigning customer multicast flows (unidirectional or
   bidirectional) to specific bidirectional tunnels in the provider
   backbone, for advertising such assignments, and for determining which
   flows have been assigned to which tunnels.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-bess-mvpn-bidir/

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-bess-mvpn-bidir-04

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=draft-ietf-bess-mvpn-bidir-04


Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

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


From nobody Wed Mar 25 08:23:21 2015
Return-Path: <antoni.przygienda@ericsson.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7EA9F1A8760 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 08:23:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.2
X-Spam-Level: 
X-Spam-Status: No, score=-4.2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id w1caiGicFPDW for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 08:23:18 -0700 (PDT)
Received: from usevmg21.ericsson.net (usevmg21.ericsson.net [198.24.6.65]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 944DF1A8761 for <bess@ietf.org>; Wed, 25 Mar 2015 08:23:09 -0700 (PDT)
X-AuditID: c6180641-f790b6d000004359-bd-551270eae064
Received: from EUSAAHC001.ericsson.se (Unknown_Domain [147.117.188.75]) by usevmg21.ericsson.net (Symantec Mail Security) with SMTP id DF.99.17241.AE072155; Wed, 25 Mar 2015 09:25:14 +0100 (CET)
Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC001.ericsson.se ([147.117.188.75]) with mapi id 14.03.0210.002; Wed, 25 Mar 2015 11:23:07 -0400
From: Antoni Przygienda <antoni.przygienda@ericsson.com>
To: "Rabadan, Jorge (Jorge) (jorge.rabadan@alcatel-lucent.com)" <jorge.rabadan@alcatel-lucent.com>, "Henderickx, Wim (Wim) (wim.henderickx@alcatel-lucent.com)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILg==
Date: Wed, 25 Mar 2015 15:23:08 +0000
Message-ID: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [147.117.188.12]
Content-Type: multipart/alternative; boundary="_000_2E4BB27CAB87BF43B4207C0E55860F18287F74eusaamb103ericsso_"
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFnrGLMWRmVeSWpSXmKPExsUyuXSPt+6rAqFQgyufrC1WHJ/JbPHt0hIW i9mPf7A5MHu0PtvL6rFkyU+mAKYoLpuU1JzMstQifbsErozP106zF0x3qZi8eQtzA2OfTRcj J4eEgInEqocb2SFsMYkL99azdTFycQgJHGGUOPP+LZSznFHi0JozrCBVbAIWEpe/PWUGSYgI HGeU+L1tFRNIgllAWeLw/PNgo4QFRCW+bXgIVMQBVCQlMel1AEhYREBP4uGjRWBzWARUJTqW fgGzeQW8JfZP+cEIYjMCXfH91BqokeISt57MZ4K4TkBiyZ7zzBC2qMTLx/9YIWwliTmvrzFD 1OdLHLy4lQlipqDEyZlPWCYwCs9CMmoWkrJZSMog4joSC3Z/YoOwtSWWLXzNDGOfOfCYCVl8 ASP7KkaO0uLUstx0I8NNjMCIOSbB5riDccEny0OMAhyMSjy8BlsEQ4VYE8uKK3MPMUpzsCiJ 85ZdORgiJJCeWJKanZpakFoUX1Sak1p8iJGJg1OqgVFWZsMiN881PFt1Vm+z85W6d+Lioo8L 8pccF2Nvny1wOM02REBu4wPney+1a7Zy9xw1Pt0f/SJWdoOAx7LQpxqB7XaOt368lmoSXxdb 5sjTuLY5vbRfROrLXav3tszTP5uLhTzaUjlDJ8L8JvenedWb5vEmiPiWTo8tD5K0lPQrCrtg ovnOQYmlOCPRUIu5qDgRAFSTWS55AgAA
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/VCdc7qqWRy__wvYtxtOhDd7bniQ>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 15:23:20 -0000

--_000_2E4BB27CAB87BF43B4207C0E55860F18287F74eusaamb103ericsso_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Watching the presentation in the group and instead crowding the mike here a=
re my comments:


a)      It is worth explaining what flavor of ARP proxy eVPN implements. 'p=
roxy ARP' I found out has different flavors including e.g. the router respo=
nding with its own MAC to requests representing remote hosts. Customers & o=
ther folks are easily confused by the overloaded 'proxy ARP' term in eVPN c=
ontext.

b)      It is worth explaining what is suggested behavior if eVPN advertise=
s the same IP with multiple MACs and what happens when e.g. the served MAC =
vanishes

c)      It is worth explaining what the suggested behavior should be when '=
local-bit' MACs are being advertised from remotes (and when those collide)

d)      eVPN draft does not explain anything about possibility to snoop @ D=
HCP

e)      the IP duplicate detection is an interesting thing given the anycas=
t/multicast and other zoo

--- tony

There are basically two types of people. People who accomplish things, and =
people who claim to have accomplished things. The first group is less crowd=
ed.<http://www.brainyquote.com/quotes/quotes/m/marktwain393535.html>
~~~ Mark Twain<http://www.brainyquote.com/quotes/quotes/m/marktwain393535.h=
tml>


--_000_2E4BB27CAB87BF43B4207C0E55860F18287F74eusaamb103ericsso_
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=3D"Generator" 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:"Bookman Old Style";
	panose-1:2 5 6 4 5 5 5 2 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:186607402;
	mso-list-type:hybrid;
	mso-list-template-ids:355776520 67698711 67698713 67698715 67698703 676987=
13 67698715 67698703 67698713 67698715;}
@list l0:level1
	{mso-level-number-format:alpha-lower;
	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=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal">Watching the presentation in the group and instead c=
rowding the mike here are my comments:<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoListParagraph" style=3D"text-indent:-.25in;mso-list:l0 level=
1 lfo1"><![if !supportLists]><span style=3D"mso-list:Ignore">a)<span style=
=3D"font:7.0pt &quot;Times New Roman&quot;">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span><![endif]>It is worth explaining what flavor of ARP proxy eVP=
N implements. &#8216;proxy ARP&#8217; I found out has different flavors inc=
luding e.g. the router responding with its own MAC to requests representing=
 remote hosts. Customers &amp; other folks are easily
 confused by the overloaded &#8216;proxy ARP&#8217; term in eVPN context. <=
o:p></o:p></p>
<p class=3D"MsoListParagraph" style=3D"text-indent:-.25in;mso-list:l0 level=
1 lfo1"><![if !supportLists]><span style=3D"mso-list:Ignore">b)<span style=
=3D"font:7.0pt &quot;Times New Roman&quot;">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span><![endif]>It is worth explaining what is suggested behavior i=
f eVPN advertises the same IP with multiple MACs and what happens when e.g.=
 the served MAC vanishes<o:p></o:p></p>
<p class=3D"MsoListParagraph" style=3D"text-indent:-.25in;mso-list:l0 level=
1 lfo1"><![if !supportLists]><span style=3D"mso-list:Ignore">c)<span style=
=3D"font:7.0pt &quot;Times New Roman&quot;">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span><![endif]>It is worth explaining what the suggested behavior =
should be when &#8216;local-bit&#8217; MACs are being advertised from remot=
es (and when those collide)
<o:p></o:p></p>
<p class=3D"MsoListParagraph" style=3D"text-indent:-.25in;mso-list:l0 level=
1 lfo1"><![if !supportLists]><span style=3D"mso-list:Ignore">d)<span style=
=3D"font:7.0pt &quot;Times New Roman&quot;">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span><![endif]>eVPN draft does not explain anything about possibil=
ity to snoop @ DHCP
<o:p></o:p></p>
<p class=3D"MsoListParagraph" style=3D"text-indent:-.25in;mso-list:l0 level=
1 lfo1"><![if !supportLists]><span style=3D"mso-list:Ignore">e)<span style=
=3D"font:7.0pt &quot;Times New Roman&quot;">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span><![endif]>the IP duplicate detection is an interesting thing =
given the anycast/multicast and other zoo
<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">--- tony <o:p></o:p></p>
<div style=3D"mso-element:para-border-div;border:none;border-bottom:solid w=
indowtext 1.0pt;padding:0in 0in 1.0pt 0in">
<p class=3D"MsoNormal" style=3D"border:none;padding:0in"><i><span style=3D"=
font-size:10.0pt;font-family:&quot;Bookman Old Style&quot;,&quot;serif&quot=
;;color:#A6A6A6"><o:p>&nbsp;</o:p></span></i></p>
</div>
<p class=3D"MsoNormal"><i><span style=3D"font-size:9.0pt;font-family:&quot;=
Bookman Old Style&quot;,&quot;serif&quot;;color:#A6A6A6"><a href=3D"http://=
www.brainyquote.com/quotes/quotes/m/marktwain393535.html" title=3D"view quo=
te"><span style=3D"color:#A6A6A6;text-decoration:none">There are
 basically two types of people. People who accomplish things, and people wh=
o claim to have accomplished things. The first group is less crowded.</span=
></a><o:p></o:p></span></i></p>
<p class=3D"MsoNormal"><i><span style=3D"font-size:8.0pt;font-family:&quot;=
Bookman Old Style&quot;,&quot;serif&quot;;color:#A6A6A6">~~~
<a href=3D"http://www.brainyquote.com/quotes/quotes/m/marktwain393535.html"=
 title=3D"view quote">
<span style=3D"color:#A6A6A6;text-decoration:none">Mark Twain</span></a><o:=
p></o:p></span></i></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
</body>
</html>

--_000_2E4BB27CAB87BF43B4207C0E55860F18287F74eusaamb103ericsso_--


From nobody Wed Mar 25 09:00:30 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CFC9E1B2A1D for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:00:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.511
X-Spam-Level: 
X-Spam-Status: No, score=-14.511 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id cO-TZbLaKQt0 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:00:28 -0700 (PDT)
Received: from rcdn-iport-5.cisco.com (rcdn-iport-5.cisco.com [173.37.86.76]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8A01C1A88C6 for <bess@ietf.org>; Wed, 25 Mar 2015 09:00:27 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=515; q=dns/txt; s=iport; t=1427299227; x=1428508827; h=from:to:cc:subject:date:message-id:in-reply-to: content-id:content-transfer-encoding:mime-version; bh=bREwIpuSr+UXMpH2OuV0L9UDEPX3s88wj5lRD3/XsL0=; b=TPWwRr8/Qs6PpPZOHJWUbfbhu3XqI2nGt9ILRb8zgfc2FmBkOxdNNoFG 6YN448AOPIRZUtPubplImZfZssstmPjzyyyWlf7kXPt5GDlCyOUEr1Vxa KnOstTlCtQczvlXmLT5MnjOtrihfHKMyMBCHRbzMTk1Ar/GSon7rUFzwe k=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0BxBQDe2hJV/5NdJa1cgwaBMMsCAoFXTAEBAQEBAX2EFQIEeRIBCHglAgQBDYg0yUUBAQEBAQEBAQEBAQEBAQEBAQEBARiLIYR2B4QtAQSKY4VtiW+BG49Kg0cig26CM38BAQE
X-IronPort-AV: E=Sophos;i="5.11,465,1422921600"; d="scan'208";a="406570227"
Received: from rcdn-core-11.cisco.com ([173.37.93.147]) by rcdn-iport-5.cisco.com with ESMTP; 25 Mar 2015 16:00:27 +0000
Received: from xhc-aln-x02.cisco.com (xhc-aln-x02.cisco.com [173.36.12.76]) by rcdn-core-11.cisco.com (8.14.5/8.14.5) with ESMTP id t2PG0QLl019717 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 25 Mar 2015 16:00:26 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.15]) by xhc-aln-x02.cisco.com ([173.36.12.76]) with mapi id 14.03.0195.001; Wed, 25 Mar 2015 11:00:26 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: Thomas Morin <thomas.morin@orange.com>, Martin Vigoureux <martin.vigoureux@alcatel-lucent.com>
Thread-Topic: [bess]   Poll for adoption: draft-sajassi-bess-evpn-etree
Thread-Index: AQHQZxTOTc7tyhhZ00ertccHJEN6bA==
Date: Wed, 25 Mar 2015 16:00:26 +0000
Message-ID: <D1383130.12DD3C%sajassi@cisco.com>
In-Reply-To: <54F85E80.2090304@orange.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.24.162.66]
Content-Type: text/plain; charset="iso-8859-1"
Content-ID: <50A88D162D8CA4429F14B174272FD9DF@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/sjt1Oe8huAoddTle-ecaurPQ5Xg>
Cc: "draft-sajassi-bess-evpn-etree@tools.ietf.org" <draft-sajassi-bess-evpn-etree@tools.ietf.org>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Poll for adoption: draft-sajassi-bess-evpn-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:00:29 -0000

Hi Thomas, Martin:

As I mentioned this morning at the BESS session, some people (including
myself and some other co-authors) missed this call and since there was no
reminder, we didn=B9t get a second chance to respond. As also mentioned
today, there is good level of interest in this draft as indicated by the
recent emails that I received and the meetings held to discuss some
aspects of this draft. Per your request, I will be next forwarding the
recent emails exchanges to the group.

Cheers,
Ali



From nobody Wed Mar 25 09:01:56 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 169E71A88C7 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:01:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IKm4fK6VPZ8X for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:01:53 -0700 (PDT)
Received: from alln-iport-3.cisco.com (alln-iport-3.cisco.com [173.37.142.90]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4B8901B2A9B for <bess@ietf.org>; Wed, 25 Mar 2015 09:01:10 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=7730; q=dns/txt; s=iport; t=1427299270; x=1428508870; h=from:to:subject:date:message-id:in-reply-to:mime-version; bh=m2LcJxvcI+HsQYa9hfrYvPQRYd8yqW6GFb05QzJ09Ew=; b=Ksxd5ObGcHTMP78Z9/Uz+g/JlShXmIJjyOU50ZBnqTrvN+gCNRF6blbM 5luH8Q+70w4nprfO9MDApDF7s3H/Zl02/XfsCBgsBm1dhVaoJ6VpPAsH2 TEAtQGUOT464hzYqc97WdGX5VBz0+0Fur0ofdsAP8oODjP9jUmhU1F+8S s=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0DZBADA2hJV/5pdJa1cgkNDUloEwzgGh0QCgVc4FAEBAQEBAQF8hBQBAgQtXgEIEQMBAig5FAcBAQUDAgQTG4gUn1mpbAEBAQEBAQQBAQEBAQEBARqLIYRlGIQtBZBQiW+ULCKDbm+BRH8BAQE
X-IronPort-AV: E=Sophos;i="5.11,465,1422921600";  d="scan'208,217";a="135276448"
Received: from rcdn-core-3.cisco.com ([173.37.93.154]) by alln-iport-3.cisco.com with ESMTP; 25 Mar 2015 16:01:09 +0000
Received: from xhc-aln-x02.cisco.com (xhc-aln-x02.cisco.com [173.36.12.76]) by rcdn-core-3.cisco.com (8.14.5/8.14.5) with ESMTP id t2PG19fK025714 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Wed, 25 Mar 2015 16:01:09 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.15]) by xhc-aln-x02.cisco.com ([173.36.12.76]) with mapi id 14.03.0195.001; Wed, 25 Mar 2015 11:01:09 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: draft-sajassi-bess-evpn-etree 
Thread-Index: AQHQZxTopI+hp9hmHkWO9ofc2Kvv9g==
Date: Wed, 25 Mar 2015 16:01:09 +0000
Message-ID: <D13845D3.12E002%sajassi@cisco.com>
In-Reply-To: <BN3PR0501MB110872D9E65D504C47B73B43A8060@BN3PR0501MB1108.namprd05.prod.outlook.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.24.162.66]
Content-Type: multipart/alternative; boundary="_000_D13845D312E002sajassiciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/aygA2QeHNY7wbDkN0SasRTSSZX4>
Subject: [bess] FW: draft-sajassi-bess-evpn-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:01:55 -0000

--_000_D13845D312E002sajassiciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable


FYI =96 per chairs request

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Wednesday, March 11, 2015 at 5:14 PM
To: "draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess=
-evpn-etree@tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<=
mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: draft-sajassi-bess-evpn-etree
Resent-From: <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Resent-To: <draft-sajassi-bess-evpn-etree@ietf.org<mailto:draft-sajassi-bes=
s-evpn-etree@ietf.org>>
Resent-Date: Thursday, March 12, 2015 at 4:50 AM

Hi ,

This is regarding the section 3.2 BUM Traffic .   From my reading of the dr=
aft it seems that the assumption here is that BUM traffic will  be send wit=
h the ESI label  even to the PEs which are not connected to the ESI.

An alternative solution could be to advertise two  IM routes per Ethernet T=
ag

1) One for  the AC in leaf mode
2) Second for the AC in Root mode

The advantage of the above solution is that we don=92t need the ESI label i=
n label stack for the packet send towards the set of PEs not connected to t=
he ESI .

The use of ESI label could be restricted  for forwarding the BUM traffic be=
tween MH PEs connected to ESI operating in all-active mode,  as per the SH =
procedures  described in the RFC 7432 .

Thanks
Nitin




--_000_D13845D312E002sajassiciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <D1A262678490124A9D2BFF24A1B7E37C@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div><br>
</div>
<div>FYI =96 per chairs request</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Nitin Singh &lt;<a href=3D"ma=
ilto:nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Wednesday, March 11, 2015 at =
5:14 PM<br>
<span style=3D"font-weight:bold">To: </span>&quot;<a href=3D"mailto:draft-s=
ajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.=
ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@tool=
s.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>draft-sajassi-bess-evpn-et=
ree <br>
<span style=3D"font-weight:bold">Resent-From: </span>&lt;<a href=3D"mailto:=
nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-To: </span>&lt;<a href=3D"mailto:dr=
aft-sajassi-bess-evpn-etree@ietf.org">draft-sajassi-bess-evpn-etree@ietf.or=
g</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-Date: </span>Thursday, March 12, 20=
15 at 4:50 AM<br>
</div>
<div><br>
</div>
<div xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micro=
soft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" x=
mlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:/=
/www.w3.org/TR/REC-html40">
<meta name=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@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:Consolas;
	panose-1:2 11 6 9 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:#0563C1;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:#954F72;
	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:Consolas;}
span.EmailStyle17
	{mso-style-type:personal-compose;
	font-family:"Calibri",sans-serif;
	color:windowtext;}
span.HTMLPreformattedChar
	{mso-style-name:"HTML Preformatted Char";
	mso-style-priority:99;
	mso-style-link:"HTML Preformatted";
	font-family:Consolas;}
.MsoChpDefault
	{mso-style-type:export-only;}
@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]-->
<div lang=3D"EN-US" link=3D"#0563C1" vlink=3D"#954F72">
<div class=3D"WordSection1">
<p class=3D"MsoNormal">Hi , <o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">This is regarding the section 3.2 BUM Traffic .&nbsp=
;&nbsp; From my reading of the draft it seems that the assumption here is t=
hat BUM traffic will&nbsp; be send with the ESI label&nbsp; even to the PEs=
 which are not connected to the ESI. &nbsp;<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">An alternative solution could be to advertise two &n=
bsp;IM routes per Ethernet Tag
<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">1) One for &nbsp;the AC in leaf mode <o:p></o:p></p>
<p class=3D"MsoNormal">2) Second for the AC in Root mode <o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">The advantage of the above solution is that we don=
=92t need the ESI label in label stack for the packet send towards the set =
of PEs not connected to the ESI .
<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">The use of ESI label could be restricted &nbsp;for f=
orwarding the BUM traffic between MH PEs connected to ESI operating in all-=
active mode, &nbsp;as per the SH procedures &nbsp;described in the RFC 7432=
 .
<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Thanks <o:p></o:p></p>
<p class=3D"MsoNormal">Nitin <o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
</div>
</div>
</span>
</body>
</html>

--_000_D13845D312E002sajassiciscocom_--


From nobody Wed Mar 25 09:02:54 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 83DED1B2ADC for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:02:52 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RCGsuyiFChqS for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:02:41 -0700 (PDT)
Received: from alln-iport-2.cisco.com (alln-iport-2.cisco.com [173.37.142.89]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 76CDF1B2A51 for <bess@ietf.org>; Wed, 25 Mar 2015 09:01:21 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=7889; q=dns/txt; s=iport; t=1427299281; x=1428508881; h=from:to:subject:date:message-id:in-reply-to:mime-version; bh=Guu0A9IvcpCbFCcZpL2VtODgIQ5XF1bfU1LRlX/aKzs=; b=Y+toVo+jj2tQjNgYl6LtAqzJjhDxX1dAr50vyV/3ixYH0Ohyk5g/7tfP C8al5I4AtPHJXYpaHe485qN2fxiNsUI/S4bIgVE+MRNx2cmzvXlAACBci ysZBPiDEwLHD1FIJHW2vuJmo/e7zzBiymmdBnoU/e3Y8lfp+aVWBio9t0 4=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0B1BQDe2hJV/5RdJa1cgkNDUloEywICgVdMAQEBAQEBfYQUAQIELV4BCBEDAQIoORQHAQEFAwIEE4gvn1mpbAEBAQEBAQQBAQEBAQEBARqLIYRlGIQtBZBQiW+BG4Mwj2Eig25vgUR/AQEB
X-IronPort-AV: E=Sophos;i="5.11,465,1422921600";  d="scan'208,217";a="135311524"
Received: from rcdn-core-12.cisco.com ([173.37.93.148]) by alln-iport-2.cisco.com with ESMTP; 25 Mar 2015 16:01:21 +0000
Received: from xhc-rcd-x06.cisco.com (xhc-rcd-x06.cisco.com [173.37.183.80]) by rcdn-core-12.cisco.com (8.14.5/8.14.5) with ESMTP id t2PG1K7o016412 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Wed, 25 Mar 2015 16:01:20 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.15]) by xhc-rcd-x06.cisco.com ([173.37.183.80]) with mapi id 14.03.0195.001; Wed, 25 Mar 2015 11:01:20 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: draft-sajassi-bess-evpn-etree 
Thread-Index: AQHQZxTupI+hp9hmHkWO9ofc2Kvv9g==
Date: Wed, 25 Mar 2015 16:01:19 +0000
Message-ID: <D13845F9.12E009%sajassi@cisco.com>
In-Reply-To: <BN3PR0501MB1108E8EDEC2CF621AD44886EA8060@BN3PR0501MB1108.namprd05.prod.outlook.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.24.162.66]
Content-Type: multipart/alternative; boundary="_000_D13845F912E009sajassiciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/GJmA0PexEybDKytTrJI1AzPlEUU>
Subject: [bess] FW: draft-sajassi-bess-evpn-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:02:52 -0000

--_000_D13845F912E009sajassiciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable



From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Wednesday, March 11, 2015 at 5:38 PM
To: "draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess=
-evpn-etree@tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<=
mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree
Resent-From: <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Resent-To: <draft-sajassi-bess-evpn-etree@ietf.org<mailto:draft-sajassi-bes=
s-evpn-etree@ietf.org>>
Resent-Date: Wednesday, March 11, 2015 at 8:11 PM

Hi ,

This is regarding the following text from the section 3.2 of the draft .

=93Furthermore, a PE advertises two special ESI MPLS labels: one for
Root and another for Leaf. These are used by remote PEs for traffic
originating from single-homed segments and for multi-homed segments
that are not connected to the advertising PE. Note that these special
labels are advertised on a per PE basis (i.e. each PE advertises only
two such special labels).=94

Can you please provide an explanation about the special ESI MPLS labels .  =
Which route type in EVPN will be used  to carry these labels ?

Thanks
Nitin

--_000_D13845F912E009sajassiciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <F42358DA113582499CBC69D199BD9314@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div><br>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Nitin Singh &lt;<a href=3D"ma=
ilto:nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Wednesday, March 11, 2015 at =
5:38 PM<br>
<span style=3D"font-weight:bold">To: </span>&quot;<a href=3D"mailto:draft-s=
ajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.=
ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@tool=
s.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>RE: draft-sajassi-bess-evp=
n-etree <br>
<span style=3D"font-weight:bold">Resent-From: </span>&lt;<a href=3D"mailto:=
nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-To: </span>&lt;<a href=3D"mailto:dr=
aft-sajassi-bess-evpn-etree@ietf.org">draft-sajassi-bess-evpn-etree@ietf.or=
g</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-Date: </span>Wednesday, March 11, 2=
015 at 8:11 PM<br>
</div>
<div><br>
</div>
<div xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micro=
soft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" x=
mlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:/=
/www.w3.org/TR/REC-html40">
<meta name=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@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:Consolas;
	panose-1:2 11 6 9 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:#0563C1;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:#954F72;
	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:Consolas;}
span.HTMLPreformattedChar
	{mso-style-name:"HTML Preformatted Char";
	mso-style-priority:99;
	mso-style-link:"HTML Preformatted";
	font-family:Consolas;}
span.EmailStyle19
	{mso-style-type:personal;
	font-family:"Calibri",sans-serif;
	color:windowtext;}
span.EmailStyle20
	{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]-->
<div lang=3D"EN-US" link=3D"#0563C1" vlink=3D"#954F72">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Hi , <o:p></o:p></span=
></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">This is regarding the =
following text from the section 3.2 of the draft .
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">=93Furthermore, a PE a=
dvertises two special ESI MPLS labels: one for<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Root and another for L=
eaf. These are used by remote PEs for traffic<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">originating from singl=
e-homed segments and for multi-homed segments<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">that are not connected=
 to the advertising PE. Note that these special<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">labels are advertised =
on a per PE basis (i.e. each PE advertises only<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">two such special label=
s).=94<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Can you please provide=
 an explanation about the special ESI MPLS labels . &nbsp;Which route type =
in EVPN will be used&nbsp; to carry these labels ?
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Thanks<o:p></o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Nitin <o:p></o:p></spa=
n></p>
</div>
</div>
</div>
</span>
</body>
</html>

--_000_D13845F912E009sajassiciscocom_--


From nobody Wed Mar 25 09:03:17 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 51FE01B2ABD for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:03:16 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id TDuJQD7Fvz0o for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:03:14 -0700 (PDT)
Received: from rcdn-iport-9.cisco.com (rcdn-iport-9.cisco.com [173.37.86.80]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E5D7D1A87A7 for <bess@ietf.org>; Wed, 25 Mar 2015 09:01:41 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=11241; q=dns/txt; s=iport; t=1427299303; x=1428508903; h=from:to:subject:date:message-id:in-reply-to:mime-version; bh=divsSRyMT8Tnfy9yJ++19QhX39nOk0AqH4s4E8YoPoQ=; b=jxVIZ0HZCOxU0Jfvj9BwsNez8s0VuGiStUziJA+ZA4+cBb++IQ9wQo+O ZteCbnYxFUXk8oW5gCtqhy4orY14M/WeIWz67hdcMHtDTTo2SGg1fnh3j LzXKewNaVoKyNMcKViBadOIUtXbkiQ7W5q8rYMGZ1Q3GpXyDbLg3rZHbg E=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0B1BQCn2hJV/5hdJa1cgkNDUloEywICgVdMAQEBAQEBfYQUAQIELV4BCBEDAQIoORQHAQEFAwEBBBMUiBufWalsAQEBAQEBBAEBAQEBAQEBGoshhGUYhC0FkFCJb4EbgzCPYSKDbm+BRH8BAQE
X-IronPort-AV: E=Sophos;i="5.11,465,1422921600";  d="scan'208,217";a="403465381"
Received: from rcdn-core-1.cisco.com ([173.37.93.152]) by rcdn-iport-9.cisco.com with ESMTP; 25 Mar 2015 16:01:42 +0000
Received: from xhc-rcd-x11.cisco.com (xhc-rcd-x11.cisco.com [173.37.183.85]) by rcdn-core-1.cisco.com (8.14.5/8.14.5) with ESMTP id t2PG1ebX005141 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Wed, 25 Mar 2015 16:01:41 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.15]) by xhc-rcd-x11.cisco.com ([173.37.183.85]) with mapi id 14.03.0195.001; Wed, 25 Mar 2015 11:01:40 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: draft-sajassi-bess-evpn-etree 
Thread-Index: AdBcUCnZNgMRxjitRJmKKmXMB3yB9gAC576gAB36U4ACkFIxAA==
Date: Wed, 25 Mar 2015 16:01:40 +0000
Message-ID: <D1384602.12E00C%sajassi@cisco.com>
In-Reply-To: <D12707EB.128D44%sajassi@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.24.162.66]
Content-Type: multipart/alternative; boundary="_000_D138460212E00Csajassiciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/4thoq6O6hSHMC4k5VveLHFIE8NY>
Subject: [bess] FW: draft-sajassi-bess-evpn-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:03:16 -0000

--_000_D138460212E00Csajassiciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable



From: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>
Date: Thursday, March 12, 2015 at 9:49 AM
To: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>, "draft-s=
ajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@=
tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft=
-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>, Cisco Employee <sajassi@cisco.co=
m<mailto:sajassi@cisco.com>>
Subject: Re: draft-sajassi-bess-evpn-etree


Hi Nitin,

This  =93ESI label=94 is advertised using =93ESI label extended community=
=94 along with the Eth-AD per ES route as always.

The purpose of this label is to have a root/leaf indication in the data-pla=
ne for traffic being sourced from single-home sites as discussed in that pa=
ragraph.

Cheers,
Ali

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Wednesday, March 11, 2015 at 5:38 PM
To: "draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess=
-evpn-etree@tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<=
mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree
Resent-From: <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Resent-To: <draft-sajassi-bess-evpn-etree@ietf.org<mailto:draft-sajassi-bes=
s-evpn-etree@ietf.org>>
Resent-Date: Wednesday, March 11, 2015 at 8:11 PM

Hi ,

This is regarding the following text from the section 3.2 of the draft .

=93Furthermore, a PE advertises two special ESI MPLS labels: one for
Root and another for Leaf. These are used by remote PEs for traffic
originating from single-homed segments and for multi-homed segments
that are not connected to the advertising PE. Note that these special
labels are advertised on a per PE basis (i.e. each PE advertises only
two such special labels).=94

Can you please provide an explanation about the special ESI MPLS labels .  =
Which route type in EVPN will be used  to carry these labels ?

Thanks
Nitin

--_000_D138460212E00Csajassiciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <7CEEF62AAFDD3F4A8EAD1F9B1F2D1242@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div><br>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Cisco Employee &lt;<a href=3D=
"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Thursday, March 12, 2015 at 9=
:49 AM<br>
<span style=3D"font-weight:bold">To: </span>Nitin Singh &lt;<a href=3D"mail=
to:nisingh@juniper.net">nisingh@juniper.net</a>&gt;, &quot;<a href=3D"mailt=
o:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etr=
ee@tools.ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-e=
tree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<b=
r>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;, Cisco Employee &lt;<a href=3D"mailto:sajassi@cisco.com">sajassi@ci=
sco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>Re: draft-sajassi-bess-evp=
n-etree <br>
</div>
<div><br>
</div>
<div>
<div style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line=
-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-famil=
y: Calibri, sans-serif;">
<div><br>
</div>
<div>Hi Nitin,</div>
<div><br>
</div>
<div>This &nbsp;=93ESI label=94 is advertised using =93ESI label extended c=
ommunity=94 along with the Eth-AD per ES route as always.</div>
<div><br>
</div>
<div>The purpose of this label is to have a root/leaf indication in the dat=
a-plane for traffic being sourced from single-home sites as discussed in th=
at paragraph.</div>
<div><br>
</div>
<div>Cheers,</div>
<div>Ali</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Nitin Singh &lt;<a href=3D"ma=
ilto:nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Wednesday, March 11, 2015 at =
5:38 PM<br>
<span style=3D"font-weight:bold">To: </span>&quot;<a href=3D"mailto:draft-s=
ajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.=
ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@tool=
s.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>RE: draft-sajassi-bess-evp=
n-etree <br>
<span style=3D"font-weight:bold">Resent-From: </span>&lt;<a href=3D"mailto:=
nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-To: </span>&lt;<a href=3D"mailto:dr=
aft-sajassi-bess-evpn-etree@ietf.org">draft-sajassi-bess-evpn-etree@ietf.or=
g</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-Date: </span>Wednesday, March 11, 2=
015 at 8:11 PM<br>
</div>
<div><br>
</div>
<div xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micro=
soft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" x=
mlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:/=
/www.w3.org/TR/REC-html40">
<meta name=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@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:Consolas;
	panose-1:2 11 6 9 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:#0563C1;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:#954F72;
	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:Consolas;}
span.HTMLPreformattedChar
	{mso-style-name:"HTML Preformatted Char";
	mso-style-priority:99;
	mso-style-link:"HTML Preformatted";
	font-family:Consolas;}
span.EmailStyle19
	{mso-style-type:personal;
	font-family:"Calibri",sans-serif;
	color:windowtext;}
span.EmailStyle20
	{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]-->
<div lang=3D"EN-US" link=3D"#0563C1" vlink=3D"#954F72">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Hi , <o:p></o:p></span=
></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">This is regarding the =
following text from the section 3.2 of the draft .
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">=93Furthermore, a PE a=
dvertises two special ESI MPLS labels: one for<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Root and another for L=
eaf. These are used by remote PEs for traffic<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">originating from singl=
e-homed segments and for multi-homed segments<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">that are not connected=
 to the advertising PE. Note that these special<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">labels are advertised =
on a per PE basis (i.e. each PE advertises only<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">two such special label=
s).=94<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Can you please provide=
 an explanation about the special ESI MPLS labels . &nbsp;Which route type =
in EVPN will be used&nbsp; to carry these labels ?
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Thanks<o:p></o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Nitin <o:p></o:p></spa=
n></p>
</div>
</div>
</div>
</span></div>
</div>
</span>
</body>
</html>

--_000_D138460212E00Csajassiciscocom_--


From nobody Wed Mar 25 09:03:28 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4C2371B2AEF for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:03:27 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hm7aZl-iDkFA for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:03:25 -0700 (PDT)
Received: from alln-iport-6.cisco.com (alln-iport-6.cisco.com [173.37.142.93]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3071B1A88D1 for <bess@ietf.org>; Wed, 25 Mar 2015 09:01:57 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=11174; q=dns/txt; s=iport; t=1427299317; x=1428508917; h=from:to:subject:date:message-id:in-reply-to:mime-version; bh=7glzesJKpip9ZI9qjOUzDMFJb1fH8/GeJdKi3jAJj8I=; b=fq+OIvQGA/nwRoML5y8GaqCLWSD7wX9laIJZ48f+NNw/tPZaH8fiQLcw WsmrKiDRs3PQ/9cAEz+Dxh2qZ/vRw2MVLSM+5MgOI1syEzkorlQK2RZyc 7Gk4SQUVTmddpHRjkf8IU3RFid+iITsrlM2Cz80QVX6Ckmh4QzC2f8HXu c=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0B1BQDe2hJV/5pdJa1cgkNDUloEywICgVdMAQEBAQEBfYQUAQIELV4BCBEDAQIoORQHAQEFAwEBBBMbiBSfWalsAQEBAQEBBAEBAQEBAQEBGoshhGUYhC0FkFCJb5QsIoNub4FEfwEBAQ
X-IronPort-AV: E=Sophos;i="5.11,465,1422921600";  d="scan'208,217";a="135310118"
Received: from rcdn-core-3.cisco.com ([173.37.93.154]) by alln-iport-6.cisco.com with ESMTP; 25 Mar 2015 16:01:56 +0000
Received: from xhc-aln-x09.cisco.com (xhc-aln-x09.cisco.com [173.36.12.83]) by rcdn-core-3.cisco.com (8.14.5/8.14.5) with ESMTP id t2PG1uM2026969 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Wed, 25 Mar 2015 16:01:56 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.15]) by xhc-aln-x09.cisco.com ([173.36.12.83]) with mapi id 14.03.0195.001; Wed, 25 Mar 2015 11:01:55 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: draft-sajassi-bess-evpn-etree 
Thread-Index: AdBcUCnZNgMRxjitRJmKKmXMB3yB9gAhSLwAAo/twoA=
Date: Wed, 25 Mar 2015 16:01:55 +0000
Message-ID: <D138461A.12E011%sajassi@cisco.com>
In-Reply-To: <D1271255.128DFA%sajassi@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.24.162.66]
Content-Type: multipart/alternative; boundary="_000_D138461A12E011sajassiciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/5TBoDkwRGKcHWtvYGTXFfPk0rok>
Subject: [bess] FW: draft-sajassi-bess-evpn-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:03:27 -0000

--_000_D138461A12E011sajassiciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable



From: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>
Date: Thursday, March 12, 2015 at 10:00 AM
To: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>, "draft-s=
ajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@=
tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft=
-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>, Cisco Employee <sajassi@cisco.co=
m<mailto:sajassi@cisco.com>>
Subject: Re: draft-sajassi-bess-evpn-etree


There are bunch of disadvantages for doing it this way:


  1.  That=92s a lot of advertisements if we do it with IM route - i.e., th=
e BGP messages for IM gets increased by 100% !! A given ES can support hund=
reds or thousands of IM routes !!
  2.  In data-plane we end up increasing the size of LFIB
  3.  There are now to different approaches for single-home versus multi-ho=
med sites

Cheers,
Ali

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Wednesday, March 11, 2015 at 5:14 PM
To: "draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess=
-evpn-etree@tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<=
mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: draft-sajassi-bess-evpn-etree
Resent-From: <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Resent-To: <draft-sajassi-bess-evpn-etree@ietf.org<mailto:draft-sajassi-bes=
s-evpn-etree@ietf.org>>
Resent-Date: Thursday, March 12, 2015 at 4:50 AM

Hi ,

This is regarding the section 3.2 BUM Traffic .   From my reading of the dr=
aft it seems that the assumption here is that BUM traffic will  be send wit=
h the ESI label  even to the PEs which are not connected to the ESI.

An alternative solution could be to advertise two  IM routes per Ethernet T=
ag

1) One for  the AC in leaf mode
2) Second for the AC in Root mode

The advantage of the above solution is that we don=92t need the ESI label i=
n label stack for the packet send towards the set of PEs not connected to t=
he ESI .

The use of ESI label could be restricted  for forwarding the BUM traffic be=
tween MH PEs connected to ESI operating in all-active mode,  as per the SH =
procedures  described in the RFC 7432 .

Thanks
Nitin




--_000_D138461A12E011sajassiciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <C9AC77C707E5194E903F16D9DE9C60A4@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div><br>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Cisco Employee &lt;<a href=3D=
"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Thursday, March 12, 2015 at 1=
0:00 AM<br>
<span style=3D"font-weight:bold">To: </span>Nitin Singh &lt;<a href=3D"mail=
to:nisingh@juniper.net">nisingh@juniper.net</a>&gt;, &quot;<a href=3D"mailt=
o:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etr=
ee@tools.ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-e=
tree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<b=
r>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;, Cisco Employee &lt;<a href=3D"mailto:sajassi@cisco.com">sajassi@ci=
sco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>Re: draft-sajassi-bess-evp=
n-etree <br>
</div>
<div><br>
</div>
<div>
<div style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line=
-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-famil=
y: Calibri, sans-serif;">
<div><br>
</div>
<div>There are bunch of disadvantages for doing it this way:</div>
<div><br>
</div>
<ol>
<li>That=92s a lot of advertisements if we do it with IM route - i.e., the =
BGP messages for IM gets increased by 100% !! A given ES can support hundre=
ds or thousands of IM routes !!</li><li>In data-plane we end up increasing =
the size of LFIB</li><li>There are now to different approaches for single-h=
ome versus multi-homed sites</li></ol>
<div><br>
</div>
<div>Cheers,</div>
<div>Ali</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Nitin Singh &lt;<a href=3D"ma=
ilto:nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Wednesday, March 11, 2015 at =
5:14 PM<br>
<span style=3D"font-weight:bold">To: </span>&quot;<a href=3D"mailto:draft-s=
ajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.=
ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@tool=
s.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>draft-sajassi-bess-evpn-et=
ree <br>
<span style=3D"font-weight:bold">Resent-From: </span>&lt;<a href=3D"mailto:=
nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-To: </span>&lt;<a href=3D"mailto:dr=
aft-sajassi-bess-evpn-etree@ietf.org">draft-sajassi-bess-evpn-etree@ietf.or=
g</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-Date: </span>Thursday, March 12, 20=
15 at 4:50 AM<br>
</div>
<div><br>
</div>
<div xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micro=
soft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" x=
mlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:/=
/www.w3.org/TR/REC-html40">
<meta name=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@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:Consolas;
	panose-1:2 11 6 9 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:#0563C1;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:#954F72;
	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:Consolas;}
span.EmailStyle17
	{mso-style-type:personal-compose;
	font-family:"Calibri",sans-serif;
	color:windowtext;}
span.HTMLPreformattedChar
	{mso-style-name:"HTML Preformatted Char";
	mso-style-priority:99;
	mso-style-link:"HTML Preformatted";
	font-family:Consolas;}
.MsoChpDefault
	{mso-style-type:export-only;}
@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]-->
<div lang=3D"EN-US" link=3D"#0563C1" vlink=3D"#954F72">
<div class=3D"WordSection1">
<p class=3D"MsoNormal">Hi , <o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">This is regarding the section 3.2 BUM Traffic .&nbsp=
;&nbsp; From my reading of the draft it seems that the assumption here is t=
hat BUM traffic will&nbsp; be send with the ESI label&nbsp; even to the PEs=
 which are not connected to the ESI. &nbsp;<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">An alternative solution could be to advertise two &n=
bsp;IM routes per Ethernet Tag
<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">1) One for &nbsp;the AC in leaf mode <o:p></o:p></p>
<p class=3D"MsoNormal">2) Second for the AC in Root mode <o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">The advantage of the above solution is that we don=
=92t need the ESI label in label stack for the packet send towards the set =
of PEs not connected to the ESI .
<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">The use of ESI label could be restricted &nbsp;for f=
orwarding the BUM traffic between MH PEs connected to ESI operating in all-=
active mode, &nbsp;as per the SH procedures &nbsp;described in the RFC 7432=
 .
<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Thanks <o:p></o:p></p>
<p class=3D"MsoNormal">Nitin <o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
</div>
</div>
</div>
</span></div>
</div>
</span>
</body>
</html>

--_000_D138461A12E011sajassiciscocom_--


From nobody Wed Mar 25 09:04:47 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A2F161B2AB8 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:04:37 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tbf2ZcaH0gOP for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:04:34 -0700 (PDT)
Received: from alln-iport-6.cisco.com (alln-iport-6.cisco.com [173.37.142.93]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D880D1B2AD8 for <bess@ietf.org>; Wed, 25 Mar 2015 09:02:51 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=22578; q=dns/txt; s=iport; t=1427299371; x=1428508971; h=from:to:subject:date:message-id:in-reply-to:mime-version; bh=TgHhLSmyebu63hsxPoaa3aqh7SDchB+g+EdiqzNN/g8=; b=F1hIJ/jfeYy0075x64+mruyv7FKzfTWIeneX/gmFuDmV8j6a4PcDxJTA nfDjsXa0R7hMOR+UAuv/rA7tkXdkxCzrLOpzHJHWkn2SDnG0ImcmS4+vj eyuQnmz4tqpfifL9hWFfVVblut1M3ow51a9/WxXU4MA5ia+DllQuT2Ukv A=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0B1BQDe2hJV/5RdJa1cgkNDgSwEywICgVdMAQEBAQEBfYQUAQEBBC1eAQgRAwEBASEHORQHAQEFAwEBBBMbiBSfWalsAQEBAQEBAQEBAQEBAQEBAQEBAQEYiyGEZRcBhC0FkFCJb4EbgzCPYSKCAhyBUG+BRH8BAQE
X-IronPort-AV: E=Sophos;i="5.11,465,1422921600";  d="scan'208,217";a="135310534"
Received: from rcdn-core-12.cisco.com ([173.37.93.148]) by alln-iport-6.cisco.com with ESMTP; 25 Mar 2015 16:02:51 +0000
Received: from xhc-aln-x11.cisco.com (xhc-aln-x11.cisco.com [173.36.12.85]) by rcdn-core-12.cisco.com (8.14.5/8.14.5) with ESMTP id t2PG2o3P017682 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Wed, 25 Mar 2015 16:02:50 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.15]) by xhc-aln-x11.cisco.com ([173.36.12.85]) with mapi id 14.03.0195.001; Wed, 25 Mar 2015 11:02:50 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: draft-sajassi-bess-evpn-etree 
Thread-Index: AdBcUCnZNgMRxjitRJmKKmXMB3yB9gAhSLwAAARqDOD//+dNAIAUdRGA
Date: Wed, 25 Mar 2015 16:02:49 +0000
Message-ID: <D1384653.12E01B%sajassi@cisco.com>
In-Reply-To: <D1271AA8.128E7B%sajassi@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.24.162.66]
Content-Type: multipart/alternative; boundary="_000_D138465312E01Bsajassiciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/hNLmeY3FMnqzdlqpb7eMFljL-jc>
Subject: [bess] FW: draft-sajassi-bess-evpn-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:04:37 -0000

--_000_D138465312E01Bsajassiciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable



From: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>
Date: Thursday, March 12, 2015 at 10:38 AM
To: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>, "draft-s=
ajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@=
tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft=
-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: Re: draft-sajassi-bess-evpn-etree
Resent-From: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>
Resent-To: <draft-sajassi-bess-evpn-etree@ietf.org<mailto:draft-sajassi-bes=
s-evpn-etree@ietf.org>>
Resent-Date: Thursday, March 12, 2015 at 10:38 AM


Hi Nitin,


From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Thursday, March 12, 2015 at 10:25 AM
To: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "draft-sa=
jassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@t=
ools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree

Hi Ali ,

Please note that IM routes are advertised at granularity of Ethernet Tag [ =
IOW , there is a common IM route for ALL Ethernet Segments ].

Hence I agree that there is doubling of IM routes per Ethernet Tag per EVI =
 but this does not have the multiplication factor of number of ESIs. Essent=
ially we will have two multicast groups one for the leafs other one for the=
 roots .

I said the number of IM routes are doubled (e.g., 100% overhead) which is t=
he case.

The issue that I see is that the use ESI label in Etree service is slightly=
 different than the one described for ELAN service in RFC 7432 .

In ELAN service , ESI label is only used between the set of MH PEs connecte=
d to the ESI , but with the approach described in ETree even the PEs not co=
nnected to the ESI have to create forwarding state/SH procedures using the =
ESI label .

That=92s not correct =96 i.e., ESI label is NOT just used among MH PEs.  Et=
her=96AD per ES route along with ESI label is advertised to all PEs partici=
pating in the VPNs. There is a forwarding state that is created for ESI lab=
el for P2MP LSP anyway (but not for ingress replication).

Also there can be single homed AC in root/leaf roles , there will be no AD =
route per ESI routes for these single homed AC hence we cannot use the SH m=
echanisms  based on ESI label .

That=92s why we only need to advertise two such labels per PE (one for root=
 and another for leaf)!

CHeers,
Ali

Thanks
Nitin





From: Ali Sajassi (sajassi) [mailto:sajassi@cisco.com]
Sent: Thursday, March 12, 2015 10:01 AM
To: Nitin Singh; draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>
Cc: Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake; Ali Sajassi=
 (sajassi)
Subject: Re: draft-sajassi-bess-evpn-etree


There are bunch of disadvantages for doing it this way:


  1.  That=92s a lot of advertisements if we do it with IM route - i.e., th=
e BGP messages for IM gets increased by 100% !! A given ES can support hund=
reds or thousands of IM routes !!
  2.  In data-plane we end up increasing the size of LFIB
  3.  There are now to different approaches for single-home versus multi-ho=
med sites


Cheers,
Ali

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Wednesday, March 11, 2015 at 5:14 PM
To: "draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess=
-evpn-etree@tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<=
mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: draft-sajassi-bess-evpn-etree
Resent-From: <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Resent-To: <draft-sajassi-bess-evpn-etree@ietf.org<mailto:draft-sajassi-bes=
s-evpn-etree@ietf.org>>
Resent-Date: Thursday, March 12, 2015 at 4:50 AM

Hi ,

This is regarding the section 3.2 BUM Traffic .   From my reading of the dr=
aft it seems that the assumption here is that BUM traffic will  be send wit=
h the ESI label  even to the PEs which are not connected to the ESI.

An alternative solution could be to advertise two  IM routes per Ethernet T=
ag

1) One for  the AC in leaf mode
2) Second for the AC in Root mode

The advantage of the above solution is that we don=92t need the ESI label i=
n label stack for the packet send towards the set of PEs not connected to t=
he ESI .

The use of ESI label could be restricted  for forwarding the BUM traffic be=
tween MH PEs connected to ESI operating in all-active mode,  as per the SH =
procedures  described in the RFC 7432 .

Thanks
Nitin





--_000_D138465312E01Bsajassiciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <9F8339475403814BB16A3D80E49F328F@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div><br>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Cisco Employee &lt;<a href=3D=
"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Thursday, March 12, 2015 at 1=
0:38 AM<br>
<span style=3D"font-weight:bold">To: </span>Nitin Singh &lt;<a href=3D"mail=
to:nisingh@juniper.net">nisingh@juniper.net</a>&gt;, &quot;<a href=3D"mailt=
o:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etr=
ee@tools.ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-e=
tree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<b=
r>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>Re: draft-sajassi-bess-evp=
n-etree <br>
<span style=3D"font-weight:bold">Resent-From: </span>Cisco Employee &lt;<a =
href=3D"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-To: </span>&lt;<a href=3D"mailto:dr=
aft-sajassi-bess-evpn-etree@ietf.org">draft-sajassi-bess-evpn-etree@ietf.or=
g</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-Date: </span>Thursday, March 12, 20=
15 at 10:38 AM<br>
</div>
<div><br>
</div>
<div>
<div style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line=
-break: after-white-space;">
<div style=3D"font-family: Calibri, sans-serif; font-size: 14px; color: rgb=
(0, 0, 0);">
<br>
</div>
<div style=3D"font-family: Calibri, sans-serif; font-size: 14px;"><font col=
or=3D"#ff0000">Hi Nitin,</font></div>
<div style=3D"font-family: Calibri, sans-serif; font-size: 14px; color: rgb=
(0, 0, 0);">
<br>
</div>
<div style=3D"font-family: Calibri, sans-serif; font-size: 14px; color: rgb=
(0, 0, 0);">
<br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION" style=3D"font-family: Calibri, sans-serif=
; font-size: 14px; color: rgb(0, 0, 0);">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Nitin Singh &lt;<a href=3D"ma=
ilto:nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Thursday, March 12, 2015 at 1=
0:25 AM<br>
<span style=3D"font-weight:bold">To: </span>Cisco Employee &lt;<a href=3D"m=
ailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;, &quot;<a href=3D"mailto=
:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etre=
e@tools.ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-et=
ree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<br=
>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>RE: draft-sajassi-bess-evp=
n-etree <br>
</div>
<div><br>
</div>
<div>
<meta name=3D"Generator" content=3D"Microsoft Exchange Server">
<!-- converted from rtf --><style><!-- .EmailQuote { margin-left: 1pt; padd=
ing-left: 4pt; border-left: #800000 2px solid; } --></style>
<div><font face=3D"Calibri Light" size=3D"3"><span style=3D"font-size:12pt;=
">
<div><font color=3D"#002060">Hi Ali , </font></div>
<div><font color=3D"#002060">&nbsp;</font></div>
<div><font color=3D"#002060">Please note that IM routes are advertised at g=
ranularity of Ethernet Tag [ IOW , there is a common IM route for ALL Ether=
net Segments ].
</font></div>
<div><font color=3D"#002060">&nbsp;</font></div>
<div><font color=3D"#002060">Hence I agree that there is doubling of IM rou=
tes per Ethernet Tag per EVI&nbsp; but this does not have the multiplicatio=
n factor of number of ESIs. Essentially we will have two multicast groups o=
ne for the leafs other one for the roots
 .</font></div>
</span></font></div>
</div>
</span>
<div style=3D"font-family: Calibri, sans-serif; font-size: 14px;"><br>
</div>
<div style=3D"font-family: Calibri, sans-serif; font-size: 14px;"><font col=
or=3D"#ff0000">I said the number of IM routes are doubled (e.g., 100% overh=
ead) which is the case.</font></div>
<span id=3D"OLK_SRC_BODY_SECTION" style=3D"font-family: Calibri, sans-serif=
; font-size: 14px; color: rgb(0, 0, 0);">
<div>
<div><font face=3D"Calibri Light" size=3D"3"><span style=3D"font-size:12pt;=
">
<div><font color=3D"#002060"></font></div>
<div><font face=3D"Calibri" size=3D"2" color=3D"#002060"><span style=3D"fon=
t-size:11pt;">&nbsp;</span></font></div>
<div><font color=3D"#002060">The issue that I see is that the use ESI label=
 in Etree service is slightly different than the one described for ELAN ser=
vice in RFC 7432 .</font></div>
<div><font face=3D"Calibri" size=3D"2" color=3D"#002060"><span style=3D"fon=
t-size:11pt;">&nbsp;</span></font></div>
<div><font color=3D"#002060">In ELAN service , ESI label is only used betwe=
en the set of MH PEs connected to the ESI , but with the approach described=
 in ETree even the PEs not connected to the ESI have to create forwarding s=
tate/SH procedures using the ESI label
 .</font></div>
</span></font></div>
</div>
</span>
<div style=3D"font-family: Calibri, sans-serif; font-size: 14px;"><br>
</div>
<div><font color=3D"#ff0000"><font face=3D"Calibri,sans-serif">That=92s not=
 correct&nbsp;=96 i.e., ESI label is NOT just used among MH PEs. &nbsp;Ethe=
r=96AD per ES route along with ESI label is advertised to all&nbsp;PEs part=
icipating in the VPNs. There is a forwarding state that is
 created for ESI label for P2MP LSP anyway (but not for ingress replication=
).</font></font></div>
<span id=3D"OLK_SRC_BODY_SECTION" style=3D"font-family: Calibri, sans-serif=
; font-size: 14px; color: rgb(0, 0, 0);">
<div>
<div><font face=3D"Calibri Light" size=3D"3"><span style=3D"font-size:12pt;=
">
<div><font color=3D"#002060"></font></div>
<div><font face=3D"Calibri" size=3D"2" color=3D"#002060"><span style=3D"fon=
t-size:11pt;">&nbsp;</span></font></div>
<div><font color=3D"#002060">Also there can be single homed AC in root/leaf=
 roles , there will be no AD route per ESI routes for these single homed AC=
 hence we cannot use the SH mechanisms&nbsp; based on ESI label .</font></d=
iv>
</span></font></div>
</div>
</span>
<div><br>
</div>
<div><font color=3D"#ff0000">That=92s why we only need to advertise two suc=
h labels per PE (one for root and another for leaf)!</font></div>
<div><font color=3D"#ff0000"><br>
</font></div>
<div><font color=3D"#ff0000">CHeers,</font></div>
<div><font color=3D"#ff0000">Ali</font></div>
<span id=3D"OLK_SRC_BODY_SECTION" style=3D"font-family: Calibri, sans-serif=
; font-size: 14px; color: rgb(0, 0, 0);">
<div>
<div><font face=3D"Calibri Light" size=3D"3"><span style=3D"font-size:12pt;=
">
<div><font color=3D"#002060"></font></div>
<div><font face=3D"Calibri" size=3D"2" color=3D"#002060"><span style=3D"fon=
t-size:11pt;">&nbsp;</span></font></div>
<div><font color=3D"#002060">Thanks </font></div>
<div><font color=3D"#002060">Nitin </font></div>
<div><font face=3D"Calibri" size=3D"2" color=3D"#1F497D"><span style=3D"fon=
t-size:11pt;">&nbsp;</span></font></div>
<div><font face=3D"Calibri" size=3D"2" color=3D"#1F497D"><span style=3D"fon=
t-size:11pt;">&nbsp;</span></font></div>
<div><font face=3D"Calibri" size=3D"2" color=3D"#1F497D"><span style=3D"fon=
t-size:11pt;">&nbsp;</span></font></div>
<div><font face=3D"Calibri" size=3D"2" color=3D"#1F497D"><span style=3D"fon=
t-size:11pt;">&nbsp;</span></font></div>
<div><font face=3D"Calibri" size=3D"2" color=3D"#1F497D"><span style=3D"fon=
t-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;"><b>From:</b> Ali Sajassi (sajassi) [<a href=3D"mai=
lto:sajassi@cisco.com">mailto:sajassi@cisco.com</a>]
<br>
<b>Sent:</b> Thursday, March 12, 2015 10:01 AM<br>
<b>To:</b> Nitin Singh; <a href=3D"mailto:draft-sajassi-bess-evpn-etree@too=
ls.ietf.org">
draft-sajassi-bess-evpn-etree@tools.ietf.org</a><br>
<b>Cc:</b> Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake; Ali =
Sajassi (sajassi)<br>
<b>Subject:</b> Re: draft-sajassi-bess-evpn-etree </span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:10.5pt;">There are bunch of disadvantages for doing it th=
is way:</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<ol style=3D"margin:0;padding-left:72pt;">
<font face=3D"Calibri" size=3D"2"><span style=3D"font-size:10.5pt;">
<li style=3D"margin-top:5pt;margin-bottom:5pt;">That=92s a lot of advertise=
ments if we do it with IM route - i.e., the BGP messages for IM gets increa=
sed by 100% !! A given ES can support hundreds or thousands of IM routes !!=
</span></font></li><li style=3D"margin-top:5pt;margin-bottom:5pt;"><font fa=
ce=3D"Calibri" size=3D"2">In data-plane we end up increasing the size of LF=
IB</font></li><li style=3D"margin-top:5pt;margin-bottom:5pt;"><font face=3D=
"Calibri" size=3D"2">There are now to different approaches for single-home =
versus multi-homed sites</font></li></ol>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:10.5pt;">Cheers,</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:10.5pt;">Ali</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Lucida Grande" size=3D"2"><=
span style=3D"font-size:11pt;"><b>From:
</b>Nitin Singh &lt;<a href=3D"mailto:nisingh@juniper.net"><font color=3D"#=
0563C1"><u>nisingh@juniper.net</u></font></a>&gt;<br>
<b>Date: </b>Wednesday, March 11, 2015 at 5:14 PM<br>
<b>To: </b>&quot;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@tools.ietf=
.org"><font color=3D"#0563C1"><u>draft-sajassi-bess-evpn-etree@tools.ietf.o=
rg</u></font></a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree=
@tools.ietf.org"><font color=3D"#0563C1"><u>draft-sajassi-bess-evpn-etree@t=
ools.ietf.org</u></font></a>&gt;<br>
<b>Cc: </b>Hassan Hosseini &lt;<a href=3D"mailto:hhosseini@juniper.net"><fo=
nt color=3D"#0563C1"><u>hhosseini@juniper.net</u></font></a>&gt;, Ryan Bick=
hart &lt;<a href=3D"mailto:rbickhart@juniper.net"><font color=3D"#0563C1"><=
u>rbickhart@juniper.net</u></font></a>&gt;, Manoj Sharma
 &lt;<a href=3D"mailto:manojs@juniper.net"><font color=3D"#0563C1"><u>manoj=
s@juniper.net</u></font></a>&gt;, John E Drake &lt;<a href=3D"mailto:jdrake=
@juniper.net"><font color=3D"#0563C1"><u>jdrake@juniper.net</u></font></a>&=
gt;<br>
<b>Subject: </b>draft-sajassi-bess-evpn-etree <br>
<b>Resent-From: </b>&lt;<a href=3D"mailto:nisingh@juniper.net"><font color=
=3D"#0563C1"><u>nisingh@juniper.net</u></font></a>&gt;<br>
<b>Resent-To: </b>&lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@ietf.=
org"><font color=3D"#0563C1"><u>draft-sajassi-bess-evpn-etree@ietf.org</u><=
/font></a>&gt;<br>
<b>Resent-Date: </b>Thursday, March 12, 2015 at 4:50 AM</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">Hi ,
</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">This is regarding the section 3.2 BUM Traffic .&nb=
sp;&nbsp; From my reading of the draft it seems that the assumption here is=
 that BUM traffic will&nbsp; be send with the ESI label&nbsp;
 even to the PEs which are not connected to the ESI. &nbsp;</span></font></=
div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">An alternative solution could be to advertise two =
&nbsp;IM routes per Ethernet Tag
</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">1) One for &nbsp;the AC in leaf mode
</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">2) Second for the AC in Root mode
</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">The advantage of the above solution is that we don=
=92t need the ESI label in label stack for the packet send towards the set =
of PEs not connected to the ESI .
</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">The use of ESI label could be restricted &nbsp;for=
 forwarding the BUM traffic between MH PEs connected to ESI operating in al=
l-active mode, &nbsp;as per the SH procedures &nbsp;described
 in the RFC 7432 . </span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">Thanks
</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">Nitin
</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
<div style=3D"padding-left:36pt;"><font face=3D"Calibri" size=3D"2"><span s=
tyle=3D"font-size:11pt;">&nbsp;</span></font></div>
</span></font></div>
</div>
</span></div>
</div>
</span>
</body>
</html>

--_000_D138465312E01Bsajassiciscocom_--


From nobody Wed Mar 25 09:05:23 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 45CF91B2A48 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:05:21 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KCf-NS0qR26K for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:05:18 -0700 (PDT)
Received: from alln-iport-1.cisco.com (alln-iport-1.cisco.com [173.37.142.88]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 6031D1A87A7 for <bess@ietf.org>; Wed, 25 Mar 2015 09:03:23 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=39130; q=dns/txt; s=iport; t=1427299403; x=1428509003; h=from:to:subject:date:message-id:in-reply-to:mime-version; bh=ckuT0SCbAYrLG9FDSXr7A5BJIQdumVZXo7PseA37Wls=; b=BOniE2PlPSvU1jzfepAvi8B9xUVh+gimFIO7UW9k3pq2pDq0zohqVypY BLTkfaMQ0il1wV6r6LWpKRsApuXUkGa7hY3ipAjc9pjQAhmwECf4n+CgU 1qFg5jdkojc5mPb0Brw2XYB+zBrLNa6GCxt05k+D/Ek9aZNV3q8Y9PAsC I=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0B2BQDP2xJV/4wNJK1cgkNDUloEywICgVdMAQEBAQEBfYQUAQEBBC1eAQgRAwEBASEBBjkUBwEBBQMCBBMbiBSfXKlvAQEBAQEBAQEBAQEBAQEBAQEBAQEBF4shhGUXAYQtBY5Cgg6Jb4EbgzCMGoNHIoICHIFQb4FEfwEBAQ
X-IronPort-AV: E=Sophos;i="5.11,465,1422921600";  d="scan'208,217";a="135258480"
Received: from alln-core-7.cisco.com ([173.36.13.140]) by alln-iport-1.cisco.com with ESMTP; 25 Mar 2015 16:03:22 +0000
Received: from xhc-rcd-x12.cisco.com (xhc-rcd-x12.cisco.com [173.37.183.86]) by alln-core-7.cisco.com (8.14.5/8.14.5) with ESMTP id t2PG3Mlb016305 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Wed, 25 Mar 2015 16:03:22 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.15]) by xhc-rcd-x12.cisco.com ([173.37.183.86]) with mapi id 14.03.0195.001; Wed, 25 Mar 2015 11:03:22 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: draft-sajassi-bess-evpn-etree 
Thread-Index: AdBcUCnZNgMRxjitRJmKKmXMB3yB9gAhSLwAAARqDOD//+dNAP//2gLAgAA0pwCAFGaPAA==
Date: Wed, 25 Mar 2015 16:03:22 +0000
Message-ID: <D1384663.12E01D%sajassi@cisco.com>
In-Reply-To: <D1272923.128F98%sajassi@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.24.162.66]
Content-Type: multipart/alternative; boundary="_000_D138466312E01Dsajassiciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/X4FXp5lHq-mCYvGoOB1Sq5xE728>
Subject: [bess] FW: draft-sajassi-bess-evpn-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:05:21 -0000

--_000_D138466312E01Dsajassiciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable



From: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>
Date: Thursday, March 12, 2015 at 11:31 AM
To: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>, "draft-s=
ajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@=
tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft=
-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: Re: draft-sajassi-bess-evpn-etree
Resent-From: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>
Resent-To: <draft-sajassi-bess-evpn-etree@ietf.org<mailto:draft-sajassi-bes=
s-evpn-etree@ietf.org>>
Resent-Date: Thursday, March 12, 2015 at 11:31 AM


Exactly!

Cheers,
Ali

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Thursday, March 12, 2015 at 11:02 AM
To: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "draft-sa=
jassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@t=
ools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree

Hi Ali ,

Thanks for your response,  it seems like that the proposal for SH segments =
 is to advertise the 2  AD routes per ESI with the ESI value encoded as 0 ,=
 one for SH segments in  Leaf role and other one the SH segments in Root mo=
de .

Regards
Nitin

From: Ali Sajassi (sajassi) [mailto:sajassi@cisco.com]
Sent: Thursday, March 12, 2015 10:39 AM
To: Nitin Singh; draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>
Cc: Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake
Subject: Re: draft-sajassi-bess-evpn-etree


Hi Nitin,


From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Thursday, March 12, 2015 at 10:25 AM
To: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "draft-sa=
jassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@t=
ools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree

Hi Ali ,

Please note that IM routes are advertised at granularity of Ethernet Tag [ =
IOW , there is a common IM route for ALL Ethernet Segments ].

Hence I agree that there is doubling of IM routes per Ethernet Tag per EVI =
 but this does not have the multiplication factor of number of ESIs. Essent=
ially we will have two multicast groups one for the leafs other one for the=
 roots .

I said the number of IM routes are doubled (e.g., 100% overhead) which is t=
he case.

The issue that I see is that the use ESI label in Etree service is slightly=
 different than the one described for ELAN service in RFC 7432 .

In ELAN service , ESI label is only used between the set of MH PEs connecte=
d to the ESI , but with the approach described in ETree even the PEs not co=
nnected to the ESI have to create forwarding state/SH procedures using the =
ESI label .

That=92s not correct =96 i.e., ESI label is NOT just used among MH PEs.  Et=
her=96AD per ES route along with ESI label is advertised to all PEs partici=
pating in the VPNs. There is a forwarding state that is created for ESI lab=
el for P2MP LSP anyway (but not for ingress replication).

Also there can be single homed AC in root/leaf roles , there will be no AD =
route per ESI routes for these single homed AC hence we cannot use the SH m=
echanisms  based on ESI label .

That=92s why we only need to advertise two such labels per PE (one for root=
 and another for leaf)!

CHeers,
Ali

Thanks
Nitin





From: Ali Sajassi (sajassi) [mailto:sajassi@cisco.com]
Sent: Thursday, March 12, 2015 10:01 AM
To: Nitin Singh; draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>
Cc: Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake; Ali Sajassi=
 (sajassi)
Subject: Re: draft-sajassi-bess-evpn-etree


There are bunch of disadvantages for doing it this way:

1.       That=92s a lot of advertisements if we do it with IM route - i.e.,=
 the BGP messages for IM gets increased by 100% !! A given ES can support h=
undreds or thousands of IM routes !!
2.      In data-plane we end up increasing the size of LFIB
3.      There are now to different approaches for single-home versus multi-=
homed sites

Cheers,
Ali

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Wednesday, March 11, 2015 at 5:14 PM
To: "draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess=
-evpn-etree@tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<=
mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: draft-sajassi-bess-evpn-etree
Resent-From: <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Resent-To: <draft-sajassi-bess-evpn-etree@ietf.org<mailto:draft-sajassi-bes=
s-evpn-etree@ietf.org>>
Resent-Date: Thursday, March 12, 2015 at 4:50 AM

Hi ,

This is regarding the section 3.2 BUM Traffic .   From my reading of the dr=
aft it seems that the assumption here is that BUM traffic will  be send wit=
h the ESI label  even to the PEs which are not connected to the ESI.

An alternative solution could be to advertise two  IM routes per Ethernet T=
ag

1) One for  the AC in leaf mode
2) Second for the AC in Root mode

The advantage of the above solution is that we don=92t need the ESI label i=
n label stack for the packet send towards the set of PEs not connected to t=
he ESI .

The use of ESI label could be restricted  for forwarding the BUM traffic be=
tween MH PEs connected to ESI operating in all-active mode,  as per the SH =
procedures  described in the RFC 7432 .

Thanks
Nitin





--_000_D138466312E01Dsajassiciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <BF76DEBE2935BD44855488E673E79BA2@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div><br>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Cisco Employee &lt;<a href=3D=
"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Thursday, March 12, 2015 at 1=
1:31 AM<br>
<span style=3D"font-weight:bold">To: </span>Nitin Singh &lt;<a href=3D"mail=
to:nisingh@juniper.net">nisingh@juniper.net</a>&gt;, &quot;<a href=3D"mailt=
o:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etr=
ee@tools.ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-e=
tree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<b=
r>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>Re: draft-sajassi-bess-evp=
n-etree <br>
<span style=3D"font-weight:bold">Resent-From: </span>Cisco Employee &lt;<a =
href=3D"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-To: </span>&lt;<a href=3D"mailto:dr=
aft-sajassi-bess-evpn-etree@ietf.org">draft-sajassi-bess-evpn-etree@ietf.or=
g</a>&gt;<br>
<span style=3D"font-weight:bold">Resent-Date: </span>Thursday, March 12, 20=
15 at 11:31 AM<br>
</div>
<div><br>
</div>
<div>
<div style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line=
-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-famil=
y: Calibri, sans-serif;">
<div><br>
</div>
<div>Exactly!&nbsp;</div>
<div><br>
</div>
<div>Cheers,</div>
<div>Ali</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Nitin Singh &lt;<a href=3D"ma=
ilto:nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Thursday, March 12, 2015 at 1=
1:02 AM<br>
<span style=3D"font-weight:bold">To: </span>Cisco Employee &lt;<a href=3D"m=
ailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;, &quot;<a href=3D"mailto=
:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etre=
e@tools.ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-et=
ree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<br=
>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>RE: draft-sajassi-bess-evp=
n-etree <br>
</div>
<div><br>
</div>
<div xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micro=
soft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" x=
mlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:/=
/www.w3.org/TR/REC-html40">
<meta name=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:"Calibri Light";
	panose-1:2 15 3 2 2 2 4 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:"Lucida Grande";
	panose-1:0 0 0 0 0 0 0 0 0 0;}
/* 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.emailquote, li.emailquote, div.emailquote
	{mso-style-name:emailquote;
	mso-margin-top-alt:auto;
	margin-right:0in;
	mso-margin-bottom-alt:auto;
	margin-left:1.0pt;
	border:none;
	padding:0in;
	font-size:12.0pt;
	font-family:"Times New Roman",serif;}
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;}
/* List Definitions */
@list l0
	{mso-list-id:1949703470;
	mso-list-template-ids:1774458198;}
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]-->
<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">Hi Ali ,
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">Thanks for your response,&nbsp; it =
seems like that the proposal for SH segments &nbsp;is to advertise the 2 &n=
bsp;AD routes per ESI with the ESI value encoded as
 0 , one for SH segments in &nbsp;Leaf role and other one the SH segments i=
n Root mode .
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">Regards<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">Nitin
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<div>
<div style=3D"border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><b><span style=3D"font-si=
ze: 11pt; font-family: Calibri, sans-serif;">From:</span></b><span style=3D=
"font-size: 11pt; font-family: Calibri, sans-serif;"> Ali Sajassi (sajassi)=
 [<a href=3D"mailto:sajassi@cisco.com">mailto:sajassi@cisco.com</a>]
<br>
<b>Sent:</b> Thursday, March 12, 2015 10:39 AM<br>
<b>To:</b> Nitin Singh; <a href=3D"mailto:draft-sajassi-bess-evpn-etree@too=
ls.ietf.org">
draft-sajassi-bess-evpn-etree@tools.ietf.org</a><br>
<b>Cc:</b> Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake<br>
<b>Subject:</b> Re: draft-sajassi-bess-evpn-etree <o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><o:p>&nbsp;</o:p></p>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: red;">Hi Nitin,</span><sp=
an style=3D"font-size: 10.5pt; font-family: Calibri, sans-serif;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></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"margin-left:.5in"><b><span style=3D"font-si=
ze: 11pt; font-family: 'Lucida Grande', serif; color: black;">From:
</span></b><span style=3D"font-size: 11pt; font-family: 'Lucida Grande', se=
rif; color: black;">Nitin Singh &lt;<a href=3D"mailto:nisingh@juniper.net">=
nisingh@juniper.net</a>&gt;<br>
<b>Date: </b>Thursday, March 12, 2015 at 10:25 AM<br>
<b>To: </b>Cisco Employee &lt;<a href=3D"mailto:sajassi@cisco.com">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@to=
ols.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&quot; &lt;<a=
 href=3D"mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi=
-bess-evpn-etree@tools.ietf.org</a>&gt;<br>
<b>Cc: </b>Hassan Hosseini &lt;<a href=3D"mailto:hhosseini@juniper.net">hho=
sseini@juniper.net</a>&gt;, Ryan Bickhart &lt;<a href=3D"mailto:rbickhart@j=
uniper.net">rbickhart@juniper.net</a>&gt;, Manoj Sharma &lt;<a href=3D"mail=
to:manojs@juniper.net">manojs@juniper.net</a>&gt;, John
 E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net</a>&g=
t;<br>
<b>Subject: </b>RE: draft-sajassi-bess-evpn-etree <o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Hi Ali ,
</span><span style=3D"font-family: 'Calibri Light', sans-serif; color: blac=
k;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Please note that IM=
 routes are advertised at granularity of Ethernet Tag [ IOW , there is a co=
mmon IM route for ALL Ethernet Segments
 ]. </span><span style=3D"font-family: 'Calibri Light', sans-serif; color: =
black;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Hence I agree that =
there is doubling of IM routes per Ethernet Tag per EVI&nbsp; but this does=
 not have the multiplication factor of number
 of ESIs. Essentially we will have two multicast groups one for the leafs o=
ther one for the roots .</span><span style=3D"font-family: 'Calibri Light',=
 sans-serif; color: black;"><o:p></o:p></span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif;"><o:p>&nbsp;</o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: red;">I said the number o=
f IM routes are doubled (e.g., 100% overhead) which is the case.</span><spa=
n style=3D"font-size: 10.5pt; font-family: Calibri, sans-serif;"><o:p></o:p=
></span></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</sp=
an><span style=3D"font-family: 'Calibri Light', sans-serif; color: black;">=
<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">The issue that I se=
e is that the use ESI label in Etree service is slightly different than the=
 one described for ELAN service in RFC
 7432 .</span><span style=3D"font-family: 'Calibri Light', sans-serif; colo=
r: black;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</sp=
an><span style=3D"font-family: 'Calibri Light', sans-serif; color: black;">=
<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">In ELAN service , E=
SI label is only used between the set of MH PEs connected to the ESI , but =
with the approach described in ETree even
 the PEs not connected to the ESI have to create forwarding state/SH proced=
ures using the ESI label .</span><span style=3D"font-family: 'Calibri Light=
', sans-serif; color: black;"><o:p></o:p></span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif;"><o:p>&nbsp;</o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: Calibri, sans-serif; color: red;">That=92s not correct&nbsp;=96 i.e., ES=
I label is NOT just used among MH PEs. &nbsp;Ether=96AD per ES route along =
with ESI label is advertised to all&nbsp;PEs participating
 in the VPNs. There is a forwarding state that is created for ESI label for=
 P2MP LSP anyway (but not for ingress replication).</span><o:p></o:p></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</sp=
an><span style=3D"font-family: 'Calibri Light', sans-serif; color: black;">=
<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Also there can be s=
ingle homed AC in root/leaf roles , there will be no AD route per ESI route=
s for these single homed AC hence we cannot
 use the SH mechanisms&nbsp; based on ESI label .</span><span style=3D"font=
-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:p></span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"color:red"=
>That=92s why we only need to advertise two such labels per PE (one for roo=
t and another for leaf)!</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><o:p>&nbsp;</o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"color:red"=
>CHeers,</span><o:p></o:p></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"color:red"=
>Ali</span><o:p></o:p></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</sp=
an><span style=3D"font-family: 'Calibri Light', sans-serif; color: black;">=
<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Thanks
</span><span style=3D"font-family: 'Calibri Light', sans-serif; color: blac=
k;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-famil=
y: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Nitin
</span><span style=3D"font-family: 'Calibri Light', sans-serif; color: blac=
k;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"font-family: 'Calibri Light', sans-serif; color: black;=
"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"font-family: 'Calibri Light', sans-serif; color: black;=
"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"font-family: 'Calibri Light', sans-serif; color: black;=
"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"font-family: 'Calibri Light', sans-serif; color: black;=
"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"font-family: 'Calibri Light', sans-serif; color: black;=
"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><b><span style=3D"font-si=
ze: 11pt; font-family: Calibri, sans-serif; color: black;">From:</span></b>=
<span style=3D"font-size: 11pt; font-family: Calibri, sans-serif; color: bl=
ack;"> Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com">mailto:s=
ajassi@cisco.com</a>]
<br>
<b>Sent:</b> Thursday, March 12, 2015 10:01 AM<br>
<b>To:</b> Nitin Singh; <a href=3D"mailto:draft-sajassi-bess-evpn-etree@too=
ls.ietf.org">
draft-sajassi-bess-evpn-etree@tools.ietf.org</a><br>
<b>Cc:</b> Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake; Ali =
Sajassi (sajassi)<br>
<b>Subject:</b> Re: draft-sajassi-bess-evpn-etree </span><span style=3D"fon=
t-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:p></span></p=
>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;">There are bunch o=
f disadvantages for doing it this way:</span><span style=3D"font-family: 'C=
alibri Light', sans-serif; color: black;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:5.0pt;margin-right:0in;m=
argin-bottom:5.0pt;margin-left:.5in;text-indent:-.25in;mso-list:l0 level1 l=
fo1">
<!--[if !supportLists]--><span style=3D"font-size:10.5pt;font-family:&quot;=
Calibri&quot;,sans-serif;color:black"><span style=3D"mso-list:Ignore">1.<sp=
an style=3D"font-style: normal; font-variant: normal; font-weight: normal; =
font-size: 7pt; line-height: normal; font-family: 'Times New Roman';">&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span></span><!--[endif]--><span style=3D"font-size: 10.5pt; font-f=
amily: Calibri, sans-serif; color: black;">That=92s a lot of advertisements=
 if we do it with IM route - i.e., the BGP messages for IM gets increased b=
y 100% !! A given ES can support hundreds
 or thousands of IM routes !!<o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:5.0pt;margin-right:0in;m=
argin-bottom:5.0pt;margin-left:.5in;text-indent:-.25in;mso-list:l0 level1 l=
fo1">
<!--[if !supportLists]--><span style=3D"font-family:&quot;Calibri Light&quo=
t;,sans-serif;color:black"><span style=3D"mso-list:Ignore">2.<span style=3D=
"font-style: normal; font-variant: normal; font-weight: normal; font-size: =
7pt; line-height: normal; font-family: 'Times New Roman';">&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;
</span></span></span><!--[endif]--><span style=3D"font-size: 10.5pt; font-f=
amily: Calibri, sans-serif; color: black;">In data-plane we end up increasi=
ng the size of LFIB</span><span style=3D"font-family: 'Calibri Light', sans=
-serif; color: black;"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:5.0pt;margin-right:0in;m=
argin-bottom:5.0pt;margin-left:.5in;text-indent:-.25in;mso-list:l0 level1 l=
fo1">
<!--[if !supportLists]--><span style=3D"font-family:&quot;Calibri Light&quo=
t;,sans-serif;color:black"><span style=3D"mso-list:Ignore">3.<span style=3D=
"font-style: normal; font-variant: normal; font-weight: normal; font-size: =
7pt; line-height: normal; font-family: 'Times New Roman';">&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;
</span></span></span><!--[endif]--><span style=3D"font-size: 10pt; font-fam=
ily: Calibri, sans-serif; color: black;">There are now to different approac=
hes for single-home versus multi-homed sites</span><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: black;"><o:p></o:p></span></p>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;">Cheers,</span><sp=
an style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p><=
/o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;">Ali</span><span s=
tyle=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:p=
></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><b><span style=3D"font-si=
ze: 11pt; font-family: 'Lucida Grande', serif; color: black;">From:
</span></b><span style=3D"font-size: 11pt; font-family: 'Lucida Grande', se=
rif; color: black;">Nitin Singh &lt;<a href=3D"mailto:nisingh@juniper.net">=
<span style=3D"color:#0563C1">nisingh@juniper.net</span></a>&gt;<br>
<b>Date: </b>Wednesday, March 11, 2015 at 5:14 PM<br>
<b>To: </b>&quot;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@tools.ietf=
.org"><span style=3D"color:#0563C1">draft-sajassi-bess-evpn-etree@tools.iet=
f.org</span></a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@=
tools.ietf.org"><span style=3D"color:#0563C1">draft-sajassi-bess-evpn-etree=
@tools.ietf.org</span></a>&gt;<br>
<b>Cc: </b>Hassan Hosseini &lt;<a href=3D"mailto:hhosseini@juniper.net"><sp=
an style=3D"color:#0563C1">hhosseini@juniper.net</span></a>&gt;, Ryan Bickh=
art &lt;<a href=3D"mailto:rbickhart@juniper.net"><span style=3D"color:#0563=
C1">rbickhart@juniper.net</span></a>&gt;, Manoj Sharma
 &lt;<a href=3D"mailto:manojs@juniper.net"><span style=3D"color:#0563C1">ma=
nojs@juniper.net</span></a>&gt;, John E Drake &lt;<a href=3D"mailto:jdrake@=
juniper.net"><span style=3D"color:#0563C1">jdrake@juniper.net</span></a>&gt=
;<br>
<b>Subject: </b>draft-sajassi-bess-evpn-etree <br>
<b>Resent-From: </b>&lt;<a href=3D"mailto:nisingh@juniper.net"><span style=
=3D"color:#0563C1">nisingh@juniper.net</span></a>&gt;<br>
<b>Resent-To: </b>&lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@ietf.=
org"><span style=3D"color:#0563C1">draft-sajassi-bess-evpn-etree@ietf.org</=
span></a>&gt;<br>
<b>Resent-Date: </b>Thursday, March 12, 2015 at 4:50 AM</span><span style=
=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:p></s=
pan></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">Hi ,
</span><span style=3D"font-family: 'Calibri Light', sans-serif; color: blac=
k;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">This is regarding t=
he section 3.2 BUM Traffic .&nbsp;&nbsp; From my reading of the draft it se=
ems that the assumption here is that BUM traffic
 will&nbsp; be send with the ESI label&nbsp; even to the PEs which are not =
connected to the ESI. &nbsp;</span><span style=3D"font-family: 'Calibri Lig=
ht', sans-serif; color: black;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">An alternative solu=
tion could be to advertise two &nbsp;IM routes per Ethernet Tag
</span><span style=3D"font-family: 'Calibri Light', sans-serif; color: blac=
k;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">1) One for &nbsp;th=
e AC in leaf mode
</span><span style=3D"font-family: 'Calibri Light', sans-serif; color: blac=
k;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">2) Second for the A=
C in Root mode
</span><span style=3D"font-family: 'Calibri Light', sans-serif; color: blac=
k;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">The advantage of th=
e above solution is that we don=92t need the ESI label in label stack for t=
he packet send towards the set of PEs not
 connected to the ESI . </span><span style=3D"font-family: 'Calibri Light',=
 sans-serif; color: black;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">The use of ESI labe=
l could be restricted &nbsp;for forwarding the BUM traffic between MH PEs c=
onnected to ESI operating in all-active mode,
 &nbsp;as per the SH procedures &nbsp;described in the RFC 7432 . </span><s=
pan style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p>=
</o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">Thanks
</span><span style=3D"font-family: 'Calibri Light', sans-serif; color: blac=
k;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">Nitin
</span><span style=3D"font-family: 'Calibri Light', sans-serif; color: blac=
k;"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span =
style=3D"font-family: 'Calibri Light', sans-serif; color: black;"><o:p></o:=
p></span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</span></div>
</div>
</span>
</body>
</html>

--_000_D138466312E01Dsajassiciscocom_--


From nobody Wed Mar 25 09:07:02 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9E29C1B2AB9 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:07:00 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Gj0oTZWcsDNP for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:06:57 -0700 (PDT)
Received: from alln-iport-7.cisco.com (alln-iport-7.cisco.com [173.37.142.94]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E82C61B2AC4 for <bess@ietf.org>; Wed, 25 Mar 2015 09:05:09 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=42414; q=dns/txt; s=iport; t=1427299509; x=1428509109; h=from:to:subject:date:message-id:in-reply-to:mime-version; bh=wSj72Cz8DG6yvSq1frmiO1k9jDcNS13z0KzX71zaKq8=; b=DVkx0vlnKtQsKEByOLgKF2T46GZzLhjapCsRVO4/IoePxRA6cDDSa6Bi qF99veXpnHHDxV8wx7CX727I4jjyEfY2quhUf+h/y75q32hR/G+t5FNL1 RQkmja88v42egh52UVqBEKCbnIsZ/Tt2+19kIjS7INN15xUmXaEsUzBvU s=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0B2BQDP2xJV/40NJK1cgkNDUloEywICgVdMAQEBAQEBfYQUAQEBBC1eAQgRAwEBASEBBjkUBwEBBQMCBBMbiBSfXKlvAQEBAQEBAQEBAQEBAQEBAQEBAQEYiyGEZRcBhC0FjkKCDolvgRuDMIwag0ciggIcgVBvgUR/AQEB
X-IronPort-AV: E=Sophos;i="5.11,465,1422921600";  d="scan'208,217";a="135313714"
Received: from alln-core-8.cisco.com ([173.36.13.141]) by alln-iport-7.cisco.com with ESMTP; 25 Mar 2015 16:05:08 +0000
Received: from xhc-rcd-x12.cisco.com (xhc-rcd-x12.cisco.com [173.37.183.86]) by alln-core-8.cisco.com (8.14.5/8.14.5) with ESMTP id t2PG581i021989 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Wed, 25 Mar 2015 16:05:08 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.15]) by xhc-rcd-x12.cisco.com ([173.37.183.86]) with mapi id 14.03.0195.001; Wed, 25 Mar 2015 11:05:08 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: draft-sajassi-bess-evpn-etree 
Thread-Index: AdBcUCnZNgMRxjitRJmKKmXMB3yB9gAhSLwAAARqDOD//+dNAP//2gLAgAA0pwD/+0kH4IAZHgWA
Date: Wed, 25 Mar 2015 16:05:08 +0000
Message-ID: <D13846A5.12E02A%sajassi@cisco.com>
In-Reply-To: <BN3PR0501MB11080BFD5C13B895763B23E4A8050@BN3PR0501MB1108.namprd05.prod.outlook.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.24.162.66]
Content-Type: multipart/alternative; boundary="_000_D13846A512E02Asajassiciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/B3TFDW-YKa5uVfWQu_OEscm2szQ>
Subject: [bess] FW: draft-sajassi-bess-evpn-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:07:00 -0000

--_000_D13846A512E02Asajassiciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable



From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Sunday, March 15, 2015 at 9:43 AM
To: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "draft-sa=
jassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@t=
ools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree

Hi Ali ,

How would the Etree service would work for Single Active Ethernet Segments =
.  Please note that AD route per ESI for single Active segments have ESI la=
bel encoded as zero as defined in RFC 7432 .

IMO  the ESI label is useful for loop prevention even in case of S/A MH seg=
ments for ELAN service too, not sure why it was decided to encode this as z=
ero in RFC 7432 .

Thanks
Nitin



From: Ali Sajassi (sajassi) [mailto:sajassi@cisco.com]
Sent: Thursday, March 12, 2015 11:31 AM
To: Nitin Singh; draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>
Cc: Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake
Subject: Re: draft-sajassi-bess-evpn-etree


Exactly!

Cheers,
Ali

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Thursday, March 12, 2015 at 11:02 AM
To: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "draft-sa=
jassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@t=
ools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree

Hi Ali ,

Thanks for your response,  it seems like that the proposal for SH segments =
 is to advertise the 2  AD routes per ESI with the ESI value encoded as 0 ,=
 one for SH segments in  Leaf role and other one the SH segments in Root mo=
de .

Regards
Nitin

From: Ali Sajassi (sajassi) [mailto:sajassi@cisco.com]
Sent: Thursday, March 12, 2015 10:39 AM
To: Nitin Singh; draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>
Cc: Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake
Subject: Re: draft-sajassi-bess-evpn-etree


Hi Nitin,


From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Thursday, March 12, 2015 at 10:25 AM
To: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "draft-sa=
jassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@t=
ools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree

Hi Ali ,

Please note that IM routes are advertised at granularity of Ethernet Tag [ =
IOW , there is a common IM route for ALL Ethernet Segments ].

Hence I agree that there is doubling of IM routes per Ethernet Tag per EVI =
 but this does not have the multiplication factor of number of ESIs. Essent=
ially we will have two multicast groups one for the leafs other one for the=
 roots .

I said the number of IM routes are doubled (e.g., 100% overhead) which is t=
he case.

The issue that I see is that the use ESI label in Etree service is slightly=
 different than the one described for ELAN service in RFC 7432 .

In ELAN service , ESI label is only used between the set of MH PEs connecte=
d to the ESI , but with the approach described in ETree even the PEs not co=
nnected to the ESI have to create forwarding state/SH procedures using the =
ESI label .

That=92s not correct =96 i.e., ESI label is NOT just used among MH PEs.  Et=
her=96AD per ES route along with ESI label is advertised to all PEs partici=
pating in the VPNs. There is a forwarding state that is created for ESI lab=
el for P2MP LSP anyway (but not for ingress replication).

Also there can be single homed AC in root/leaf roles , there will be no AD =
route per ESI routes for these single homed AC hence we cannot use the SH m=
echanisms  based on ESI label .

That=92s why we only need to advertise two such labels per PE (one for root=
 and another for leaf)!

CHeers,
Ali

Thanks
Nitin





From: Ali Sajassi (sajassi) [mailto:sajassi@cisco.com]
Sent: Thursday, March 12, 2015 10:01 AM
To: Nitin Singh; draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>
Cc: Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake; Ali Sajassi=
 (sajassi)
Subject: Re: draft-sajassi-bess-evpn-etree


There are bunch of disadvantages for doing it this way:

1.      That=92s a lot of advertisements if we do it with IM route - i.e., =
the BGP messages for IM gets increased by 100% !! A given ES can support hu=
ndreds or thousands of IM routes !!
2.      In data-plane we end up increasing the size of LFIB
3.      There are now to different approaches for single-home versus multi-=
homed sites

Cheers,
Ali

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Wednesday, March 11, 2015 at 5:14 PM
To: "draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess=
-evpn-etree@tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<=
mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: draft-sajassi-bess-evpn-etree
Resent-From: <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Resent-To: <draft-sajassi-bess-evpn-etree@ietf.org<mailto:draft-sajassi-bes=
s-evpn-etree@ietf.org>>
Resent-Date: Thursday, March 12, 2015 at 4:50 AM

Hi ,

This is regarding the section 3.2 BUM Traffic .   From my reading of the dr=
aft it seems that the assumption here is that BUM traffic will  be send wit=
h the ESI label  even to the PEs which are not connected to the ESI.

An alternative solution could be to advertise two  IM routes per Ethernet T=
ag

1) One for  the AC in leaf mode
2) Second for the AC in Root mode

The advantage of the above solution is that we don=92t need the ESI label i=
n label stack for the packet send towards the set of PEs not connected to t=
he ESI .

The use of ESI label could be restricted  for forwarding the BUM traffic be=
tween MH PEs connected to ESI operating in all-active mode,  as per the SH =
procedures  described in the RFC 7432 .

Thanks
Nitin





--_000_D13846A512E02Asajassiciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <713C46E652979E4890D521E129BCC3EB@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div><br>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Nitin Singh &lt;<a href=3D"ma=
ilto:nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Sunday, March 15, 2015 at 9:4=
3 AM<br>
<span style=3D"font-weight:bold">To: </span>Cisco Employee &lt;<a href=3D"m=
ailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;, &quot;<a href=3D"mailto=
:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etre=
e@tools.ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-et=
ree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<br=
>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>RE: draft-sajassi-bess-evp=
n-etree <br>
</div>
<div><br>
</div>
<div xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micro=
soft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" x=
mlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:/=
/www.w3.org/TR/REC-html40">
<meta name=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:"Calibri Light";
	panose-1:2 15 3 2 2 2 4 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:"Lucida Grande";
	panose-1:0 0 0 0 0 0 0 0 0 0;}
/* 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.emailquote, li.emailquote, div.emailquote
	{mso-style-name:emailquote;
	mso-margin-top-alt:auto;
	margin-right:0in;
	mso-margin-bottom-alt:auto;
	margin-left:1.0pt;
	font-size:12.0pt;
	font-family:"Times New Roman",serif;}
span.EmailStyle18
	{mso-style-type:personal;
	font-family:"Calibri",sans-serif;
	color:#1F497D;}
span.EmailStyle19
	{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;}
/* List Definitions */
@list l0
	{mso-list-id:1949703470;
	mso-list-template-ids:1774458198;}
@list l0:level1
	{mso-level-tab-stop:.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level2
	{mso-level-tab-stop:1.0in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level3
	{mso-level-tab-stop:1.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level4
	{mso-level-tab-stop:2.0in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level5
	{mso-level-tab-stop:2.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level6
	{mso-level-tab-stop:3.0in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level7
	{mso-level-tab-stop:3.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level8
	{mso-level-tab-stop:4.0in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level9
	{mso-level-tab-stop:4.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
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]-->
<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">Hi Ali ,
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">How would the Etree service would w=
ork for Single Active Ethernet Segments .&nbsp; Please note that AD route p=
er ESI for single Active segments have ESI
 label encoded as zero as defined in RFC 7432 . <o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">IMO &nbsp;the ESI label is useful f=
or loop prevention even in case of S/A MH segments for ELAN service too, no=
t sure why it was decided to encode this
 as zero in RFC 7432 . <o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">Thanks
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">Nitin
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<div>
<div style=3D"border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><b><span style=3D"font-si=
ze: 11pt; font-family: Calibri, sans-serif;">From:</span></b><span style=3D=
"font-size: 11pt; font-family: Calibri, sans-serif;"> Ali Sajassi (sajassi)=
 [<a href=3D"mailto:sajassi@cisco.com">mailto:sajassi@cisco.com</a>]
<br>
<b>Sent:</b> Thursday, March 12, 2015 11:31 AM<br>
<b>To:</b> Nitin Singh; <a href=3D"mailto:draft-sajassi-bess-evpn-etree@too=
ls.ietf.org">
draft-sajassi-bess-evpn-etree@tools.ietf.org</a><br>
<b>Cc:</b> Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake<br>
<b>Subject:</b> Re: draft-sajassi-bess-evpn-etree <o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><o:p>&nbsp;</o:p></p>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;">Exactly!&nbsp;<o:=
p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;">Cheers,<o:p></o:p=
></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;">Ali<o:p></o:p></s=
pan></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></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"margin-left:.5in"><b><span style=3D"font-si=
ze: 11pt; font-family: 'Lucida Grande', serif; color: black;">From:
</span></b><span style=3D"font-size: 11pt; font-family: 'Lucida Grande', se=
rif; color: black;">Nitin Singh &lt;<a href=3D"mailto:nisingh@juniper.net">=
nisingh@juniper.net</a>&gt;<br>
<b>Date: </b>Thursday, March 12, 2015 at 11:02 AM<br>
<b>To: </b>Cisco Employee &lt;<a href=3D"mailto:sajassi@cisco.com">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@to=
ols.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&quot; &lt;<a=
 href=3D"mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi=
-bess-evpn-etree@tools.ietf.org</a>&gt;<br>
<b>Cc: </b>Hassan Hosseini &lt;<a href=3D"mailto:hhosseini@juniper.net">hho=
sseini@juniper.net</a>&gt;, Ryan Bickhart &lt;<a href=3D"mailto:rbickhart@j=
uniper.net">rbickhart@juniper.net</a>&gt;, Manoj Sharma &lt;<a href=3D"mail=
to:manojs@juniper.net">manojs@juniper.net</a>&gt;, John
 E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net</a>&g=
t;<br>
<b>Subject: </b>RE: draft-sajassi-bess-evpn-etree <o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">Hi Ali ,
</span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">Thanks f=
or your response,&nbsp; it seems like that the proposal for SH segments &nb=
sp;is to advertise the 2 &nbsp;AD routes per ESI with
 the ESI value encoded as 0 , one for SH segments in &nbsp;Leaf role and ot=
her one the SH segments in Root mode .
</span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">Regards<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">Nitin
</span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div style=3D"border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><b><span style=3D"font-s=
ize: 11pt; font-family: Calibri, sans-serif; color: black;">From:</span></b=
><span style=3D"font-size: 11pt; font-family: Calibri, sans-serif; color: b=
lack;"> Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com">mailto:=
sajassi@cisco.com</a>]
<br>
<b>Sent:</b> Thursday, March 12, 2015 10:39 AM<br>
<b>To:</b> Nitin Singh; <a href=3D"mailto:draft-sajassi-bess-evpn-etree@too=
ls.ietf.org">
draft-sajassi-bess-evpn-etree@tools.ietf.org</a><br>
<b>Cc:</b> Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake<br>
<b>Subject:</b> Re: draft-sajassi-bess-evpn-etree </span><span style=3D"col=
or:black"><o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:bla=
ck">&nbsp;<o:p></o:p></span></p>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: red;">Hi Nitin,</span><s=
pan style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></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"margin-left:1.0in"><b><span style=3D"font-s=
ize: 11pt; font-family: 'Lucida Grande', serif; color: black;">From:
</span></b><span style=3D"font-size: 11pt; font-family: 'Lucida Grande', se=
rif; color: black;">Nitin Singh &lt;<a href=3D"mailto:nisingh@juniper.net">=
nisingh@juniper.net</a>&gt;<br>
<b>Date: </b>Thursday, March 12, 2015 at 10:25 AM<br>
<b>To: </b>Cisco Employee &lt;<a href=3D"mailto:sajassi@cisco.com">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@to=
ols.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&quot; &lt;<a=
 href=3D"mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi=
-bess-evpn-etree@tools.ietf.org</a>&gt;<br>
<b>Cc: </b>Hassan Hosseini &lt;<a href=3D"mailto:hhosseini@juniper.net">hho=
sseini@juniper.net</a>&gt;, Ryan Bickhart &lt;<a href=3D"mailto:rbickhart@j=
uniper.net">rbickhart@juniper.net</a>&gt;, Manoj Sharma &lt;<a href=3D"mail=
to:manojs@juniper.net">manojs@juniper.net</a>&gt;, John
 E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net</a>&g=
t;<br>
<b>Subject: </b>RE: draft-sajassi-bess-evpn-etree </span><span style=3D"col=
or:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Hi Ali ,
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Please note that I=
M routes are advertised at granularity of Ethernet Tag [ IOW , there is a c=
ommon IM route for ALL Ethernet Segments
 ]. </span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Hence I agree that=
 there is doubling of IM routes per Ethernet Tag per EVI&nbsp; but this doe=
s not have the multiplication factor of number
 of ESIs. Essentially we will have two multicast groups one for the leafs o=
ther one for the roots .</span><span style=3D"color:black"><o:p></o:p></spa=
n></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: red;">I said the number =
of IM routes are doubled (e.g., 100% overhead) which is the case.</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</s=
pan><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">The issue that I s=
ee is that the use ESI label in Etree service is slightly different than th=
e one described for ELAN service in RFC
 7432 .</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</s=
pan><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">In ELAN service , =
ESI label is only used between the set of MH PEs connected to the ESI , but=
 with the approach described in ETree
 even the PEs not connected to the ESI have to create forwarding state/SH p=
rocedures using the ESI label .</span><span style=3D"color:black"><o:p></o:=
p></span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: Calibri, sans-serif; color: red;">That=92s not correct&nbsp;=96 i.e., E=
SI label is NOT just used among MH PEs. &nbsp;Ether=96AD per ES route along=
 with ESI label is advertised to all&nbsp;PEs participating
 in the VPNs. There is a forwarding state that is created for ESI label for=
 P2MP LSP anyway (but not for ingress replication).</span><span style=3D"co=
lor:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</s=
pan><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Also there can be =
single homed AC in root/leaf roles , there will be no AD route per ESI rout=
es for these single homed AC hence we
 cannot use the SH mechanisms&nbsp; based on ESI label .</span><span style=
=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:bla=
ck">&nbsp;<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:red=
">That=92s why we only need to advertise two such labels per PE (one for ro=
ot and another for leaf)!</span><span style=3D"color:black"><o:p></o:p></sp=
an></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:bla=
ck">&nbsp;<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:red=
">CHeers,</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:red=
">Ali</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</s=
pan><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Thanks
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Nitin
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><b><span style=3D"font-s=
ize: 11pt; font-family: Calibri, sans-serif; color: black;">From:</span></b=
><span style=3D"font-size: 11pt; font-family: Calibri, sans-serif; color: b=
lack;"> Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com">mailto:=
sajassi@cisco.com</a>]
<br>
<b>Sent:</b> Thursday, March 12, 2015 10:01 AM<br>
<b>To:</b> Nitin Singh; <a href=3D"mailto:draft-sajassi-bess-evpn-etree@too=
ls.ietf.org">
draft-sajassi-bess-evpn-etree@tools.ietf.org</a><br>
<b>Cc:</b> Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake; Ali =
Sajassi (sajassi)<br>
<b>Subject:</b> Re: draft-sajassi-bess-evpn-etree </span><span style=3D"col=
or:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">There are bunch =
of disadvantages for doing it this way:</span><span style=3D"color:black"><=
o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:5.0pt;margin-right:0in;m=
argin-bottom:5.0pt;margin-left:1.0in;text-indent:-.25in;mso-list:l0 level1 =
lfo2">
<!--[if !supportLists]--><span style=3D"color:black"><span style=3D"mso-lis=
t:Ignore">1.<span style=3D"font-style: normal; font-variant: normal; font-w=
eight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New=
 Roman';">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span></span><!--[endif]--><span style=3D"font-size: 10.5pt; font-f=
amily: Calibri, sans-serif; color: black;">That=92s a lot of advertisements=
 if we do it with IM route - i.e., the BGP messages for IM gets increased b=
y 100% !! A given ES can support hundreds
 or thousands of IM routes !!</span><span style=3D"color:black"><o:p></o:p>=
</span></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:5.0pt;margin-right:0in;m=
argin-bottom:5.0pt;margin-left:1.0in;text-indent:-.25in;mso-list:l0 level1 =
lfo2">
<!--[if !supportLists]--><span style=3D"color:black"><span style=3D"mso-lis=
t:Ignore">2.<span style=3D"font-style: normal; font-variant: normal; font-w=
eight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New=
 Roman';">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span></span><!--[endif]--><span style=3D"font-size: 10.5pt; font-f=
amily: Calibri, sans-serif; color: black;">In data-plane we end up increasi=
ng the size of LFIB</span><span style=3D"color:black"><o:p></o:p></span></p=
>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:5.0pt;margin-right:0in;m=
argin-bottom:5.0pt;margin-left:1.0in;text-indent:-.25in;mso-list:l0 level1 =
lfo2">
<!--[if !supportLists]--><span style=3D"color:black"><span style=3D"mso-lis=
t:Ignore">3.<span style=3D"font-style: normal; font-variant: normal; font-w=
eight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New=
 Roman';">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span></span><!--[endif]--><span style=3D"font-size: 10pt; font-fam=
ily: Calibri, sans-serif; color: black;">There are now to different approac=
hes for single-home versus multi-homed sites</span><span style=3D"color:bla=
ck"><o:p></o:p></span></p>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">Cheers,</span><s=
pan style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">Ali</span><span =
style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><b><span style=3D"font-s=
ize: 11pt; font-family: 'Lucida Grande', serif; color: black;">From:
</span></b><span style=3D"font-size: 11pt; font-family: 'Lucida Grande', se=
rif; color: black;">Nitin Singh &lt;<a href=3D"mailto:nisingh@juniper.net">=
<span style=3D"color:#0563C1">nisingh@juniper.net</span></a>&gt;<br>
<b>Date: </b>Wednesday, March 11, 2015 at 5:14 PM<br>
<b>To: </b>&quot;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@tools.ietf=
.org"><span style=3D"color:#0563C1">draft-sajassi-bess-evpn-etree@tools.iet=
f.org</span></a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@=
tools.ietf.org"><span style=3D"color:#0563C1">draft-sajassi-bess-evpn-etree=
@tools.ietf.org</span></a>&gt;<br>
<b>Cc: </b>Hassan Hosseini &lt;<a href=3D"mailto:hhosseini@juniper.net"><sp=
an style=3D"color:#0563C1">hhosseini@juniper.net</span></a>&gt;, Ryan Bickh=
art &lt;<a href=3D"mailto:rbickhart@juniper.net"><span style=3D"color:#0563=
C1">rbickhart@juniper.net</span></a>&gt;, Manoj Sharma
 &lt;<a href=3D"mailto:manojs@juniper.net"><span style=3D"color:#0563C1">ma=
nojs@juniper.net</span></a>&gt;, John E Drake &lt;<a href=3D"mailto:jdrake@=
juniper.net"><span style=3D"color:#0563C1">jdrake@juniper.net</span></a>&gt=
;<br>
<b>Subject: </b>draft-sajassi-bess-evpn-etree <br>
<b>Resent-From: </b>&lt;<a href=3D"mailto:nisingh@juniper.net"><span style=
=3D"color:#0563C1">nisingh@juniper.net</span></a>&gt;<br>
<b>Resent-To: </b>&lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@ietf.=
org"><span style=3D"color:#0563C1">draft-sajassi-bess-evpn-etree@ietf.org</=
span></a>&gt;<br>
<b>Resent-Date: </b>Thursday, March 12, 2015 at 4:50 AM</span><span style=
=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">Hi ,
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">This is regarding =
the section 3.2 BUM Traffic .&nbsp;&nbsp; From my reading of the draft it s=
eems that the assumption here is that BUM traffic
 will&nbsp; be send with the ESI label&nbsp; even to the PEs which are not =
connected to the ESI. &nbsp;</span><span style=3D"color:black"><o:p></o:p><=
/span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">An alternative sol=
ution could be to advertise two &nbsp;IM routes per Ethernet Tag
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">1) One for &nbsp;t=
he AC in leaf mode
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">2) Second for the =
AC in Root mode
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">The advantage of t=
he above solution is that we don=92t need the ESI label in label stack for =
the packet send towards the set of PEs not
 connected to the ESI . </span><span style=3D"color:black"><o:p></o:p></spa=
n></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">The use of ESI lab=
el could be restricted &nbsp;for forwarding the BUM traffic between MH PEs =
connected to ESI operating in all-active mode,
 &nbsp;as per the SH procedures &nbsp;described in the RFC 7432 . </span><s=
pan style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">Thanks
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">Nitin
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</span>
</body>
</html>

--_000_D13846A512E02Asajassiciscocom_--


From nobody Wed Mar 25 09:07:39 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 664421A889D for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:07:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GXn1kUz2aBI5 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:07:32 -0700 (PDT)
Received: from rcdn-iport-6.cisco.com (rcdn-iport-6.cisco.com [173.37.86.77]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8FCC21A8966 for <bess@ietf.org>; Wed, 25 Mar 2015 09:05:53 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=47893; q=dns/txt; s=iport; t=1427299553; x=1428509153; h=from:to:subject:date:message-id:in-reply-to:mime-version; bh=zeoi6p6cF0YJ6JCQNsKcnXb36xeZ9FY8DUcpRF8Z8rA=; b=IJZuhs0lkAoVoateceqoPXBdloZplScTmEF4riwX67VZQW+oAK3On8Ei jS66mLn6eCN9FT6329+KpC3kyUDZJqaFDoSJrNOgGnvsYfWa6TfwYw/i1 QkkVtMUbQRexNMJ9en5eLXN1oNxWv5qyLpNGHzJI9v6ZI9O625rWWvuku c=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0B2BQAM3BJV/4QNJK1cgkNDUloEywICgVdMAQEBAQEBfYQUAQEBBC1eAQgRAwEBASEBBjkUBwEBBQMCBBMUB4gUn2CpbwEBAQEBAQEBAQEBAQEBAQEBAQEBGIshhGUXAYQtBY5Cgg6Jb4EbgzCCWYlBg0ciggIcgVBvgUR/AQEB
X-IronPort-AV: E=Sophos;i="5.11,465,1422921600";  d="scan'208,217";a="406588699"
Received: from alln-core-10.cisco.com ([173.36.13.132]) by rcdn-iport-6.cisco.com with ESMTP; 25 Mar 2015 16:05:52 +0000
Received: from xhc-aln-x09.cisco.com (xhc-aln-x09.cisco.com [173.36.12.83]) by alln-core-10.cisco.com (8.14.5/8.14.5) with ESMTP id t2PG5qur001852 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Wed, 25 Mar 2015 16:05:52 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.15]) by xhc-aln-x09.cisco.com ([173.36.12.83]) with mapi id 14.03.0195.001; Wed, 25 Mar 2015 11:05:52 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: draft-sajassi-bess-evpn-etree 
Thread-Index: AdBcUCnZNgMRxjitRJmKKmXMB3yB9gAhSLwAAARqDOD//+dNAP//2gLAgAA0pwD/+0kH4IALsDkAgA1uAgA=
Date: Wed, 25 Mar 2015 16:05:51 +0000
Message-ID: <D13846F7.12E037%sajassi@cisco.com>
In-Reply-To: <D12CFFEA.12B3E4%sajassi@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.24.162.66]
Content-Type: multipart/alternative; boundary="_000_D13846F712E037sajassiciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/HreCLj3MyV0owzYYFD0ooOlATRo>
Subject: [bess] FW: draft-sajassi-bess-evpn-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:07:38 -0000

--_000_D13846F712E037sajassiciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable



From: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>
Date: Monday, March 16, 2015 at 10:00 PM
To: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>, "draft-s=
ajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@=
tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft=
-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>, Cisco Employee <sajassi@cisco.co=
m<mailto:sajassi@cisco.com>>
Subject: Re: draft-sajassi-bess-evpn-etree


Hi Nitin,

Where did you read in RFC7432 that says ESI label is set to zero for single=
-active? Section 8.2.1 says:

"  If Single-Active redundancy mode is desired, then the "Single-Active"

   bit in the flags of the ESI Label extended community MUST be set to 1
   and the ESI label SHOULD be set to a valid MPLS label value.=94


And section 8.3 says:


"When a set of PEs are operating in Single-Active redundancy mode, the

   use of this split-horizon filtering mechanism is highly recommended
   because it prevents transient loops at the time of failure or
   recovery that would impact the Ethernet segment"


Cheers,

Ali

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Sunday, March 15, 2015 at 9:43 AM
To: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "draft-sa=
jassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@t=
ools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree

Hi Ali ,

How would the Etree service would work for Single Active Ethernet Segments =
.  Please note that AD route per ESI for single Active segments have ESI la=
bel encoded as zero as defined in RFC 7432 .

IMO  the ESI label is useful for loop prevention even in case of S/A MH seg=
ments for ELAN service too, not sure why it was decided to encode this as z=
ero in RFC 7432 .

Thanks
Nitin



From: Ali Sajassi (sajassi) [mailto:sajassi@cisco.com]
Sent: Thursday, March 12, 2015 11:31 AM
To: Nitin Singh; draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>
Cc: Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake
Subject: Re: draft-sajassi-bess-evpn-etree


Exactly!

Cheers,
Ali

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Thursday, March 12, 2015 at 11:02 AM
To: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "draft-sa=
jassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@t=
ools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree

Hi Ali ,

Thanks for your response,  it seems like that the proposal for SH segments =
 is to advertise the 2  AD routes per ESI with the ESI value encoded as 0 ,=
 one for SH segments in  Leaf role and other one the SH segments in Root mo=
de .

Regards
Nitin

From: Ali Sajassi (sajassi) [mailto:sajassi@cisco.com]
Sent: Thursday, March 12, 2015 10:39 AM
To: Nitin Singh; draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>
Cc: Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake
Subject: Re: draft-sajassi-bess-evpn-etree


Hi Nitin,


From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Thursday, March 12, 2015 at 10:25 AM
To: Cisco Employee <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "draft-sa=
jassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess-evpn-etree@t=
ools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: RE: draft-sajassi-bess-evpn-etree

Hi Ali ,

Please note that IM routes are advertised at granularity of Ethernet Tag [ =
IOW , there is a common IM route for ALL Ethernet Segments ].

Hence I agree that there is doubling of IM routes per Ethernet Tag per EVI =
 but this does not have the multiplication factor of number of ESIs. Essent=
ially we will have two multicast groups one for the leafs other one for the=
 roots .

I said the number of IM routes are doubled (e.g., 100% overhead) which is t=
he case.

The issue that I see is that the use ESI label in Etree service is slightly=
 different than the one described for ELAN service in RFC 7432 .

In ELAN service , ESI label is only used between the set of MH PEs connecte=
d to the ESI , but with the approach described in ETree even the PEs not co=
nnected to the ESI have to create forwarding state/SH procedures using the =
ESI label .

That=92s not correct =96 i.e., ESI label is NOT just used among MH PEs.  Et=
her=96AD per ES route along with ESI label is advertised to all PEs partici=
pating in the VPNs. There is a forwarding state that is created for ESI lab=
el for P2MP LSP anyway (but not for ingress replication).

Also there can be single homed AC in root/leaf roles , there will be no AD =
route per ESI routes for these single homed AC hence we cannot use the SH m=
echanisms  based on ESI label .

That=92s why we only need to advertise two such labels per PE (one for root=
 and another for leaf)!

CHeers,
Ali

Thanks
Nitin





From: Ali Sajassi (sajassi) [mailto:sajassi@cisco.com]
Sent: Thursday, March 12, 2015 10:01 AM
To: Nitin Singh; draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-=
sajassi-bess-evpn-etree@tools.ietf.org>
Cc: Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake; Ali Sajassi=
 (sajassi)
Subject: Re: draft-sajassi-bess-evpn-etree


There are bunch of disadvantages for doing it this way:

1.      That=92s a lot of advertisements if we do it with IM route - i.e., =
the BGP messages for IM gets increased by 100% !! A given ES can support hu=
ndreds or thousands of IM routes !!
2.      In data-plane we end up increasing the size of LFIB
3.      There are now to different approaches for single-home versus multi-=
homed sites

Cheers,
Ali

From: Nitin Singh <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Date: Wednesday, March 11, 2015 at 5:14 PM
To: "draft-sajassi-bess-evpn-etree@tools.ietf.org<mailto:draft-sajassi-bess=
-evpn-etree@tools.ietf.org>" <draft-sajassi-bess-evpn-etree@tools.ietf.org<=
mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org>>
Cc: Hassan Hosseini <hhosseini@juniper.net<mailto:hhosseini@juniper.net>>, =
Ryan Bickhart <rbickhart@juniper.net<mailto:rbickhart@juniper.net>>, Manoj =
Sharma <manojs@juniper.net<mailto:manojs@juniper.net>>, John E Drake <jdrak=
e@juniper.net<mailto:jdrake@juniper.net>>
Subject: draft-sajassi-bess-evpn-etree
Resent-From: <nisingh@juniper.net<mailto:nisingh@juniper.net>>
Resent-To: <draft-sajassi-bess-evpn-etree@ietf.org<mailto:draft-sajassi-bes=
s-evpn-etree@ietf.org>>
Resent-Date: Thursday, March 12, 2015 at 4:50 AM

Hi ,

This is regarding the section 3.2 BUM Traffic .   From my reading of the dr=
aft it seems that the assumption here is that BUM traffic will  be send wit=
h the ESI label  even to the PEs which are not connected to the ESI.

An alternative solution could be to advertise two  IM routes per Ethernet T=
ag

1) One for  the AC in leaf mode
2) Second for the AC in Root mode

The advantage of the above solution is that we don=92t need the ESI label i=
n label stack for the packet send towards the set of PEs not connected to t=
he ESI .

The use of ESI label could be restricted  for forwarding the BUM traffic be=
tween MH PEs connected to ESI operating in all-active mode,  as per the SH =
procedures  described in the RFC 7432 .

Thanks
Nitin





--_000_D13846F712E037sajassiciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <97FA42E8E8F1F6458E31A916A0473891@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div><br>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Cisco Employee &lt;<a href=3D=
"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Monday, March 16, 2015 at 10:=
00 PM<br>
<span style=3D"font-weight:bold">To: </span>Nitin Singh &lt;<a href=3D"mail=
to:nisingh@juniper.net">nisingh@juniper.net</a>&gt;, &quot;<a href=3D"mailt=
o:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etr=
ee@tools.ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-e=
tree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<b=
r>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;, Cisco Employee &lt;<a href=3D"mailto:sajassi@cisco.com">sajassi@ci=
sco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>Re: draft-sajassi-bess-evp=
n-etree <br>
</div>
<div><br>
</div>
<div>
<div style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line=
-break: after-white-space; color: rgb(0, 0, 0);">
<div style=3D"font-size: 14px; font-family: Calibri, sans-serif;"><br>
</div>
<div style=3D"font-size: 14px; font-family: Calibri, sans-serif;">Hi Nitin,=
</div>
<div style=3D"font-size: 14px; font-family: Calibri, sans-serif;"><br>
</div>
<div style=3D"font-size: 14px; font-family: Calibri, sans-serif;">Where did=
 you read in RFC7432 that says ESI label is set to zero for single-active? =
Section 8.2.1 says:</div>
<div style=3D"font-size: 14px; font-family: Calibri, sans-serif;"><br>
</div>
<div><font face=3D"Courier">&quot; &nbsp;<span style=3D"line-height: 1.2em;=
">If Single-Active redundancy mode is desired, then the &quot;Single-Active=
&quot;</span></font></div>
<pre style=3D"line-height: 1.2em; margin-top: 0px; margin-bottom: 0px;"><fo=
nt face=3D"Courier">   bit in the flags of the ESI Label extended community=
 MUST be set to 1
   and the ESI label SHOULD be set to a valid MPLS label value.=94</font></=
pre>
<pre style=3D"line-height: 1.2em; margin-top: 0px; margin-bottom: 0px;"><br=
></pre>
<pre style=3D"line-height: 1.2em; margin-top: 0px; margin-bottom: 0px;">And=
 section 8.3 says:</pre>
<pre style=3D"line-height: 1.2em; margin-top: 0px; margin-bottom: 0px;"><br=
></pre>
<pre style=3D"line-height: 1.2em; margin-top: 0px; margin-bottom: 0px;"><sp=
an style=3D"font-size: 14px;">&quot;<span style=3D"line-height: 1.2em;">Whe=
n a set of PEs are operating in Single-Active redundancy mode, the</span></=
span></pre>
<pre style=3D"line-height: 1.2em; margin-top: 0px; margin-bottom: 0px;"><sp=
an style=3D"font-size: 14px;">   use of this split-horizon filtering mechan=
ism is highly recommended
   because it prevents transient loops at the time of failure or
   recovery that would impact the Ethernet segment&quot;</span></pre>
<pre style=3D"font-size: 13px; font-family: Calibri, sans-serif; line-heigh=
t: 1.2em; margin-top: 0px; margin-bottom: 0px;"><br></pre>
<pre style=3D"font-family: Calibri, sans-serif; line-height: 1.2em; margin-=
top: 0px; margin-bottom: 0px;">Cheers,</pre>
<pre style=3D"font-family: Calibri, sans-serif; line-height: 1.2em; margin-=
top: 0px; margin-bottom: 0px;">Ali</pre>
<div style=3D"font-size: 14px; font-family: Calibri, sans-serif;"><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION" style=3D"font-size: 14px; font-family: Ca=
libri, sans-serif;">
<div style=3D"font-family:Lucida Grande; font-size:11pt; text-align:left; c=
olor:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-B=
OTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt =
solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Nitin Singh &lt;<a href=3D"ma=
ilto:nisingh@juniper.net">nisingh@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Sunday, March 15, 2015 at 9:4=
3 AM<br>
<span style=3D"font-weight:bold">To: </span>Cisco Employee &lt;<a href=3D"m=
ailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;, &quot;<a href=3D"mailto=
:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi-bess-evpn-etre=
e@tools.ietf.org</a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-et=
ree@tools.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&gt;<br=
>
<span style=3D"font-weight:bold">Cc: </span>Hassan Hosseini &lt;<a href=3D"=
mailto:hhosseini@juniper.net">hhosseini@juniper.net</a>&gt;, Ryan Bickhart =
&lt;<a href=3D"mailto:rbickhart@juniper.net">rbickhart@juniper.net</a>&gt;,=
 Manoj Sharma &lt;<a href=3D"mailto:manojs@juniper.net">manojs@juniper.net<=
/a>&gt;,
 John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net<=
/a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>RE: draft-sajassi-bess-evp=
n-etree <br>
</div>
<div><br>
</div>
<div xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micro=
soft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" x=
mlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:/=
/www.w3.org/TR/REC-html40">
<meta name=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:"Calibri Light";
	panose-1:2 15 3 2 2 2 4 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:"Lucida Grande";
	panose-1:0 0 0 0 0 0 0 0 0 0;}
/* 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.emailquote, li.emailquote, div.emailquote
	{mso-style-name:emailquote;
	mso-margin-top-alt:auto;
	margin-right:0in;
	mso-margin-bottom-alt:auto;
	margin-left:1.0pt;
	font-size:12.0pt;
	font-family:"Times New Roman",serif;}
span.EmailStyle18
	{mso-style-type:personal;
	font-family:"Calibri",sans-serif;
	color:#1F497D;}
span.EmailStyle19
	{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;}
/* List Definitions */
@list l0
	{mso-list-id:1949703470;
	mso-list-template-ids:1774458198;}
@list l0:level1
	{mso-level-tab-stop:.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level2
	{mso-level-tab-stop:1.0in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level3
	{mso-level-tab-stop:1.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level4
	{mso-level-tab-stop:2.0in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level5
	{mso-level-tab-stop:2.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level6
	{mso-level-tab-stop:3.0in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level7
	{mso-level-tab-stop:3.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level8
	{mso-level-tab-stop:4.0in;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level9
	{mso-level-tab-stop:4.5in;
	mso-level-number-position:left;
	text-indent:-.25in;}
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]-->
<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">Hi Ali ,
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">How would the Etree service would w=
ork for Single Active Ethernet Segments .&nbsp; Please note that AD route p=
er ESI for single Active segments have ESI
 label encoded as zero as defined in RFC 7432 . <o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">IMO &nbsp;the ESI label is useful f=
or loop prevention even in case of S/A MH segments for ELAN service too, no=
t sure why it was decided to encode this
 as zero in RFC 7432 . <o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">Thanks
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);">Nitin
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size: 11pt; font-family: Calibri=
, sans-serif; color: rgb(31, 73, 125);"><o:p>&nbsp;</o:p></span></p>
<div>
<div style=3D"border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><b><span style=3D"font-si=
ze: 11pt; font-family: Calibri, sans-serif;">From:</span></b><span style=3D=
"font-size: 11pt; font-family: Calibri, sans-serif;"> Ali Sajassi (sajassi)=
 [<a href=3D"mailto:sajassi@cisco.com">mailto:sajassi@cisco.com</a>]
<br>
<b>Sent:</b> Thursday, March 12, 2015 11:31 AM<br>
<b>To:</b> Nitin Singh; <a href=3D"mailto:draft-sajassi-bess-evpn-etree@too=
ls.ietf.org">
draft-sajassi-bess-evpn-etree@tools.ietf.org</a><br>
<b>Cc:</b> Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake<br>
<b>Subject:</b> Re: draft-sajassi-bess-evpn-etree <o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><o:p>&nbsp;</o:p></p>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;">Exactly!&nbsp;<o:=
p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;">Cheers,<o:p></o:p=
></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;">Ali<o:p></o:p></s=
pan></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></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"margin-left:.5in"><b><span style=3D"font-si=
ze: 11pt; font-family: 'Lucida Grande', serif; color: black;">From:
</span></b><span style=3D"font-size: 11pt; font-family: 'Lucida Grande', se=
rif; color: black;">Nitin Singh &lt;<a href=3D"mailto:nisingh@juniper.net">=
nisingh@juniper.net</a>&gt;<br>
<b>Date: </b>Thursday, March 12, 2015 at 11:02 AM<br>
<b>To: </b>Cisco Employee &lt;<a href=3D"mailto:sajassi@cisco.com">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@to=
ols.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&quot; &lt;<a=
 href=3D"mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi=
-bess-evpn-etree@tools.ietf.org</a>&gt;<br>
<b>Cc: </b>Hassan Hosseini &lt;<a href=3D"mailto:hhosseini@juniper.net">hho=
sseini@juniper.net</a>&gt;, Ryan Bickhart &lt;<a href=3D"mailto:rbickhart@j=
uniper.net">rbickhart@juniper.net</a>&gt;, Manoj Sharma &lt;<a href=3D"mail=
to:manojs@juniper.net">manojs@juniper.net</a>&gt;, John
 E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net</a>&g=
t;<br>
<b>Subject: </b>RE: draft-sajassi-bess-evpn-etree <o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 10.5pt; font-family: Calibri, sans-serif; color: black;"><o:p>&nbsp;</o:p>=
</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">Hi Ali ,
</span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">Thanks f=
or your response,&nbsp; it seems like that the proposal for SH segments &nb=
sp;is to advertise the 2 &nbsp;AD routes per ESI with
 the ESI value encoded as 0 , one for SH segments in &nbsp;Leaf role and ot=
her one the SH segments in Root mode .
</span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">Regards<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">Nitin
</span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal" style=3D"margin-left:.5in"><span style=3D"font-size:=
 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;</=
span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div style=3D"border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><b><span style=3D"font-s=
ize: 11pt; font-family: Calibri, sans-serif; color: black;">From:</span></b=
><span style=3D"font-size: 11pt; font-family: Calibri, sans-serif; color: b=
lack;"> Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com">mailto:=
sajassi@cisco.com</a>]
<br>
<b>Sent:</b> Thursday, March 12, 2015 10:39 AM<br>
<b>To:</b> Nitin Singh; <a href=3D"mailto:draft-sajassi-bess-evpn-etree@too=
ls.ietf.org">
draft-sajassi-bess-evpn-etree@tools.ietf.org</a><br>
<b>Cc:</b> Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake<br>
<b>Subject:</b> Re: draft-sajassi-bess-evpn-etree </span><span style=3D"col=
or:black"><o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:bla=
ck">&nbsp;<o:p></o:p></span></p>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: red;">Hi Nitin,</span><s=
pan style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></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"margin-left:1.0in"><b><span style=3D"font-s=
ize: 11pt; font-family: 'Lucida Grande', serif; color: black;">From:
</span></b><span style=3D"font-size: 11pt; font-family: 'Lucida Grande', se=
rif; color: black;">Nitin Singh &lt;<a href=3D"mailto:nisingh@juniper.net">=
nisingh@juniper.net</a>&gt;<br>
<b>Date: </b>Thursday, March 12, 2015 at 10:25 AM<br>
<b>To: </b>Cisco Employee &lt;<a href=3D"mailto:sajassi@cisco.com">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@to=
ols.ietf.org">draft-sajassi-bess-evpn-etree@tools.ietf.org</a>&quot; &lt;<a=
 href=3D"mailto:draft-sajassi-bess-evpn-etree@tools.ietf.org">draft-sajassi=
-bess-evpn-etree@tools.ietf.org</a>&gt;<br>
<b>Cc: </b>Hassan Hosseini &lt;<a href=3D"mailto:hhosseini@juniper.net">hho=
sseini@juniper.net</a>&gt;, Ryan Bickhart &lt;<a href=3D"mailto:rbickhart@j=
uniper.net">rbickhart@juniper.net</a>&gt;, Manoj Sharma &lt;<a href=3D"mail=
to:manojs@juniper.net">manojs@juniper.net</a>&gt;, John
 E Drake &lt;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net</a>&g=
t;<br>
<b>Subject: </b>RE: draft-sajassi-bess-evpn-etree </span><span style=3D"col=
or:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Hi Ali ,
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Please note that I=
M routes are advertised at granularity of Ethernet Tag [ IOW , there is a c=
ommon IM route for ALL Ethernet Segments
 ]. </span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Hence I agree that=
 there is doubling of IM routes per Ethernet Tag per EVI&nbsp; but this doe=
s not have the multiplication factor of number
 of ESIs. Essentially we will have two multicast groups one for the leafs o=
ther one for the roots .</span><span style=3D"color:black"><o:p></o:p></spa=
n></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: red;">I said the number =
of IM routes are doubled (e.g., 100% overhead) which is the case.</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</s=
pan><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">The issue that I s=
ee is that the use ESI label in Etree service is slightly different than th=
e one described for ELAN service in RFC
 7432 .</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</s=
pan><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">In ELAN service , =
ESI label is only used between the set of MH PEs connected to the ESI , but=
 with the approach described in ETree
 even the PEs not connected to the ESI have to create forwarding state/SH p=
rocedures using the ESI label .</span><span style=3D"color:black"><o:p></o:=
p></span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><sp=
an style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: Calibri, sans-serif; color: red;">That=92s not correct&nbsp;=96 i.e., E=
SI label is NOT just used among MH PEs. &nbsp;Ether=96AD per ES route along=
 with ESI label is advertised to all&nbsp;PEs participating
 in the VPNs. There is a forwarding state that is created for ESI label for=
 P2MP LSP anyway (but not for ingress replication).</span><span style=3D"co=
lor:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</s=
pan><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Also there can be =
single homed AC in root/leaf roles , there will be no AD route per ESI rout=
es for these single homed AC hence we
 cannot use the SH mechanisms&nbsp; based on ESI label .</span><span style=
=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:bla=
ck">&nbsp;<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:red=
">That=92s why we only need to advertise two such labels per PE (one for ro=
ot and another for leaf)!</span><span style=3D"color:black"><o:p></o:p></sp=
an></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:bla=
ck">&nbsp;<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:red=
">CHeers,</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"color:red=
">Ali</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(0, 32, 96);">&nbsp;</s=
pan><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Thanks
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-fami=
ly: 'Calibri Light', sans-serif; color: rgb(0, 32, 96);">Nitin
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);">&nbsp;<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><b><span style=3D"font-s=
ize: 11pt; font-family: Calibri, sans-serif; color: black;">From:</span></b=
><span style=3D"font-size: 11pt; font-family: Calibri, sans-serif; color: b=
lack;"> Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com">mailto:=
sajassi@cisco.com</a>]
<br>
<b>Sent:</b> Thursday, March 12, 2015 10:01 AM<br>
<b>To:</b> Nitin Singh; <a href=3D"mailto:draft-sajassi-bess-evpn-etree@too=
ls.ietf.org">
draft-sajassi-bess-evpn-etree@tools.ietf.org</a><br>
<b>Cc:</b> Hassan Hosseini; Ryan Bickhart; Manoj Sharma; John E Drake; Ali =
Sajassi (sajassi)<br>
<b>Subject:</b> Re: draft-sajassi-bess-evpn-etree </span><span style=3D"col=
or:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">There are bunch =
of disadvantages for doing it this way:</span><span style=3D"color:black"><=
o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:5.0pt;margin-right:0in;m=
argin-bottom:5.0pt;margin-left:1.0in;text-indent:-.25in;mso-list:l0 level1 =
lfo2">
<!--[if !supportLists]--><span style=3D"color:black"><span style=3D"mso-lis=
t:Ignore">1.<span style=3D"font-style: normal; font-variant: normal; font-w=
eight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New=
 Roman';">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span></span><!--[endif]--><span style=3D"font-size: 10.5pt; font-f=
amily: Calibri, sans-serif; color: black;">That=92s a lot of advertisements=
 if we do it with IM route - i.e., the BGP messages for IM gets increased b=
y 100% !! A given ES can support hundreds
 or thousands of IM routes !!</span><span style=3D"color:black"><o:p></o:p>=
</span></p>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:5.0pt;margin-right:0in;m=
argin-bottom:5.0pt;margin-left:1.0in;text-indent:-.25in;mso-list:l0 level1 =
lfo2">
<!--[if !supportLists]--><span style=3D"color:black"><span style=3D"mso-lis=
t:Ignore">2.<span style=3D"font-style: normal; font-variant: normal; font-w=
eight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New=
 Roman';">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span></span><!--[endif]--><span style=3D"font-size: 10.5pt; font-f=
amily: Calibri, sans-serif; color: black;">In data-plane we end up increasi=
ng the size of LFIB</span><span style=3D"color:black"><o:p></o:p></span></p=
>
<p class=3D"MsoNormal" style=3D"mso-margin-top-alt:5.0pt;margin-right:0in;m=
argin-bottom:5.0pt;margin-left:1.0in;text-indent:-.25in;mso-list:l0 level1 =
lfo2">
<!--[if !supportLists]--><span style=3D"color:black"><span style=3D"mso-lis=
t:Ignore">3.<span style=3D"font-style: normal; font-variant: normal; font-w=
eight: normal; font-size: 7pt; line-height: normal; font-family: 'Times New=
 Roman';">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
</span></span></span><!--[endif]--><span style=3D"font-size: 10pt; font-fam=
ily: Calibri, sans-serif; color: black;">There are now to different approac=
hes for single-home versus multi-homed sites</span><span style=3D"color:bla=
ck"><o:p></o:p></span></p>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">Cheers,</span><s=
pan style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 10.5pt; font-family: Calibri, sans-serif; color: black;">Ali</span><span =
style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><b><span style=3D"font-s=
ize: 11pt; font-family: 'Lucida Grande', serif; color: black;">From:
</span></b><span style=3D"font-size: 11pt; font-family: 'Lucida Grande', se=
rif; color: black;">Nitin Singh &lt;<a href=3D"mailto:nisingh@juniper.net">=
<span style=3D"color:#0563C1">nisingh@juniper.net</span></a>&gt;<br>
<b>Date: </b>Wednesday, March 11, 2015 at 5:14 PM<br>
<b>To: </b>&quot;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@tools.ietf=
.org"><span style=3D"color:#0563C1">draft-sajassi-bess-evpn-etree@tools.iet=
f.org</span></a>&quot; &lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@=
tools.ietf.org"><span style=3D"color:#0563C1">draft-sajassi-bess-evpn-etree=
@tools.ietf.org</span></a>&gt;<br>
<b>Cc: </b>Hassan Hosseini &lt;<a href=3D"mailto:hhosseini@juniper.net"><sp=
an style=3D"color:#0563C1">hhosseini@juniper.net</span></a>&gt;, Ryan Bickh=
art &lt;<a href=3D"mailto:rbickhart@juniper.net"><span style=3D"color:#0563=
C1">rbickhart@juniper.net</span></a>&gt;, Manoj Sharma
 &lt;<a href=3D"mailto:manojs@juniper.net"><span style=3D"color:#0563C1">ma=
nojs@juniper.net</span></a>&gt;, John E Drake &lt;<a href=3D"mailto:jdrake@=
juniper.net"><span style=3D"color:#0563C1">jdrake@juniper.net</span></a>&gt=
;<br>
<b>Subject: </b>draft-sajassi-bess-evpn-etree <br>
<b>Resent-From: </b>&lt;<a href=3D"mailto:nisingh@juniper.net"><span style=
=3D"color:#0563C1">nisingh@juniper.net</span></a>&gt;<br>
<b>Resent-To: </b>&lt;<a href=3D"mailto:draft-sajassi-bess-evpn-etree@ietf.=
org"><span style=3D"color:#0563C1">draft-sajassi-bess-evpn-etree@ietf.org</=
span></a>&gt;<br>
<b>Resent-Date: </b>Thursday, March 12, 2015 at 4:50 AM</span><span style=
=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">Hi ,
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">This is regarding =
the section 3.2 BUM Traffic .&nbsp;&nbsp; From my reading of the draft it s=
eems that the assumption here is that BUM traffic
 will&nbsp; be send with the ESI label&nbsp; even to the PEs which are not =
connected to the ESI. &nbsp;</span><span style=3D"color:black"><o:p></o:p><=
/span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">An alternative sol=
ution could be to advertise two &nbsp;IM routes per Ethernet Tag
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">1) One for &nbsp;t=
he AC in leaf mode
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">2) Second for the =
AC in Root mode
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">The advantage of t=
he above solution is that we don=92t need the ESI label in label stack for =
the packet send towards the set of PEs not
 connected to the ESI . </span><span style=3D"color:black"><o:p></o:p></spa=
n></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">The use of ESI lab=
el could be restricted &nbsp;for forwarding the BUM traffic between MH PEs =
connected to ESI operating in all-active mode,
 &nbsp;as per the SH procedures &nbsp;described in the RFC 7432 . </span><s=
pan style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">Thanks
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">Nitin
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-left:1.0in"><span style=3D"font-size=
: 11pt; font-family: Calibri, sans-serif; color: black;">&nbsp;</span><span=
 style=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</span></div>
</div>
</span>
</body>
</html>

--_000_D13846F712E037sajassiciscocom_--


From nobody Wed Mar 25 09:21:13 2015
Return-Path: <sajassi@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0071C1A8825 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:21:12 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.51
X-Spam-Level: 
X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id E9xPLiILHW4u for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:21:11 -0700 (PDT)
Received: from rcdn-iport-2.cisco.com (rcdn-iport-2.cisco.com [173.37.86.73]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CCC091A8754 for <bess@ietf.org>; Wed, 25 Mar 2015 09:21:10 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1120; q=dns/txt; s=iport; t=1427300472; x=1428510072; h=from:to:subject:date:message-id:mime-version; bh=0alkxilxTjZD0hRmy1sKZ8GPU2rs3E91YAMxrNSFYfI=; b=V6XzDCARX1gWHdttm2MBW0I2EjbuZbTPsBpsaaQo0xAi9WFlawyOBfDd phsteIYFT9E9MQr9m0Bt6BPp6P8NZa/hiCx8BkzuIw86JRNKcUakCYWcd c4nDW6t3j+U68xkACCc2hLrvJPoakgF0zYw6oakuu26oXLJ0H2cH/Wrr1 c=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0ANBQBw3xJV/4ENJK1cgkNDUl7FD4VzgVpMAQEBAQEBfYQbgQsBCwF0JwSIQg2fUalvAQEBBwEBAQEaBJAThDgFkFCDb4YAlCwig26CM38BAQE
X-IronPort-AV: E=Sophos;i="5.11,465,1422921600";  d="scan'208,217";a="406938019"
Received: from alln-core-9.cisco.com ([173.36.13.129]) by rcdn-iport-2.cisco.com with ESMTP; 25 Mar 2015 16:21:12 +0000
Received: from xhc-aln-x13.cisco.com (xhc-aln-x13.cisco.com [173.36.12.87]) by alln-core-9.cisco.com (8.14.5/8.14.5) with ESMTP id t2PGLA7U007266 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Wed, 25 Mar 2015 16:21:10 GMT
Received: from xmb-aln-x13.cisco.com ([169.254.15.15]) by xhc-aln-x13.cisco.com ([173.36.12.87]) with mapi id 14.03.0195.001; Wed, 25 Mar 2015 11:21:09 -0500
From: "Ali Sajassi (sajassi)" <sajassi@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZxezUi+2HXd7c06Z2rr8uw2a2Q==
Date: Wed, 25 Mar 2015 16:21:09 +0000
Message-ID: <D1384AA7.12E0C1%sajassi@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.3.8.130913
x-originating-ip: [10.24.162.66]
Content-Type: multipart/alternative; boundary="_000_D1384AA712E0C1sajassiciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/Iq1-tV-AZtIhD0idVVsc3G5x6eA>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:21:12 -0000

--_000_D1384AA712E0C1sajassiciscocom_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable


FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_D1384AA712E0C1sajassiciscocom_
Content-Type: text/html; charset="us-ascii"
Content-ID: <8C0360A1E4BB9642A30F90CA865E6E86@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div><br>
</div>
<div>FYI- First published July 4, 2011</div>
<div><br>
</div>
<div><a href=3D"https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-s=
egment-route">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-seg=
ment-route</a>/</div>
<div><br>
</div>
<div><br>
</div>
<div>-Ali</div>
</body>
</html>

--_000_D1384AA712E0C1sajassiciscocom_--


From nobody Wed Mar 25 09:28:57 2015
Return-Path: <shares@ndzh.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8C0C21A86DD for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:28:56 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -99.054
X-Spam-Level: 
X-Spam-Status: No, score=-99.054 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DOS_OUTLOOK_TO_MX=2.845, HTML_MESSAGE=0.001, USER_IN_WHITELIST=-100] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HPMAaKDRJTMO for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:28:55 -0700 (PDT)
Received: from hickoryhill-consulting.com (hhc-web3.hickoryhill-consulting.com [64.9.205.143]) by ietfa.amsl.com (Postfix) with ESMTP id 293DB1A8903 for <bess@ietf.org>; Wed, 25 Mar 2015 09:28:45 -0700 (PDT)
X-Default-Received-SPF: pass (skip=loggedin (res=PASS)) x-ip-name=31.133.166.41; 
From: "Susan Hares" <shares@ndzh.com>
To: <bess@ietf.org>
Date: Wed, 25 Mar 2015 12:28:39 -0400
Message-ID: <00f601d06718$c08c98a0$41a5c9e0$@ndzh.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_NextPart_000_00F7_01D066F7.397C5830"
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AdBnGFfTTvhRjp7USyuYLMi+bjkdGw==
Content-Language: en-us
X-Authenticated-User: skh@ndzh.com 
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/vFSlLvZ_MkuPRBkJRn4FVUwykCk>
Cc: "Alvaro Retana \(aretana\)" <aretana@cisco.com>, Thomas Morin <thomas.morin@orange.com>, Martin Vigoureux <martin.vigoureux@alcatel-lucent.com>
Subject: [bess] IDR BGP modules
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:28:56 -0000

This is a multipart message in MIME format.

------=_NextPart_000_00F7_01D066F7.397C5830
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit

Two models have been presented at IDR interims and the last 2-3 IETFs: 

 

http://datatracker.ietf.org/doc/draft-zhdankin-idr-bgp-cfg/

 

http://datatracker.ietf.org/doc/draft-shaikh-idr-bgp-model/

(Open Config group) 

 

The authors of these two BGP yang models have been working on creating one
model. 

 

Sue 


------=_NextPart_000_00F7_01D066F7.397C5830
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-microsoft-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"Microsoft Word 14 =
(filtered medium)"><style><!--
/* Font Definitions */
@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;}
/* 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 =
vlink=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal>Two models =
have been presented at IDR interims and the last 2-3 IETFs: =
<o:p></o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p =
class=3DMsoNormal><a =
href=3D"http://datatracker.ietf.org/doc/draft-zhdankin-idr-bgp-cfg/">http=
://datatracker.ietf.org/doc/draft-zhdankin-idr-bgp-cfg/</a><o:p></o:p></p=
><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal><a =
href=3D"http://datatracker.ietf.org/doc/draft-shaikh-idr-bgp-model/">http=
://datatracker.ietf.org/doc/draft-shaikh-idr-bgp-model/</a><o:p></o:p></p=
><p class=3DMsoNormal>(Open Config group) <o:p></o:p></p><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>The authors =
of these two BGP yang models have been working on creating one model. =
<o:p></o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p =
class=3DMsoNormal>Sue <o:p></o:p></p></div></body></html>
------=_NextPart_000_00F7_01D066F7.397C5830--


From nobody Wed Mar 25 09:59:48 2015
Return-Path: <shares@ndzh.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A764F1A8A43 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:59:46 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -99.054
X-Spam-Level: 
X-Spam-Status: No, score=-99.054 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DOS_OUTLOOK_TO_MX=2.845, HTML_MESSAGE=0.001, USER_IN_WHITELIST=-100] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QHRhNVJag-mG for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 09:59:45 -0700 (PDT)
Received: from hickoryhill-consulting.com (hhc-web3.hickoryhill-consulting.com [64.9.205.143]) by ietfa.amsl.com (Postfix) with ESMTP id ED1371A88EA for <bess@ietf.org>; Wed, 25 Mar 2015 09:59:44 -0700 (PDT)
X-Default-Received-SPF: pass (skip=forwardok (res=PASS)) x-ip-name=31.133.140.50; 
From: "Susan Hares" <shares@ndzh.com>
To: <bess@ietf.org>
References: <00f601d06718$c08c98a0$41a5c9e0$@ndzh.com>
In-Reply-To: <00f601d06718$c08c98a0$41a5c9e0$@ndzh.com>
Date: Wed, 25 Mar 2015 12:59:39 -0400
Message-ID: <000901d0671d$15ce0480$416a0d80$@ndzh.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_NextPart_000_000A_01D066FB.8ECBA6C0"
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQGNDExOacXcXogIjTmvaAkGOX8DkJ20H8DA
Content-Language: en-us
X-Authenticated-User: skh@ndzh.com 
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/19ZNfEkRwpXPVCGKm4DWe9VHY6g>
Cc: "'Alvaro Retana \(aretana\)'" <aretana@cisco.com>, 'Thomas Morin' <thomas.morin@orange.com>, 'Martin Vigoureux' <martin.vigoureux@alcatel-lucent.com>
Subject: Re: [bess] IDR BGP modules
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 16:59:46 -0000

This is a multipart message in MIME format.

------=_NextPart_000_000A_01D066FB.8ECBA6C0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit

Deepak, Kishore, Tapraj, Ali, and Luay: 

 

Thank you for making the effort to create an EVPN draft. 

 

 <http://tools.ietf.org/html?draft=draft-tsingh-bess-pbb-evpn-yang-cfg>
draft-tsingh-bess-pbb-evpn-yang-cfg

 

I look forward to your revision with a more holistic view that includes the
draft below. 

 

Sue 

 

 

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Susan Hares
Sent: Wednesday, March 25, 2015 12:29 PM
To: bess@ietf.org
Cc: Alvaro Retana (aretana); Thomas Morin; Martin Vigoureux
Subject: [bess] IDR BGP modules

 

Two models have been presented at IDR interims and the last 2-3 IETFs: 

 

http://datatracker.ietf.org/doc/draft-zhdankin-idr-bgp-cfg/

 

http://datatracker.ietf.org/doc/draft-shaikh-idr-bgp-model/

(Open Config group) 

 

The authors of these two BGP yang models have been working on creating one
model. 

 

Sue 


------=_NextPart_000_000A_01D066FB.8ECBA6C0
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-microsoft-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"Microsoft Word 14 =
(filtered medium)"><style><!--
/* Font Definitions */
@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;}
/* 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 =
vlink=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><span =
style=3D'color:#1F497D'>Deepak, Kishore, Tapraj, Ali, and Luay: =
<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'>Thank you for making the =
effort to create an EVPN draft. <o:p></o:p></span></p><p =
class=3DMsoNormal><span =
style=3D'color:#1F497D'><o:p>&nbsp;</o:p></span></p><p =
class=3DMsoNormal><a =
href=3D"http://tools.ietf.org/html?draft=3Ddraft-tsingh-bess-pbb-evpn-yan=
g-cfg"><span style=3D'font-size:10.0pt;font-family:"Courier =
New";color:#440088'>draft-tsingh-bess-pbb-evpn-yang-cfg</span></a><o:p></=
o:p></p><p class=3DMsoNormal><span =
style=3D'color:#1F497D'><o:p>&nbsp;</o:p></span></p><p =
class=3DMsoNormal><span style=3D'color:#1F497D'>I look forward to your =
revision with a more holistic view that includes the draft below. =
<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'>Sue =
<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'><o:p>&nbsp;</o:p></span></p><div><div =
style=3D'border:none;border-top:solid #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"'> =
BESS [mailto:bess-bounces@ietf.org] <b>On Behalf Of </b>Susan =
Hares<br><b>Sent:</b> Wednesday, March 25, 2015 12:29 PM<br><b>To:</b> =
bess@ietf.org<br><b>Cc:</b> Alvaro Retana (aretana); Thomas Morin; =
Martin Vigoureux<br><b>Subject:</b> [bess] IDR BGP =
modules<o:p></o:p></span></p></div></div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>Two models =
have been presented at IDR interims and the last 2-3 IETFs: =
<o:p></o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p =
class=3DMsoNormal><a =
href=3D"http://datatracker.ietf.org/doc/draft-zhdankin-idr-bgp-cfg/">http=
://datatracker.ietf.org/doc/draft-zhdankin-idr-bgp-cfg/</a><o:p></o:p></p=
><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal><a =
href=3D"http://datatracker.ietf.org/doc/draft-shaikh-idr-bgp-model/">http=
://datatracker.ietf.org/doc/draft-shaikh-idr-bgp-model/</a><o:p></o:p></p=
><p class=3DMsoNormal>(Open Config group) <o:p></o:p></p><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>The authors =
of these two BGP yang models have been working on creating one model. =
<o:p></o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p =
class=3DMsoNormal>Sue <o:p></o:p></p></div></body></html>
------=_NextPart_000_000A_01D066FB.8ECBA6C0--


From nobody Wed Mar 25 13:14:21 2015
Return-Path: <tsingh@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B04C31B2B45 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 13:14:19 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.901
X-Spam-Level: 
X-Spam-Status: No, score=-1.901 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oJT8EWmFw0Oo for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 13:14:17 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1bon0724.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::1:724]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 842D51A9008 for <bess@ietf.org>; Wed, 25 Mar 2015 13:14:17 -0700 (PDT)
Received: from BY2PR05MB061.namprd05.prod.outlook.com (10.242.34.141) by BY2PR05MB063.namprd05.prod.outlook.com (10.242.34.151) with Microsoft SMTP Server (TLS) id 15.1.118.21; Wed, 25 Mar 2015 20:13:57 +0000
Received: from BY2PR05MB061.namprd05.prod.outlook.com ([169.254.3.187]) by BY2PR05MB061.namprd05.prod.outlook.com ([169.254.3.187]) with mapi id 15.01.0125.002; Wed, 25 Mar 2015 20:13:57 +0000
From: Tapraj Singh <tsingh@juniper.net>
To: Susan Hares <shares@ndzh.com>
Thread-Topic: [bess] IDR BGP modules
Thread-Index: AdBnGFfTTvhRjp7USyuYLMi+bjkdGwABLy2AAAbJPmo=
Date: Wed, 25 Mar 2015 20:13:57 +0000
Message-ID: <78116A8F-C2BE-4D9E-934E-01C364BC0549@juniper.net>
References: <00f601d06718$c08c98a0$41a5c9e0$@ndzh.com>, <000901d0671d$15ce0480$416a0d80$@ndzh.com>
In-Reply-To: <000901d0671d$15ce0480$416a0d80$@ndzh.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [31.133.163.138]
authentication-results: spf=none (sender IP is ) smtp.mailfrom=tsingh@juniper.net; 
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BY2PR05MB063;
x-microsoft-antispam-prvs: <BY2PR05MB063F1385CC8EBF9DFBC1923DB0B0@BY2PR05MB063.namprd05.prod.outlook.com>
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(51414003)(24454002)(199003)(377454003)(189002)(19617315012)(50986999)(54356999)(76176999)(66066001)(87936001)(40100003)(83716003)(2656002)(82746002)(19580395003)(97736003)(122556002)(110136001)(19580405001)(64706001)(46102003)(16236675004)(15975445007)(102836002)(33656002)(19625215002)(99286002)(101416001)(86362001)(92566002)(36756003)(561944003)(68736005)(62966003)(77156002)(106356001)(2950100001)(2900100001)(105586002)(217873001)(104396002); DIR:OUT; SFP:1102; SCL:1; SRVR:BY2PR05MB063; H:BY2PR05MB061.namprd05.prod.outlook.com; FPR:; SPF:None; PTR:InfoNoRecords; A:1; MX:1; LANG:en; 
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5005006)(5002010); SRVR:BY2PR05MB063; BCL:0; PCL:0; RULEID:; SRVR:BY2PR05MB063; 
x-forefront-prvs: 052670E5A4
received-spf: None (protection.outlook.com: juniper.net does not designate permitted sender hosts)
Content-Type: multipart/alternative; boundary="_000_78116A8FC2BE4D9E934E01C364BC0549junipernet_"
MIME-Version: 1.0
X-OriginatorOrg: juniper.net
X-MS-Exchange-CrossTenant-originalarrivaltime: 25 Mar 2015 20:13:57.6338 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BY2PR05MB063
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/A4OmHrfeX-QUKgP0Xbu3pnSCpVw>
Cc: "Alvaro Retana \(aretana\)" <aretana@cisco.com>, "EXT - thomas.morin@orange.com" <thomas.morin@orange.com>, Martin Vigoureux <martin.vigoureux@alcatel-lucent.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] IDR BGP modules
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 20:14:19 -0000

--_000_78116A8FC2BE4D9E934E01C364BC0549junipernet_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Hi Susan,
    Thank you for your inputs and we will go through these drafts to align =
our draft proposal.
    Our idea here is to have  yang data modeling PBB EVPN only. We will ref=
er to BGP related section to BGP yang modeling drafts proposal.

Thanks
Tapraj



On Mar 25, 2015, at 12:00 PM, Susan Hares <shares@ndzh.com<mailto:shares@nd=
zh.com>> wrote:

Deepak, Kishore, Tapraj, Ali, and Luay:

Thank you for making the effort to create an EVPN draft.

draft-tsingh-bess-pbb-evpn-yang-cfg<http://tools.ietf.org/html?draft=3Ddraf=
t-tsingh-bess-pbb-evpn-yang-cfg>

I look forward to your revision with a more holistic view that includes the=
 draft below.

Sue


From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Susan Hares
Sent: Wednesday, March 25, 2015 12:29 PM
To: bess@ietf.org<mailto:bess@ietf.org>
Cc: Alvaro Retana (aretana); Thomas Morin; Martin Vigoureux
Subject: [bess] IDR BGP modules

Two models have been presented at IDR interims and the last 2-3 IETFs:

http://datatracker.ietf.org/doc/draft-zhdankin-idr-bgp-cfg/

http://datatracker.ietf.org/doc/draft-shaikh-idr-bgp-model/
(Open Config group)

The authors of these two BGP yang models have been working on creating one =
model.

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

--_000_78116A8FC2BE4D9E934E01C364BC0549junipernet_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
</head>
<body dir=3D"auto">
<div>Hi Susan,</div>
<div>&nbsp; &nbsp; Thank you for your inputs and we will go through these d=
rafts to align our draft proposal.&nbsp;</div>
<div>&nbsp; &nbsp; Our idea here is to have &nbsp;yang data modeling PBB EV=
PN only. We will refer to BGP related section to BGP yang modeling drafts p=
roposal.</div>
<div><br>
</div>
<div>Thanks</div>
<div>Tapraj&nbsp;<br>
<br>
<br>
</div>
<div><br>
On Mar 25, 2015, at 12:00 PM, Susan Hares &lt;<a href=3D"mailto:shares@ndzh=
.com">shares@ndzh.com</a>&gt; wrote:<br>
<br>
</div>
<blockquote type=3D"cite">
<div>
<meta name=3D"Generator" content=3D"Microsoft Word 14 (filtered medium)">
<style><!--
/* Font Definitions */
@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;}
/* 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]-->
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Deepak, Kishore, Tapra=
j, Ali, and Luay:
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Thank you for making t=
he effort to create an EVPN draft.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><a href=3D"http://tools.ietf.org/html?draft=3Ddraft-=
tsingh-bess-pbb-evpn-yang-cfg"><span style=3D"font-size:10.0pt;font-family:=
&quot;Courier New&quot;;color:#440088">draft-tsingh-bess-pbb-evpn-yang-cfg<=
/span></a><o:p></o:p></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">I look forward to your=
 revision with a more holistic view that includes the draft below.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D">Sue <o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"color:#1F497D"><o:p>&nbsp;</o:p></spa=
n></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;"> BESS [<a=
 href=3D"mailto:bess-bounces@ietf.org">mailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Susan Hares<br>
<b>Sent:</b> Wednesday, March 25, 2015 12:29 PM<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org">bess@ietf.org</a><br>
<b>Cc:</b> Alvaro Retana (aretana); Thomas Morin; Martin Vigoureux<br>
<b>Subject:</b> [bess] IDR BGP modules<o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Two models have been presented at IDR interims and t=
he last 2-3 IETFs:
<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><a href=3D"http://datatracker.ietf.org/doc/draft-zhd=
ankin-idr-bgp-cfg/">http://datatracker.ietf.org/doc/draft-zhdankin-idr-bgp-=
cfg/</a><o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal"><a href=3D"http://datatracker.ietf.org/doc/draft-sha=
ikh-idr-bgp-model/">http://datatracker.ietf.org/doc/draft-shaikh-idr-bgp-mo=
del/</a><o:p></o:p></p>
<p class=3D"MsoNormal">(Open Config group) <o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">The authors of these two BGP yang models have been w=
orking on creating one model.
<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Sue <o:p></o:p></p>
</div>
</div>
</blockquote>
<blockquote type=3D"cite">
<div><span>_______________________________________________</span><br>
<span>BESS mailing list</span><br>
<span><a href=3D"mailto:BESS@ietf.org">BESS@ietf.org</a></span><br>
<span><a href=3D"https://www.ietf.org/mailman/listinfo/bess">https://www.ie=
tf.org/mailman/listinfo/bess</a></span><br>
</div>
</blockquote>
</body>
</html>

--_000_78116A8FC2BE4D9E934E01C364BC0549junipernet_--


From nobody Wed Mar 25 14:34:13 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5F8931A8AE4 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 14:34:12 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.21
X-Spam-Level: 
X-Spam-Status: No, score=-4.21 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id V50o3XYXQv1u for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 14:34:10 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 72F711A1A2D for <bess@ietf.org>; Wed, 25 Mar 2015 14:34:10 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BUD25734; Wed, 25 Mar 2015 21:34:09 +0000 (GMT)
Received: from NKGEML403-HUB.china.huawei.com (10.98.56.34) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Wed, 25 Mar 2015 21:34:07 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml403-hub.china.huawei.com ([10.98.56.34]) with mapi id 14.03.0158.001; Thu, 26 Mar 2015 05:34:05 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Ali Sajassi (sajassi)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZxezUi+2HXd7c06Z2rr8uw2a2Z0ttLRN
Date: Wed, 25 Mar 2015 21:34:04 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com>
In-Reply-To: <D1384AA7.12E0C1%sajassi@cisco.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.153.46]
Content-Type: multipart/alternative; boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F8571E7nkgeml501mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/ukQTCksCOVMWuiZEnoOMViQArTk>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 21:34:12 -0000

--_000_DD5FC8DE455C3348B94340C0AB5517334F8571E7nkgeml501mbschi_
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org] on behalf of Ali Sajassi (sajassi) [saja=
ssi@cisco.com]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m


FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_DD5FC8DE455C3348B94340C0AB5517334F8571E7nkgeml501mbschi_
Content-Type: text/html; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Diso-8859-=
1">
<style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
</head>
<body style=3D"FONT-FAMILY: Calibri,sans-serif; WORD-WRAP: break-word; COLO=
R: rgb(0,0,0); FONT-SIZE: 14px" fPStyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3">Hi Ali,</font></p>
<p><font size=3D"3">Thanks for your information. I scanned through this dra=
ft, it really introduces inter-chassis message for DF election handshaking,=
 the requirements in this draft is to eliminate transiet Loop and traffic d=
uplication. Current EVPN DF election
 mechanism already eliminated loop and traffic duplication by configuring l=
ong reception timer on each multi-homed PE, but up to reception timer traff=
ic disruption issue still exist. EVPN for DCI is an important use case for =
EVPN, up to reception timer traffic
 disruption can't be tolerated for service providers, it should be improved=
.</font></p>
<p><font size=3D"3">Also for accuracy, i think handshaking state machine on=
 each multi-homed PE is also needed. From solution perspective, in my draft=
, no inter-chassis message is introduced, only one new extended community i=
s introduced, i think the process
 is comparatively simple than your following draft.</font></p>
<p><font size=3D"3"></font>&nbsp;</p>
<p><font size=3D"3">Current EVPN DF election has some drawbacks, so there a=
re three new drafts about DF election emerged. I think BESS WG can consider=
 these three drafts in global view,&nbsp; a single,comprehensive new DF ele=
ction draft is hoped.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3">thanks.</font></p>
<p><font size=3D"3"></font>&nbsp;</p>
<p><font size=3D"3">Thanks,</font></p>
<p><font size=3D"3">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF484031"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> BESS [bess-bounces@ietf.org] on beha=
lf of Ali Sajassi (sajassi) [sajassi@cisco.com]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> bess@ietf.org<br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm<br>
</font><br>
</div>
<div></div>
<div>
<div><br>
</div>
<div>FYI- First published July 4, 2011</div>
<div><br>
</div>
<div><a href=3D"https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-s=
egment-route" target=3D"_blank">https://datatracker.ietf.org/doc/draft-saja=
ssi-l2vpn-evpn-segment-route</a>/</div>
<div><br>
</div>
<div><br>
</div>
<div>-Ali</div>
</div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F8571E7nkgeml501mbschi_--


From nobody Wed Mar 25 16:17:11 2015
Return-Path: <jdrake@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 370C31A87C1 for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 16:17:10 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.901
X-Spam-Level: 
X-Spam-Status: No, score=-1.901 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id poXyK2BdzTeY for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 16:17:07 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1on0779.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::779]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2E6241A8825 for <bess@ietf.org>; Wed, 25 Mar 2015 16:17:07 -0700 (PDT)
Received: from BLUPR05MB561.namprd05.prod.outlook.com (10.141.202.139) by BLUPR05MB117.namprd05.prod.outlook.com (10.255.214.11) with Microsoft SMTP Server (TLS) id 15.1.112.16; Wed, 25 Mar 2015 23:16:49 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) by BLUPR05MB561.namprd05.prod.outlook.com (10.141.202.139) with Microsoft SMTP Server (TLS) id 15.1.118.21; Wed, 25 Mar 2015 23:16:48 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) by BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) with mapi id 15.01.0118.022; Wed, 25 Mar 2015 23:16:48 +0000
From: John E Drake <jdrake@juniper.net>
To: Haoweiguo <haoweiguo@huawei.com>, "Ali Sajassi (sajassi)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ0Nzx4kiG6xy2E29vzlr08PJ+J0tz1Yw
Date: Wed, 25 Mar 2015 23:16:47 +0000
Message-ID: <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [66.129.239.11]
authentication-results: huawei.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB561; UriScan:; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB117; 
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(69234005)(51444003)(377454003)(164054003)(19609705001)(77096005)(102836002)(86362001)(561944003)(19580395003)(19580405001)(16236675004)(66066001)(50986999)(19625215002)(33656002)(54356999)(76176999)(74316001)(15975445007)(46102003)(62966003)(77156002)(2501003)(107886001)(2900100001)(40100003)(92566002)(2950100001)(122556002)(2656002)(87936001)(76576001)(106116001)(19617315012)(99286002)(19300405004); DIR:OUT; SFP:1102; SCL:1; SRVR:BLUPR05MB561; H:BLUPR05MB562.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BLUPR05MB561CEF1CBA9AA76D3DD385AC70B0@BLUPR05MB561.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BLUPR05MB561; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB561; 
x-forefront-prvs: 052670E5A4
Content-Type: multipart/alternative; boundary="_000_BLUPR05MB5629642F4A86ED28469DF19C70B0BLUPR05MB562namprd_"
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-originalarrivaltime: 25 Mar 2015 23:16:47.5725 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BLUPR05MB561
X-OriginatorOrg: juniper.net
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/G2xAqHcJ5LCKoijMptYrQ0Dh8TA>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 23:17:10 -0000

--_000_BLUPR05MB5629642F4A86ED28469DF19C70B0BLUPR05MB562namprd_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can't  use the ES route to co-ordinate DF election because you would ne=
ed to carry your new extended community for each EVI and they would not all=
 fit.  You also can't use the Per EVI Ethernet AD route because that is pro=
cessed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya's draft is sufficiently stable that it renders your draft moot, even i=
f it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org] on behalf of Ali Sajassi (sajassi) [saja=
ssi@cisco.com]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_BLUPR05MB5629642F4A86ED28469DF19C70B0BLUPR05MB562namprd_
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=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<!--[if !mso]><style>v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style><![endif]--><style><!--
/* Font Definitions */
@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;}
/* 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;
	margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman",serif;}
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=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;,sans-serif;color:#1F497D">Weiguo,<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">Your proposal introduces a control pl=
ane processing load that is O(#EVIs * PEs) per DF election and given that t=
here can be 4K EVIs per ES, this looks like a
 *<b>substantial</b>* load.&nbsp; Furthermore,<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">you can&#8217;t &nbsp;use the ES rout=
e to co-ordinate DF election because you would need to carry your new exten=
ded community for each EVI and they would not all fit.&nbsp; You
 also can&#8217;t use the Per EVI Ethernet AD route because that is process=
ed by all PEs in the EVI.<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">I think that from a practical perspec=
tive the new DF election proposed in Satya&#8217;s draft is sufficiently st=
able that it renders your draft moot, even if it could
 be made to work.<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D"><o:p></o:p></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Yours Irrespectively,<o:p></o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">John<o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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 #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif">From:</span></b><span style=3D"font-size:11.0pt;=
font-family:&quot;Calibri&quot;,sans-serif"> BESS [mailto:bess-bounces@ietf=
.org]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<div>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Hi=
 Ali,</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Th=
anks for your information. I scanned through this draft, it really introduc=
es inter-chassis message for DF election handshaking, the requirements in t=
his draft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black"><o:p><=
/o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Al=
so for accuracy, i think handshaking state machine on each multi-homed PE i=
s also needed. From solution perspective, in my draft, no inter-chassis mes=
sage is introduced, only one new extended community
 is introduced, i think the process is comparatively simple than your follo=
wing draft.</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&=
quot;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-seri=
f;color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Cu=
rrent EVPN DF election has some drawbacks, so there are three new drafts ab=
out DF election emerged. I think BESS WG can consider these three drafts in=
 global view,&nbsp; a single,comprehensive new DF
 election draft is hoped.</span><span style=3D"font-size:10.0pt;font-family=
:&quot;Tahoma&quot;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-seri=
f;color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">th=
anks.</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-seri=
f;color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Th=
anks,</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">we=
iguo</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,s=
ans-serif;color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF484031">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> BESS [bess-bounces@ietf.org] on behalf
 of Ali Sajassi (sajassi) [sajassi@cisco.com]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org">bess@ietf.org</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><o:p>&nbsp;</o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">FYI- First published Jul=
y 4, 2011<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><o:p>&nbsp;</o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><a href=3D"https://datat=
racker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_blan=
k">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route<=
/a>/<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><o:p>&nbsp;</o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><o:p>&nbsp;</o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">-Ali<o:p></o:p></span></=
p>
</div>
</div>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_BLUPR05MB5629642F4A86ED28469DF19C70B0BLUPR05MB562namprd_--


From nobody Wed Mar 25 16:32:53 2015
Return-Path: <nordmark@acm.org>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4F19E1A899C for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 16:32:52 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.735
X-Spam-Level: 
X-Spam-Status: No, score=-0.735 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_12=0.6, J_CHICKENPOX_14=0.6, RCVD_IN_DNSWL_LOW=-0.7, SPF_SOFTFAIL=0.665] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Q0107UfCcFFJ for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 16:32:51 -0700 (PDT)
Received: from c.mail.sonic.net (c.mail.sonic.net [64.142.111.80]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 194E21A1B51 for <bess@ietf.org>; Wed, 25 Mar 2015 16:32:51 -0700 (PDT)
Received: from [172.22.240.95] ([162.210.130.3]) (authenticated bits=0) by c.mail.sonic.net (8.15.1/8.15.1) with ESMTPSA id t2PNWcYe028869 (version=TLSv1.2 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Wed, 25 Mar 2015 16:32:39 -0700
Message-ID: <55134595.7060209@acm.org>
Date: Wed, 25 Mar 2015 18:32:37 -0500
From: Erik Nordmark <nordmark@acm.org>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: Antoni Przygienda <antoni.przygienda@ericsson.com>, "Rabadan, Jorge (Jorge) (jorge.rabadan@alcatel-lucent.com)" <jorge.rabadan@alcatel-lucent.com>, "Henderickx, Wim (Wim) (wim.henderickx@alcatel-lucent.com)" <wim.henderickx@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se>
In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 8bit
X-Sonic-CAuth: UmFuZG9tSVZRERbiynesqq7kqyblYC2tbVkjGK5vRDmNXpFhIbx1/U0tj+pjig6uOGCCscetZKjumy51PRg+AAnYhPcqiAa3
X-Sonic-ID: C;alcOOUfT5BG9L9BwQIsAyQ== M;0DbmOUfT5BG9L9BwQIsAyQ==
X-Sonic-Spam-Details: 0.0/5.0 by cerberusd
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/whuY90QdR15fetR6YNgrgOGhQqY>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 23:32:52 -0000

On 3/25/15 10:23 AM, Antoni Przygienda wrote:
>
> Watching the presentation in the group and instead crowding the mike 
> here are my comments:
>
> a)It is worth explaining what flavor of ARP proxy eVPN implements. 
> ‘proxy ARP’ I found out has different flavors including e.g. the 
> router responding with its own MAC to requests representing remote 
> hosts. Customers & other folks are easily confused by the overloaded 
> ‘proxy ARP’ term in eVPN context.
>
Yes, that is a part that is underspecified for EVPN. I assume that the 
response would include the MAC address of the CE instead of the router's 
own MAC address.

> b)It is worth explaining what is suggested behavior if eVPN advertises 
> the same IP with multiple MACs and what happens when e.g. the served 
> MAC vanishes
>
Doesn't the EVPN RFC already stating that the routes would be withdrawn 
in that case?
>
> c)It is worth explaining what the suggested behavior should be when 
> ‘local-bit’ MACs are being advertised from remotes (and when those 
> collide)
>
Does L2VPN handle that in any interesting way? I don't think EVPN 
introduces any new failure modes for duplicate MAC addresses.
>
> d)eVPN draft does not explain anything about possibility to snoop @ DHCP
>

> e)the IP duplicate detection is an interesting thing given the 
> anycast/multicast and other zoo
>
Presumably not only anycast, but also for some host that might have a 
single IP address and use it with different MAC addresses (I think 
certain configurations of NIC bonding has that effect.)

FWIW the SAVI RFC provides an approach to tell whether it was a change 
in the IP to MAC binding or the IP address being reachable at multiple 
MAC addresses at the same time (which is basically to unicast an NS to 
the old MAC address - I don't know if that would work will for ARP as well).

That still leaves out what to do in the case of anycast addresses etc.

Regards,
    Erik



> --- tony
>
> //
>
> /There are basically two types of people. People who accomplish 
> things, and people who claim to have accomplished things. The first 
> group is less crowded. 
> <http://www.brainyquote.com/quotes/quotes/m/marktwain393535.html>/
>
> /~~~ Mark Twain 
> <http://www.brainyquote.com/quotes/quotes/m/marktwain393535.html>/
>
>
>
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess


From nobody Wed Mar 25 16:56:23 2015
Return-Path: <antoni.przygienda@ericsson.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1035C1A92FF for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 16:56:22 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.601
X-Spam-Level: 
X-Spam-Status: No, score=-3.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id j_mVdUO3ubfr for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 16:56:20 -0700 (PDT)
Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 07C0B1A9025 for <bess@ietf.org>; Wed, 25 Mar 2015 16:56:19 -0700 (PDT)
X-AuditID: c618062d-f79686d0000030a8-f9-5512f55536c1
Received: from EUSAAHC005.ericsson.se (Unknown_Domain [147.117.188.87]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id C4.C8.12456.555F2155; Wed, 25 Mar 2015 18:50:13 +0100 (CET)
Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC005.ericsson.se ([147.117.188.87]) with mapi id 14.03.0210.002; Wed, 25 Mar 2015 19:56:17 -0400
From: Antoni Przygienda <antoni.przygienda@ericsson.com>
To: Erik Nordmark <nordmark@acm.org>, "Rabadan, Jorge (Jorge) (jorge.rabadan@alcatel-lucent.com)" <jorge.rabadan@alcatel-lucent.com>, "Henderickx, Wim (Wim) (wim.henderickx@alcatel-lucent.com)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAZ/NmAAAeuvZA=
Date: Wed, 25 Mar 2015 23:56:17 +0000
Message-ID: <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org>
In-Reply-To: <55134595.7060209@acm.org>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [147.117.188.12]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFnrPLMWRmVeSWpSXmKPExsUyuXRPuG7oV6FQg7Mr5C1WHJ/JbPHt0hIW i/6pO5ksZj/+webA4nH5irdH67O9rB5LlvxkCmCO4rJJSc3JLEst0rdL4Mpo2ryEseA0S8XN zYvZGhgPMHcxcnJICJhI7Ov+xAZhi0lcuLceyObiEBI4wiix7t5RJpCEkMByRokDu7hBbDYB C4nL354ygxSJCLxmlJizeg4rSIJZQFni8Pzz7CC2sICCRHvHV7ANIgKKEvOfr2OEsK0k7nxY BmazCKhKrOw6wAJi8wp4S3zfMIcZYlmGxJb/m8BmcgqoS8x/tABsJiPQdd9PrWGC2CUucevJ fCaIqwUkluw5D/WNqMTLx/9YIWwliTmvrzFD1OtILNgN8SWzgLbEsoWvmSH2CkqcnPmEZQKj 2CwkY2chaZmFpGUWkpYFjCyrGDlKi1PLctONDDYxAmPomASb7g7GPS8tDzEKcDAq8fAWvBEM FWJNLCuuzD3EKM3BoiTOu+jBwRAhgfTEktTs1NSC1KL4otKc1OJDjEwcnFINjCv+h53NKVaf ss4wVosvsCSb+QlHxF+OSwz218QS1s198PTG5M9HQ5eeOF7y6n7kzQ/8+9fb8/mcDLdZ92mX aEDdOzljny/rM7a8cDp2hPVtxRPL94JvDlhq/Xt7X1jee2JM9YM1mwI37bdY63JHZpey8S+9 5va6qjkqiefbLl1O2fJy18QI6wYlluKMREMt5qLiRAAguW+dggIAAA==
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/DlJWUqOTA8h1kJ-LM3o2qtbmDkg>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 23:56:22 -0000

> > b)It is worth explaining what is suggested behavior if eVPN advertises
> > the same IP with multiple MACs and what happens when e.g. the served
> > MAC vanishes
> >
> Doesn't the EVPN RFC already stating that the routes would be withdrawn i=
n
> that case?

The scenario I had in mind was when eVPN PE receives=20

>From PE2  IP1/M1  and  later
>From PE3  IP1/M2

while having answered with IP1/M1 per proxy alrady. Additionally, in such s=
ituation ends up seeing

>From PE2   IP1/<no MAC>

So the answer it gave is not valid anymore all of a sudden.

--- tony=20


From nobody Wed Mar 25 20:38:41 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 89EAF1A700B for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 20:38:40 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.61
X-Spam-Level: 
X-Spam-Status: No, score=-3.61 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id aF_S42VyDaUi for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 20:38:38 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A255E1A6F5D for <bess@ietf.org>; Wed, 25 Mar 2015 20:38:37 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQS25794; Thu, 26 Mar 2015 03:38:36 +0000 (GMT)
Received: from NKGEML403-HUB.china.huawei.com (10.98.56.34) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Thu, 26 Mar 2015 03:38:34 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml403-hub.china.huawei.com ([10.98.56.34]) with mapi id 14.03.0158.001; Thu, 26 Mar 2015 11:38:31 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: John E Drake <jdrake@juniper.net>, "Ali Sajassi (sajassi)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ1HNDitJcP09SEmMV2KHyzZvtJ0uF9eC
Date: Thu, 26 Mar 2015 03:38:30 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com>
In-Reply-To: <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.146.35]
Content-Type: multipart/alternative; boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F857294nkgeml501mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/6jVax9BMZVjqJSi-kICvvgOLyg0>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 03:38:40 -0000

--_000_DD5FC8DE455C3348B94340C0AB5517334F857294nkgeml501mbschi_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org] on behalf of Ali Sajassi (sajassi) [saja=
ssi@cisco.com]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_DD5FC8DE455C3348B94340C0AB5517334F857294nkgeml501mbschi_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
<style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.MsoChpDefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple" fPStyle=3D"1" ocsi=3D"0=
">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Hi John,<=
/font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">&nbsp;Fir=
stly i think EVPN community should reach consensus on the issues of current=
 DF election mechanism. All these issues should be resolved in a single new=
 DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3"></font></=
font></font>&nbsp;</p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Thanks,</=
font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">weiguo</f=
ont></p>
</font></font>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF271558"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [jdrake@juniper.net]<br=
>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Your proposal introduces a control plane pro=
cessing load that is O(#EVIs * PEs) per DF election and given that there ca=
n be 4K EVIs per ES, this looks like
 a *<b>substantial</b>* load.&nbsp; Furthermore,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">you can=92t &nbsp;use the ES route to co-ord=
inate DF election because you would need to carry your new extended communi=
ty for each EVI and they would not all fit.&nbsp;
 You also can=92t use the Per EVI Ethernet AD route because that is process=
ed by all PEs in the EVI.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">I think that from a practical perspective th=
e new DF election proposed in Satya=92s draft is sufficiently stable that i=
t renders your draft moot, even if it
 could be made to work.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif;=
 FONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY: 'Calibri',san=
s-serif; FONT-SIZE: 11pt"> BESS [mailto:bess-bounces@ietf.org]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Hi Ali,</=
span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SI=
ZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Thanks fo=
r your information. I scanned through this draft, it really introduces inte=
r-chassis message for DF election handshaking, the requirements in this dra=
ft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"FON=
T-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Also for =
accuracy, i think handshaking state machine on each multi-homed PE is also =
needed. From solution perspective, in my draft, no inter-chassis message is=
 introduced, only one new extended
 community is introduced, i think the process is comparatively simple than =
your following draft.</span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif=
; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE=
: 10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Current E=
VPN DF election has some drawbacks, so there are three new drafts about DF =
election emerged. I think BESS WG can consider these three drafts in global=
 view,&nbsp; a single,comprehensive new
 DF election draft is hoped.</span><span style=3D"FONT-FAMILY: 'Tahoma',san=
s-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE=
: 10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">thanks.</=
span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SI=
ZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE=
: 10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Thanks,</=
span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SI=
ZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">weiguo</s=
pan><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZ=
E: 10pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></=
b><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE:=
 10pt"> BESS [bess-bounces@ietf.org] on
 behalf of Ali Sajassi (sajassi) [sajassi@cisco.com]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F857294nkgeml501mbschi_--


From nobody Wed Mar 25 22:04:41 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A602D1ACCFB for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 22:04:37 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -5.71
X-Spam-Level: 
X-Spam-Status: No, score=-5.71 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_12=0.6, J_CHICKENPOX_14=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Whhezepsm7AE for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 22:04:35 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 59FC01ACC92 for <bess@ietf.org>; Wed, 25 Mar 2015 22:04:35 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id AE1743BC39E7F; Thu, 26 Mar 2015 05:04:32 +0000 (GMT)
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2Q54XwL009094 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 26 Mar 2015 06:04:33 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR712WXCHHUB03.zeu.alcatel-lucent.com ([135.239.2.74]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 06:04:33 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Erik Nordmark <nordmark@acm.org>, Antoni Przygienda <antoni.przygienda@ericsson.com>, "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAPgqSA///mFgA=
Date: Thu, 26 Mar 2015 05:04:32 +0000
Message-ID: <D138D9BF.6C4CE%jorge.rabadan@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org>
In-Reply-To: <55134595.7060209@acm.org>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.39]
Content-Type: text/plain; charset="utf-8"
Content-ID: <21AA555885E98445B42E9990725BFA6B@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/CiMQEXwPiCWprpRBmopA4He4hwI>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 05:04:37 -0000
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==


From nobody Wed Mar 25 22:04:50 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 59A2B1ACD0A for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 22:04:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id h_t_mIkmOqVS for <bess@ietfa.amsl.com>; Wed, 25 Mar 2015 22:04:40 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpgre-esg-01.alcatel-lucent.com [135.245.210.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8A1AF1ACD01 for <bess@ietf.org>; Wed, 25 Mar 2015 22:04:38 -0700 (PDT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (unknown [135.239.2.122]) by Websense Email Security Gateway with ESMTPS id AECEB2DF945AB; Thu, 26 Mar 2015 05:04:36 +0000 (GMT)
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id t2Q54bFw012222 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 26 Mar 2015 06:04:37 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR711WXCHHUB01.zeu.alcatel-lucent.com ([135.239.2.111]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 06:04:37 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Antoni Przygienda <antoni.przygienda@ericsson.com>, Erik Nordmark <nordmark@acm.org>, "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAPgqSAAADTmID//+C+gA==
Date: Thu, 26 Mar 2015 05:04:36 +0000
Message-ID: <D138E061.6C519%jorge.rabadan@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se>
In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.39]
Content-Type: text/plain; charset="utf-8"
Content-ID: <3D0A28177ADE1744A7E4AC2E2347C4FE@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/IAeEXSBAzs6fATRktCXvkH5P6rI>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 05:04:41 -0000
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From nobody Thu Mar 26 02:41:05 2015
Return-Path: <satyamoh@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A5DC01ACE26 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 02:41:03 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -13.91
X-Spam-Level: 
X-Spam-Status: No, score=-13.91 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Q6nKiRqn2hZR for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 02:41:01 -0700 (PDT)
Received: from alln-iport-1.cisco.com (alln-iport-1.cisco.com [173.37.142.88]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B49251A90BC for <bess@ietf.org>; Thu, 26 Mar 2015 02:41:00 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=19247; q=dns/txt; s=iport; t=1427362860; x=1428572460; h=from:to:subject:date:message-id:references:in-reply-to: mime-version; bh=wiostwQONZbZrJJ1dpSDDUJ2KJROxKBpBSdEPcrlet8=; b=V3JY1C8lJa0yJPmP2kBT5wGrvexQ90B5S4B60DvAnyF82e5WyZH865+G cgfO51UMs2X3srxsHiYMDwM239+kOzS2e7nGEuWxTlZ35xgfzr81mmKe0 dPVphn6D0zv++yrUOLj7kHg+eMNdyfkRReDzrNPOq4F/QRAwslybpq2r7 U=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0BGBQBT0xNV/4QNJK1cgkNDUloExRuFcwKBX0wBAQEBAQF9hBQBAQEEZyICAQgRAwEBASgHMhQJCAIEARIUB4gUDcowAQEBAQEBAQEBAQEBAQEBAQEBAQEBEwSKIn+EE1INCgEGhCcFjkKCDoNvhgCBG4YJiUGDRyKCAR2BUG+BRH8BAQE
X-IronPort-AV: E=Sophos;i="5.11,470,1422921600";  d="scan'208,217";a="135502651"
Received: from alln-core-10.cisco.com ([173.36.13.132]) by alln-iport-1.cisco.com with ESMTP; 26 Mar 2015 09:40:59 +0000
Received: from xhc-aln-x09.cisco.com (xhc-aln-x09.cisco.com [173.36.12.83]) by alln-core-10.cisco.com (8.14.5/8.14.5) with ESMTP id t2Q9exQi014779 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 26 Mar 2015 09:40:59 GMT
Received: from xmb-aln-x07.cisco.com ([169.254.2.199]) by xhc-aln-x09.cisco.com ([173.36.12.83]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 04:40:59 -0500
From: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>
To: Haoweiguo <haoweiguo@huawei.com>, John E Drake <jdrake@juniper.net>, "Ali Sajassi (sajassi)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ1HYE/ZJUJbtWUGUXbn7g+eTDJ0ucdUA///v54A=
Date: Thu, 26 Mar 2015 09:40:59 +0000
Message-ID: <D1391B5F.2C9D0%satyamoh@cisco.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.65.53.206]
Content-Type: multipart/alternative; boundary="_000_D1391B5F2C9D0satyamohciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/cAc9CA65PjotQZ84wPkFZEqbN94>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 09:41:03 -0000

--_000_D1391B5F2C9D0satyamohciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Hi Weiguo,

I read your draft.

Your proposal introduces quite a bit of overhead as John has pointed out.
It is indeed bringing the role of AD routes into the DF election which has =
been intentionally kept separate so far.
As John points out, the AD routes will now be advertised ever so often, eve=
n to remote PEs  that do not host the same Ethernet Segment.
The per ES AD route also carries all the Route-Targets of all the EVIs, so =
anytime you want to advertise it again, it will need to be advertised with =
all these Route-Targets.
It imposes quite a processing overhead as far as the extended communities a=
re concerned.


By this proposed feedback mechanism, it is going to delay the DF election.
In section 3 of https://tools.ietf.org/html/draft-hao-bess-evpn-df-handshak=
ing-00 you mention a sequence number but it is not clear how you use it.
How does the scheme work when the PE reboots ?
How will it remember the sequence number?
Its a bit uncooked in the draft.

Best,
--Satya

From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 8:38 PM
To: "jdrake@juniper.net<mailto:jdrake@juniper.net>" <jdrake@juniper.net<mai=
lto:jdrake@juniper.net>>, "Ali Sajassi (sajassi)" <sajassi@cisco.com<mailto=
:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.org>" <bess@ietf.org<=
mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_D1391B5F2C9D0satyamohciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <BC822974D2549D4F8E73DA28EA41DB2E@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif; ">
<div>Hi Weiguo,</div>
<div><br>
</div>
<div>I read your draft.</div>
<div><br>
</div>
<div>Your proposal introduces quite a bit of overhead as John has pointed o=
ut.</div>
<div>It is indeed bringing the role of AD routes into the DF election which=
 has been intentionally kept separate so far.&nbsp;</div>
<div>As John points out, the AD routes will now be advertised ever so often=
, even to remote PEs &nbsp;that do not host the same Ethernet Segment.</div=
>
<div>The per ES AD route also carries all the Route-Targets of all the EVIs=
, so anytime you want to advertise it again, it will need to be advertised =
with all these Route-Targets.</div>
<div>It imposes quite a processing overhead as far as the extended communit=
ies are concerned.</div>
<div><br>
</div>
<div><br>
</div>
<div>By this proposed feedback mechanism, it is going to delay the DF elect=
ion.</div>
<div>In section 3 of&nbsp;<a href=3D"https://tools.ietf.org/html/draft-hao-=
bess-evpn-df-handshaking-00">https://tools.ietf.org/html/draft-hao-bess-evp=
n-df-handshaking-00</a>&nbsp;you mention a sequence number but it is not cl=
ear how you use it.</div>
<div>How does the scheme work when the PE reboots ?</div>
<div>How will it remember the sequence number?</div>
<div>Its a bit uncooked in the draft.</div>
<div><br>
</div>
<div>Best,</div>
<div>--Satya</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Calibri; font-size:11pt; text-align:left; color:b=
lack; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM:=
 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid;=
 BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Haoweiguo &lt;<a href=3D"mail=
to:haoweiguo@huawei.com">haoweiguo@huawei.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Wednesday, March 25, 2015 8:3=
8 PM<br>
<span style=3D"font-weight:bold">To: </span>&quot;<a href=3D"mailto:jdrake@=
juniper.net">jdrake@juniper.net</a>&quot; &lt;<a href=3D"mailto:jdrake@juni=
per.net">jdrake@juniper.net</a>&gt;, &quot;Ali Sajassi (sajassi)&quot; &lt;=
<a href=3D"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;, &quot;<a hr=
ef=3D"mailto:bess@ietf.org">bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org">bess@ietf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>Re: [bess] Handshaking amo=
ng PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<div dir=3D"ltr"><style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.MsoChpDefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple" fpstyle=3D"1" ocsi=3D"0"=
>
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Hi John,<=
/font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">&nbsp;Fir=
stly i think EVPN community should reach consensus on the issues of current=
 DF election mechanism. All these issues should be resolved in a single new=
 DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3"></font></=
font></font>&nbsp;</p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Thanks,</=
font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">weiguo</f=
ont></font></font></p>
<font face=3D"MingLiU"><font face=3D"Calibri"></font></font>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF271558"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g">bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; ">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; "></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; ">Your proposal introduces a control=
 plane processing load that is O(#EVIs * PEs) per DF election and given tha=
t there can be 4K EVIs per ES, this
 looks like a *<b>substantial</b>* load.&nbsp; Furthermore,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; ">you can=92t &nbsp;use the ES route=
 to co-ordinate DF election because you would need to carry your new extend=
ed community for each EVI and they would not
 all fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because =
that is processed by all PEs in the EVI.</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; "></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; ">I think that from a practical pers=
pective the new DF election proposed in Satya=92s draft is sufficiently sta=
ble that it renders your draft moot, even
 if it could be made to work.</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; "></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; ">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; "></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; ">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt; "></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
font-size: 11pt; ">From:</span></b><span style=3D"font-family: Calibri, san=
s-serif; font-size: 11pt; "> BESS [<a href=3D"mailto:bess-bounces@ietf.org"=
>mailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org">bess@iet=
f.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; ">Hi Ali,<=
/span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-si=
ze: 10pt; "></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; ">Thanks f=
or your information. I scanned through this draft, it really introduces int=
er-chassis message for DF election handshaking, the requirements in this dr=
aft is to eliminate transiet Loop
 and traffic duplication. Current EVPN DF election mechanism already elimin=
ated loop and traffic duplication by configuring long reception timer on ea=
ch multi-homed PE, but up to reception timer traffic disruption issue still=
 exist. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"fon=
t-family: Tahoma, sans-serif; color: black; font-size: 10pt; "></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; ">Also for=
 accuracy, i think handshaking state machine on each multi-homed PE is also=
 needed. From solution perspective, in my draft, no inter-chassis message i=
s introduced, only one new extended
 community is introduced, i think the process is comparatively simple than =
your following draft.</span><span style=3D"font-family: Tahoma, sans-serif;=
 color: black; font-size: 10pt; "></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt; "></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; ">Current =
EVPN DF election has some drawbacks, so there are three new drafts about DF=
 election emerged. I think BESS WG can consider these three drafts in globa=
l view,&nbsp; a single,comprehensive new
 DF election draft is hoped.</span><span style=3D"font-family: Tahoma, sans=
-serif; color: black; font-size: 10pt; "></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt; "></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; ">thanks.<=
/span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-si=
ze: 10pt; "></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt; "></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; ">Thanks,<=
/span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-si=
ze: 10pt; "></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; ">weiguo</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt; "></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt; ">From:</span><=
/b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt; "> BESS [<a href=3D"mailto:bess-bounces@ietf.org">bess-bounces@ietf.=
org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com">s=
ajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</span>
</body>
</html>

--_000_D1391B5F2C9D0satyamohciscocom_--


From nobody Thu Mar 26 04:01:11 2015
Return-Path: <wim.henderickx@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 61E201A1B15 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 04:01:09 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id MasBMLb0kDvQ for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 04:01:05 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpgre-esg-01.alcatel-lucent.com [135.245.210.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 080601A0151 for <bess@ietf.org>; Thu, 26 Mar 2015 04:01:04 -0700 (PDT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (unknown [135.239.2.122]) by Websense Email Security Gateway with ESMTPS id 66B20FD2E3825; Thu, 26 Mar 2015 11:01:00 +0000 (GMT)
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id t2QB10iM013996 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 26 Mar 2015 12:01:00 +0100
Received: from FR711WXCHMBA07.zeu.alcatel-lucent.com ([169.254.3.17]) by FR711WXCHHUB01.zeu.alcatel-lucent.com ([135.239.2.111]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 12:01:00 +0100
From: "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
To: Antoni Przygienda <antoni.przygienda@ericsson.com>, Erik Nordmark <nordmark@acm.org>, "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AQHQZ1QBsq1vN6Z2REGnqTtRu+9A/p0tzyWAgABl5AA=
Date: Thu, 26 Mar 2015 11:00:59 +0000
Message-ID: <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se>
In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se>
Accept-Language: nl-BE, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/15.8.1.150311
x-originating-ip: [135.239.27.38]
Content-Type: text/plain; charset="utf-8"
Content-ID: <95D557233C6961429EF679C09F0124E0@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/bxQWDAFw2W9KEc8OIU8QMcx4C7U>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 11:01:09 -0000
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From nobody Thu Mar 26 05:54:30 2015
Return-Path: <pbrisset@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 634DC1B2CEB for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 05:54:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -13.91
X-Spam-Level: 
X-Spam-Status: No, score=-13.91 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id rLjBEgPel2AI for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 05:54:25 -0700 (PDT)
Received: from rcdn-iport-4.cisco.com (rcdn-iport-4.cisco.com [173.37.86.75]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 09C911B2CCE for <bess@ietf.org>; Thu, 26 Mar 2015 05:54:24 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=21263; q=dns/txt; s=iport; t=1427374465; x=1428584065; h=from:to:subject:date:message-id:references:in-reply-to: mime-version; bh=TgjQUzoUMOm91uNxKhIqRZ7ji74uF9DMbLtg8VdSYt0=; b=LBAuWy8i906eUv33JvyTgoT8oZ2OXpBwQPvMz3aPDMnLEex9qJ+KEQ6h 0hVROmvN37D29Pz4EIPcbEkE7Vi1QP9pvuTsXCM/j+jYd3OlW9KfZZUIJ eIvKTgOfxrQtSmly/CMGW4QQrdgbE06nxB1AOIW7a7sCad85opNzH/zFt g=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0C/CQAOABRV/4cNJK1ZA4JDQ1JaBMJugj2FcwKBXUwBAQEBAQF9hBQBAQEEZgMBHQICAQgRAwEBASgHGxcUCQgCBAESFAeIFA2yBZhdAQEBAQEBAQEBAQEBAQEBAQEBAQEBFwSBMYlshBMBEQEkBhUBDAoBBguCaYEzBY5Cgg6Db4YAgRs6gnaMGoNHIoIBHYFQbwGBAwcXIn8BAQE
X-IronPort-AV: E=Sophos;i="5.11,471,1422921600";  d="scan'208,217";a="406913483"
Received: from alln-core-2.cisco.com ([173.36.13.135]) by rcdn-iport-4.cisco.com with ESMTP; 26 Mar 2015 12:54:24 +0000
Received: from xhc-rcd-x03.cisco.com (xhc-rcd-x03.cisco.com [173.37.183.77]) by alln-core-2.cisco.com (8.14.5/8.14.5) with ESMTP id t2QCsNO8030479 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 26 Mar 2015 12:54:24 GMT
Received: from xmb-aln-x09.cisco.com ([169.254.4.43]) by xhc-rcd-x03.cisco.com ([173.37.183.77]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 07:54:23 -0500
From: "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
To: Haoweiguo <haoweiguo@huawei.com>, John E Drake <jdrake@juniper.net>, "Ali Sajassi (sajassi)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8P7HRUda35AYUyDZ9cQKQx3Aw==
Date: Thu, 26 Mar 2015 12:54:22 +0000
Message-ID: <D1396B6C.4EDD3%pbrisset@cisco.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.89.13.188]
Content-Type: multipart/alternative; boundary="_000_D1396B6C4EDD3pbrissetciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/HhEldwD-fS0YllNdaM4Jla5E7sw>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 12:54:28 -0000

--_000_D1396B6C4EDD3pbrissetciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_D1396B6C4EDD3pbrissetciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <885F1EE4F00FCA4AB2675936C5E1B53E@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space;">
<div>
<div>
<div style=3D"color: rgb(0, 0, 0); font-family: Calibri, sans-serif; font-s=
ize: 14px;">
Weiguo,</div>
<div style=3D"color: rgb(0, 0, 0); font-family: Calibri, sans-serif; font-s=
ize: 14px;">
<br>
</div>
<div><font face=3D"Calibri,sans-serif">You mention &quot;</font>But if your=
 draft have not solved all issues=94,</div>
<div>Can you explain what Satya=92s draft is not solving?</div>
<div style=3D"color: rgb(0, 0, 0); font-family: Calibri, sans-serif; font-s=
ize: 14px;">
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"padding-left: 24px; padding-bottom: 0;" align=3D"left" nowrap=
=3D"nowrap" valign=3D"top">
<p style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 11px; col=
or: rgb(102, 102, 102);">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"color:#666666; text-decoration: none;" href=3D"mailto:pbrisset@=
cisco.com">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"padding-left: 20px; padding-bottom: 10px; padding-top: 15px;" =
valign=3D"top" width=3D"50%">
<p style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 11px; col=
or: rgb(102, 102, 102);">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"color:#666666; text-decoration: none;" href=3D"http://www.cisco=
.com/global/CA/">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 10px; co=
lor: #999999; padding-left: 24px; padding-right: 20px; padding-top: 5px; pa=
dding-bottom: 0px; color:#009900; ">
<img src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif" al=
t=3D""> Think before you print.</td>
</tr>
<tr>
<td style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 10px; co=
lor: #999999; padding-left: 24px; padding-right: 20px; padding-top: 7px; pa=
dding-bottom: 6px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a href=3D"http://www.cisco.com/web/about/doing_business/legal/cr=
i/index.html" title=3D"Legal Information" style=3D"color: #0E58A0;">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
</div>
<div style=3D"color: rgb(0, 0, 0); font-family: Calibri, sans-serif; font-s=
ize: 14px;">
<br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION" style=3D"color: rgb(0, 0, 0); font-family=
: Calibri, sans-serif; font-size: 14px;">
<div style=3D"font-family:Calibri; font-size:11pt; text-align:left; color:b=
lack; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM:=
 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid;=
 BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Haoweiguo &lt;<a href=3D"mail=
to:haoweiguo@huawei.com">haoweiguo@huawei.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Wednesday, March 25, 2015 at =
10:38 PM<br>
<span style=3D"font-weight:bold">To: </span>John E Drake &lt;<a href=3D"mai=
lto:jdrake@juniper.net">jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=
=3D"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;, &quot;<a href=3D"m=
ailto:bess@ietf.org">bess@ietf.org</a>&quot; &lt;<a href=3D"mailto:bess@iet=
f.org">bess@ietf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>Re: [bess] Handshaking amo=
ng PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE" style=3D"BORDER-LEFT:=
 #b5c4df 5 solid; PADDING:0 0 0 5; MARGIN:0 0 0 5;">
<div dir=3D"ltr"><style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.MsoChpDefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple" fpstyle=3D"1" ocsi=3D"0"=
>
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Hi John,<=
/font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">&nbsp;Fir=
stly i think EVPN community should reach consensus on the issues of current=
 DF election mechanism. All these issues should be resolved in a single new=
 DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3"></font></=
font></font>&nbsp;</p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Thanks,</=
font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">weiguo</f=
ont></font></font></p>
<font face=3D"MingLiU"><font face=3D"Calibri"></font></font>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF271558"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g">bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Your proposal introduces a control =
plane processing load that is O(#EVIs * PEs) per DF election and given that=
 there can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">you can=92t &nbsp;use the ES route =
to co-ordinate DF election because you would need to carry your new extende=
d community for each EVI and they would not
 all fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because =
that is processed by all PEs in the EVI.</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">I think that from a practical persp=
ective the new DF election proposed in Satya=92s draft is sufficiently stab=
le that it renders your draft moot, even
 if it could be made to work.</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
font-size: 11pt;">From:</span></b><span style=3D"font-family: Calibri, sans=
-serif; font-size: 11pt;"> BESS [<a href=3D"mailto:bess-bounces@ietf.org">m=
ailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org">bess@iet=
f.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Hi Ali,</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Thanks fo=
r your information. I scanned through this draft, it really introduces inte=
r-chassis message for DF election handshaking, the requirements in this dra=
ft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"fon=
t-family: Tahoma, sans-serif; color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Also for =
accuracy, i think handshaking state machine on each multi-homed PE is also =
needed. From solution perspective, in my draft, no inter-chassis message is=
 introduced, only one new extended
 community is introduced, i think the process is comparatively simple than =
your following draft.</span><span style=3D"font-family: Tahoma, sans-serif;=
 color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Current E=
VPN DF election has some drawbacks, so there are three new drafts about DF =
election emerged. I think BESS WG can consider these three drafts in global=
 view,&nbsp; a single,comprehensive new
 DF election draft is hoped.</span><span style=3D"font-family: Tahoma, sans=
-serif; color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">thanks.</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Thanks,</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">weiguo</s=
pan><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size=
: 10pt;"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> BESS [<a href=3D"mailto:bess-bounces@ietf.org">bess-bounces@ietf.or=
g</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com">s=
ajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span>
</body>
</html>

--_000_D1396B6C4EDD3pbrissetciscocom_--


From nobody Thu Mar 26 08:13:12 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 648C61A036B for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 08:13:11 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.61
X-Spam-Level: 
X-Spam-Status: No, score=-3.61 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9lZqXsCaUbxg for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 08:13:07 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E87A81A01FA for <bess@ietf.org>; Thu, 26 Mar 2015 08:13:06 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQS91754; Thu, 26 Mar 2015 15:13:05 +0000 (GMT)
Received: from nkgeml405-hub.china.huawei.com (10.98.56.36) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Thu, 26 Mar 2015 15:13:03 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml405-hub.china.huawei.com ([10.98.56.36]) with mapi id 14.03.0158.001; Thu, 26 Mar 2015 23:12:58 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, John E Drake <jdrake@juniper.net>, "Ali Sajassi (sajassi)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ6j80pJncGbUN0++hT+ovUCRX50u0Xii
Date: Thu, 26 Mar 2015 15:12:57 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F8573B9@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com>, <D1391B5F.2C9D0%satyamoh@cisco.com>
In-Reply-To: <D1391B5F.2C9D0%satyamoh@cisco.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.145.49]
Content-Type: multipart/alternative; boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F8573B9nkgeml501mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/0xdPHe7xN-UwtfLcIkPJCseGlK4>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 15:13:11 -0000

--_000_DD5FC8DE455C3348B94340C0AB5517334F8573B9nkgeml501mbschi_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Hi Satya,

Pls see below with [weiguo]. Further discussions are welcome.



Thanks,

weiguo

________________________________
From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
Sent: Thursday, March 26, 2015 17:40
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Hi Weiguo,

I read your draft.

Your proposal introduces quite a bit of overhead as John has pointed out.
[weiguo]; More AD routes advertisements overhead is introduced,but i think =
it's not so much. I think the benefit deserves for the overhead. It can eff=
ectively reduce traffic disruption during transiet time.

It is indeed bringing the role of AD routes into the DF election which has =
been intentionally kept separate so far.
[weiguo]: Yes, AD routes are re-used for DF election handshaking process. A=
s draft draft-rabadan-bess-evpn-ac-df-01 points out, DF election on each mu=
lti-homed PE also should consider AD route, not just ES route.

As John points out, the AD routes will now be advertised ever so often, eve=
n to remote PEs  that do not host the same Ethernet Segment.
[weiguo]: Not so often, only in the DF election and re-election phase, one =
more copy of AD routes per PE are advertised.

The per ES AD route also carries all the Route-Targets of all the EVIs, so =
anytime you want to advertise it again, it will need to be advertised with =
all these Route-Targets.
It imposes quite a processing overhead as far as the extended communities a=
re concerned.
[weiguo]: Only when VLAN configuration change, PE joins or leaves ESI and e=
tc will trigger the DF election and re-election, in normal operating envion=
ment, these events will not happen frequently.

By this proposed feedback mechanism, it is going to delay the DF election.
[weiguo]: Yes, to ensure that all PEs of one ESI take action consistently. =
If DF PE and non-DF PE take action independantly, there will be some time t=
raffic disruption. Handshaking mechanism is common in native ethernet world=
, such as MSTP. EVPN access network belongs to native ethernet world, it ne=
eds a more accurate mechanism.

In section 3 of https://tools.ietf.org/html/draft-hao-bess-evpn-df-handshak=
ing-00 you mention a sequence number but it is not clear how you use it.
[weiguo]:  The seq number is generated at DF PE. When a PE is elected as DF=
 PE, it should generate a seq number. Each handshaking message will use the=
 seq number to ensure it is for same DF election process , it is used to av=
oid message interaction disorder.
Assuming PE1 is DF PE, it generates seq number 5, PE2 and PE3 use seq 5 to =
communicate with PE1. Before the handshaking negotication ends, another PE4=
 joins same ESI, PE1 is still DF, so PE1 wants to start a new handshaking p=
rocess, it needs to generate a new seq number 6 to avoid message interactio=
n confusing. I think the seq number mechanism can enhance reliability. If y=
ou and the comunity experts have better mechanism, i can also change it.

How does the scheme work when the PE reboots ?
[weiguo]: When PE reboots,  the process:
1. Local ESI discovery
2.Multi-homed PE auto-discovery
3.When reception timer expires, DF election handshaking.
4. DF PE and non-DF PE take effect, and enters into stable state.

How will it remember the sequence number?
[weiguo]: Don't need remember the seq. When PE restarts, the seq is generat=
ed from the start. Normally a device reboot timer needs several min.

Its a bit uncooked in the draft.

Best,
--Satya

From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 8:38 PM
To: "jdrake@juniper.net<mailto:jdrake@juniper.net>" <jdrake@juniper.net<mai=
lto:jdrake@juniper.net>>, "Ali Sajassi (sajassi)" <sajassi@cisco.com<mailto=
:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.org>" <bess@ietf.org<=
mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_DD5FC8DE455C3348B94340C0AB5517334F8573B9nkgeml501mbschi_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"FONT-FAMILY: Calibri,sans-serif; WORD-WRAP: break-word; COLO=
R: rgb(0,0,0); FONT-SIZE: 14px" fPStyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3">Hi Satya,</font></p>
<p><font size=3D"3">Pls see below with [weiguo]. Further discussions are we=
lcome.</font></p>
<p><font size=3D"3"></font>&nbsp;</p>
<p><font size=3D"3">Thanks,</font></p>
<p><font size=3D"3">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF748017"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> Satya Mohanty (satyamoh) [satyamoh@c=
isco.com]<br>
<b>Sent:</b> Thursday, March 26, 2015 17:40<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<br=
>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>Hi Weiguo,</div>
<div><br>
</div>
<div>I read your draft.</div>
<div><br>
</div>
<div>Your proposal introduces quite a bit of overhead as John has pointed o=
ut.</div>
<div>[weiguo]; More AD routes advertisements overhead is introduced,but i t=
hink it's not so much. I think the benefit deserves for the overhead. It ca=
n effectively reduce traffic disruption during transiet time.</div>
<div>&nbsp;</div>
<div>It is indeed bringing the role of AD routes into the DF election which=
 has been intentionally kept separate so far.&nbsp;</div>
<div>[weiguo]: Yes, AD routes are re-used for DF election handshaking proce=
ss. As draft
<font face=3D"Calibri">draft-rabadan-bess-evpn-ac-df-01 points out, DF elec=
tion on each multi-homed PE also should consider AD route, not just ES rout=
e.</font></div>
<div>&nbsp;</div>
<div>As John points out, the AD routes will now be advertised ever so often=
, even to remote PEs &nbsp;that do not host the same Ethernet Segment.</div=
>
<div>[weiguo]: Not so often, only in the DF election and re-election phase,=
 one more copy of AD routes per PE are advertised.</div>
<div>&nbsp;</div>
<div>The per ES AD route also carries all the Route-Targets of all the EVIs=
, so anytime you want to advertise it again, it will need to be advertised =
with all these Route-Targets.</div>
<div>It imposes quite a processing overhead as far as the extended communit=
ies are concerned.</div>
<div>[weiguo]: Only when VLAN configuration change, PE joins or leaves ESI =
and etc will trigger the DF election and re-election, in normal operating e=
nvionment, these events will not happen frequently.<br>
</div>
<div><br>
</div>
<div>By this proposed feedback mechanism, it is going to delay the DF elect=
ion.</div>
<div>[weiguo]: Yes, to ensure that all PEs of one ESI take action consisten=
tly. If DF PE and non-DF PE take action independantly, there will be some t=
ime traffic disruption. Handshaking mechanism is common in native ethernet =
world, such as MSTP. EVPN access
 network belongs to native ethernet world, it needs a more accurate mechani=
sm.</div>
<div>&nbsp;</div>
<div>In section 3 of&nbsp;<a href=3D"https://tools.ietf.org/html/draft-hao-=
bess-evpn-df-handshaking-00" target=3D"_blank">https://tools.ietf.org/html/=
draft-hao-bess-evpn-df-handshaking-00</a>&nbsp;you mention a sequence numbe=
r but it is not clear how you use it.</div>
<div>[weiguo]:&nbsp;&nbsp;The seq number is generated at DF PE. When a PE i=
s elected as DF PE, it should generate a seq number. Each handshaking messa=
ge will use the seq number to ensure it is for same DF election process , i=
t is used to avoid message interaction disorder.</div>
<div>Assuming PE1 is DF PE, it generates seq number 5, PE2 and PE3 use seq =
5 to communicate with PE1. Before the handshaking negotication ends, anothe=
r PE4 joins same ESI, PE1 is still DF, so PE1 wants to start a new handshak=
ing process, it needs to generate
 a new seq number 6&nbsp;to avoid message interaction confusing. I think th=
e seq number mechanism can enhance reliability. If you and the comunity exp=
erts have better mechanism, i can also change it.</div>
<div>&nbsp;</div>
<div>How does the scheme work when the PE reboots ?</div>
<div>[weiguo]: When PE reboots,&nbsp; the process:</div>
<div>1. Local ESI discovery</div>
<div>2.Multi-homed PE auto-discovery</div>
<div>3.When reception timer expires, DF election handshaking.</div>
<div>4. DF PE and non-DF PE take effect, and enters into stable state.</div=
>
<div>&nbsp;</div>
<div>How will it remember the sequence number?</div>
<div>[weiguo]: Don't need remember the seq. When PE restarts, the seq is ge=
nerated from the start. Normally a device reboot timer needs several min.</=
div>
<div>&nbsp;</div>
<div>Its a bit uncooked in the draft.</div>
<div><br>
</div>
<div>Best,</div>
<div>--Satya</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Wednesday, March 25, 2015 8:=
38 PM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>&quot;<a href=3D"mailto:jdrake=
@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&quot; &lt;<a href=3D=
"mailto:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, &=
quot;Ali Sajassi (sajassi)&quot; &lt;<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>Re: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<div dir=3D"ltr"><style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.MsoChpDefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Hi John,<=
/font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">&nbsp;Fir=
stly i think EVPN community should reach consensus on the issues of current=
 DF election mechanism. All these issues should be resolved in a single new=
 DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3"></font></=
font></font>&nbsp;</p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Thanks,</=
font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">weiguo</f=
ont></font></font></p>
<font face=3D"MingLiU"><font face=3D"Calibri"></font></font>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF271558"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net" target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Your proposal introduces a control plan=
e processing load that is O(#EVIs * PEs) per DF election and given that the=
re can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">you can=92t &nbsp;use the ES route to c=
o-ordinate DF election because you would need to carry your new extended co=
mmunity for each EVI and they would not all
 fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because that=
 is processed by all PEs in the EVI.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">I think that from a practical perspecti=
ve the new DF election proposed in Satya=92s draft is sufficiently stable t=
hat it renders your draft moot, even if
 it could be made to work.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: Calibri,sans-serif; F=
ONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY: Calibri,sans-se=
rif; FONT-SIZE: 11pt"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" targe=
t=3D"_blank">mailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Hi Ali,</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks for =
your information. I scanned through this draft, it really introduces inter-=
chassis message for DF election handshaking, the requirements in this draft=
 is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"FON=
T-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Also for ac=
curacy, i think handshaking state machine on each multi-homed PE is also ne=
eded. From solution perspective, in my draft, no inter-chassis message is i=
ntroduced, only one new extended community
 is introduced, i think the process is comparatively simple than your follo=
wing draft.</span><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: bla=
ck; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Current EVP=
N DF election has some drawbacks, so there are three new drafts about DF el=
ection emerged. I think BESS WG can consider these three drafts in global v=
iew,&nbsp; a single,comprehensive new DF
 election draft is hoped.</span><span style=3D"FONT-FAMILY: Tahoma,sans-ser=
if; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">thanks.</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks,</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">weiguo</spa=
n><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 1=
0pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></b>=
<span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10p=
t"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">bess-b=
ounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</span></div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F8573B9nkgeml501mbschi_--


From nobody Thu Mar 26 08:24:40 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DEA501AD0A5 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 08:24:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.61
X-Spam-Level: 
X-Spam-Status: No, score=-3.61 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id G1D175yoWgTF for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 08:24:36 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 261201A86FE for <bess@ietf.org>; Thu, 26 Mar 2015 08:24:35 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml402-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQS92836; Thu, 26 Mar 2015 15:24:33 +0000 (GMT)
Received: from NKGEML403-HUB.china.huawei.com (10.98.56.34) by lhreml402-hub.china.huawei.com (10.201.5.241) with Microsoft SMTP Server (TLS) id 14.3.158.1; Thu, 26 Mar 2015 15:24:32 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml403-hub.china.huawei.com ([10.98.56.34]) with mapi id 14.03.0158.001; Thu, 26 Mar 2015 23:24:26 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>, John E Drake <jdrake@juniper.net>, "Ali Sajassi (sajassi)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F
Date: Thu, 26 Mar 2015 15:24:26 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com>, <D1396B6C.4EDD3%pbrisset@cisco.com>
In-Reply-To: <D1396B6C.4EDD3%pbrisset@cisco.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.145.49]
Content-Type: multipart/alternative; boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F8573CCnkgeml501mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/ezRz1ztRUWtYaTRx-gtOWQX7G3M>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 15:24:39 -0000

--_000_DD5FC8DE455C3348B94340C0AB5517334F8573CCnkgeml501mbschi_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_DD5FC8DE455C3348B94340C0AB5517334F8573CCnkgeml501mbschi_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"WORD-WRAP: break-word" fPStyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3" face=3D"Calibri">Hi Patrice,</font></p>
<p><font size=3D"3" face=3D"Calibri">Up to reception timer traffic disrupti=
on in transient phase is one of the issues.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF422254"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> Patrice Brissette (pbrisset) [pbriss=
et@cisco.com]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<br=
>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>
<div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
Weiguo,</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<div><font face=3D"Calibri,sans-serif">You mention &quot;</font>But if your=
 draft have not solved all issues=94,</div>
<div>Can you explain what Satya=92s draft is not solving?</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZ=
E: 14px" id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Wednesday, March 25, 2015 at=
 10:38 PM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>John E Drake &lt;<a href=3D"ma=
ilto:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali =
Sajassi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank"=
>bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>Re: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr"><style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.MsoChpDefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Hi John,<=
/font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">&nbsp;Fir=
stly i think EVPN community should reach consensus on the issues of current=
 DF election mechanism. All these issues should be resolved in a single new=
 DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3"></font></=
font></font>&nbsp;</p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Thanks,</=
font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">weiguo</f=
ont></font></font></p>
<font face=3D"MingLiU"><font face=3D"Calibri"></font></font>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF271558"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net" target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Your proposal introduces a control plan=
e processing load that is O(#EVIs * PEs) per DF election and given that the=
re can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">you can=92t &nbsp;use the ES route to c=
o-ordinate DF election because you would need to carry your new extended co=
mmunity for each EVI and they would not all
 fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because that=
 is processed by all PEs in the EVI.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">I think that from a practical perspecti=
ve the new DF election proposed in Satya=92s draft is sufficiently stable t=
hat it renders your draft moot, even if
 it could be made to work.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: Calibri,sans-serif; F=
ONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY: Calibri,sans-se=
rif; FONT-SIZE: 11pt"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" targe=
t=3D"_blank">mailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Hi Ali,</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks for =
your information. I scanned through this draft, it really introduces inter-=
chassis message for DF election handshaking, the requirements in this draft=
 is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"FON=
T-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Also for ac=
curacy, i think handshaking state machine on each multi-homed PE is also ne=
eded. From solution perspective, in my draft, no inter-chassis message is i=
ntroduced, only one new extended community
 is introduced, i think the process is comparatively simple than your follo=
wing draft.</span><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: bla=
ck; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Current EVP=
N DF election has some drawbacks, so there are three new drafts about DF el=
ection emerged. I think BESS WG can consider these three drafts in global v=
iew,&nbsp; a single,comprehensive new DF
 election draft is hoped.</span><span style=3D"FONT-FAMILY: Tahoma,sans-ser=
if; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">thanks.</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks,</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">weiguo</spa=
n><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 1=
0pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></b>=
<span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10p=
t"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">bess-b=
ounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F8573CCnkgeml501mbschi_--


From nobody Thu Mar 26 08:34:32 2015
Return-Path: <pbrisset@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6A6D21A1BA3 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 08:34:31 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -13.91
X-Spam-Level: 
X-Spam-Status: No, score=-13.91 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7EAlOEb9upv4 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 08:34:28 -0700 (PDT)
Received: from rcdn-iport-9.cisco.com (rcdn-iport-9.cisco.com [173.37.86.80]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 132901A7D82 for <bess@ietf.org>; Thu, 26 Mar 2015 08:34:28 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=29152; q=dns/txt; s=iport; t=1427384068; x=1428593668; h=from:to:subject:date:message-id:references:in-reply-to: mime-version; bh=AAIkTo17w5R+Iy/Lq47NncYyd7nJlL8dJAuKX15g8Is=; b=F+rxXUdHxr3hViuHgV4MxqUay98eM65aDMvt+y/C89XIXzN3Fw2U+ruS YJKdULza2k6sXd3n0QQBrFa6Re2hEe1rYVmNHDhbLCGTiqEpk9O8mVq5p lNI3PZPcnbVjV8kRc4SjgX5IHsT5LkdNgQRD0YNKAbtKVPlephMxeWUQK w=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0A1CACAJhRV/4wNJK1ZA4JDQ1JaBLU2jTmCPYVzAoFYTAEBAQEBAX2EFAEBAQRmAwEdAgIBCBEDAQEBIQcHGxcUCQgCBAESFAeIFA2ySZhKAQEBAQEBAQEBAQEBAQEBAQEBAQEBFwSBMYlzhBUBEQEkBhQBAQYGCgEGC4JpgTMFjkKCDoNvhgCBGzqCdowbg0ciggEdgVBvAYEDBxcifwEBAQ
X-IronPort-AV: E=Sophos;i="5.11,472,1422921600";  d="scan'208,217";a="403777673"
Received: from alln-core-7.cisco.com ([173.36.13.140]) by rcdn-iport-9.cisco.com with ESMTP; 26 Mar 2015 15:34:27 +0000
Received: from xhc-aln-x14.cisco.com (xhc-aln-x14.cisco.com [173.36.12.88]) by alln-core-7.cisco.com (8.14.5/8.14.5) with ESMTP id t2QFYQ6T024075 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 26 Mar 2015 15:34:26 GMT
Received: from xmb-aln-x09.cisco.com ([169.254.4.43]) by xhc-aln-x14.cisco.com ([173.36.12.88]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 10:34:26 -0500
From: "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
To: Haoweiguo <haoweiguo@huawei.com>, John E Drake <jdrake@juniper.net>, "Ali Sajassi (sajassi)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ9pX3YIyCBnGV069NWaTut3WMg==
Date: Thu, 26 Mar 2015 15:34:25 +0000
Message-ID: <D13990E2.4EE3E%pbrisset@cisco.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.89.7.222]
Content-Type: multipart/alternative; boundary="_000_D13990E24EE3Epbrissetciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/ZdCRPfMvj8-nkWosRK3SXWPpCtY>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 15:34:31 -0000

--_000_D13990E24EE3Epbrissetciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Weiguo,

I=92m not sure I=92m following here.
Don=92t you have the same issue with your handshaking mechanism?
If you don=92t know your peer, how can you handshake?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_D13990E24EE3Epbrissetciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <8F2F1D2CD40F0A44AA5A87BA60E410C2@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div>
<div>Weiguo,</div>
<div><br>
</div>
<div>I=92m not sure I=92m following here.</div>
<div>Don=92t you have the same issue with your handshaking mechanism?&nbsp;=
</div>
<div>If you don=92t know your peer, how can you handshake?</div>
<div>
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"padding-left: 24px; padding-bottom: 0;" align=3D"left" nowrap=
=3D"nowrap" valign=3D"top">
<p style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 11px; col=
or: rgb(102, 102, 102);">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"color:#666666; text-decoration: none;" href=3D"mailto:pbrisset@=
cisco.com">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"padding-left: 20px; padding-bottom: 10px; padding-top: 15px;" =
valign=3D"top" width=3D"50%">
<p style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 11px; col=
or: rgb(102, 102, 102);">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"color:#666666; text-decoration: none;" href=3D"http://www.cisco=
.com/global/CA/">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 10px; co=
lor: #999999; padding-left: 24px; padding-right: 20px; padding-top: 5px; pa=
dding-bottom: 0px; color:#009900; ">
<img src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif" al=
t=3D""> Think before you print.</td>
</tr>
<tr>
<td style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 10px; co=
lor: #999999; padding-left: 24px; padding-right: 20px; padding-top: 7px; pa=
dding-bottom: 6px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a href=3D"http://www.cisco.com/web/about/doing_business/legal/cr=
i/index.html" title=3D"Legal Information" style=3D"color: #0E58A0;">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Calibri; font-size:11pt; text-align:left; color:b=
lack; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM:=
 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid;=
 BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Haoweiguo &lt;<a href=3D"mail=
to:haoweiguo@huawei.com">haoweiguo@huawei.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Thursday, March 26, 2015 at 1=
0:24 AM<br>
<span style=3D"font-weight:bold">To: </span>Patrice Brissette &lt;<a href=
=3D"mailto:pbrisset@cisco.com">pbrisset@cisco.com</a>&gt;, John E Drake &lt=
;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net</a>&gt;, Ali Saja=
ssi &lt;<a href=3D"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org">bess@ietf.org</a>&quot; &lt;<a href=
=3D"mailto:bess@ietf.org">bess@ietf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>RE: [bess] Handshaking amo=
ng PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE" style=3D"BORDER-LEFT:=
 #b5c4df 5 solid; PADDING:0 0 0 5; MARGIN:0 0 0 5;">
<div dir=3D"ltr">
<div style=3D"WORD-WRAP: break-word" fpstyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3" face=3D"Calibri">Hi Patrice,</font></p>
<p><font size=3D"3" face=3D"Calibri">Up to reception timer traffic disrupti=
on in transient phase is one of the issues.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF422254"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> Patrice Brissette (pbrisset) [<a hre=
f=3D"mailto:pbrisset@cisco.com">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>
<div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
Weiguo,</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<div><font face=3D"Calibri,sans-serif">You mention &quot;</font>But if your=
 draft have not solved all issues=94,</div>
<div>Can you explain what Satya=92s draft is not solving?</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<span style=3D"font-family: Calibri, sans-serif; color: rgb(0, 0, 0); font-=
size: 14px;" id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Wednesday, March 25, 2015 at=
 10:38 PM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>John E Drake &lt;<a href=3D"ma=
ilto:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali =
Sajassi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank"=
>bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>Re: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr"><style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.MsoChpDefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Hi John,<=
/font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">&nbsp;Fir=
stly i think EVPN community should reach consensus on the issues of current=
 DF election mechanism. All these issues should be resolved in a single new=
 DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3"></font></=
font></font>&nbsp;</p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Thanks,</=
font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">weiguo</f=
ont></font></font></p>
<font face=3D"MingLiU"><font face=3D"Calibri"></font></font>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF271558"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net" target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Your proposal introduces a control =
plane processing load that is O(#EVIs * PEs) per DF election and given that=
 there can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">you can=92t &nbsp;use the ES route =
to co-ordinate DF election because you would need to carry your new extende=
d community for each EVI and they would not
 all fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because =
that is processed by all PEs in the EVI.</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">I think that from a practical persp=
ective the new DF election proposed in Satya=92s draft is sufficiently stab=
le that it renders your draft moot, even
 if it could be made to work.</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
font-size: 11pt;">From:</span></b><span style=3D"font-family: Calibri, sans=
-serif; font-size: 11pt;"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" t=
arget=3D"_blank">mailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Hi Ali,</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Thanks fo=
r your information. I scanned through this draft, it really introduces inte=
r-chassis message for DF election handshaking, the requirements in this dra=
ft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"fon=
t-family: Tahoma, sans-serif; color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Also for =
accuracy, i think handshaking state machine on each multi-homed PE is also =
needed. From solution perspective, in my draft, no inter-chassis message is=
 introduced, only one new extended
 community is introduced, i think the process is comparatively simple than =
your following draft.</span><span style=3D"font-family: Tahoma, sans-serif;=
 color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Current E=
VPN DF election has some drawbacks, so there are three new drafts about DF =
election emerged. I think BESS WG can consider these three drafts in global=
 view,&nbsp; a single,comprehensive new
 DF election draft is hoped.</span><span style=3D"font-family: Tahoma, sans=
-serif; color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">thanks.</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Thanks,</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">weiguo</s=
pan><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size=
: 10pt;"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">be=
ss-bounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</div>
</div>
</div>
</div>
</blockquote>
</span>
</body>
</html>

--_000_D13990E24EE3Epbrissetciscocom_--


From nobody Thu Mar 26 10:28:54 2015
Return-Path: <jdrake@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 92BE51A8998 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 10:28:52 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.301
X-Spam-Level: 
X-Spam-Status: No, score=-1.301 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id prt2FlWoNI6H for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 10:28:49 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1bon0732.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::1:732]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3C2291A877B for <bess@ietf.org>; Thu, 26 Mar 2015 10:28:49 -0700 (PDT)
Received: from BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) by BLUPR05MB564.namprd05.prod.outlook.com (10.141.202.150) with Microsoft SMTP Server (TLS) id 15.1.118.21; Thu, 26 Mar 2015 17:28:30 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) by BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) with mapi id 15.01.0118.022; Thu, 26 Mar 2015 17:28:30 +0000
From: John E Drake <jdrake@juniper.net>
To: "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8P/o146fJTTukSNC2GolBinQZ0u4lsAgAACyoCAAB/gjQ==
Date: Thu, 26 Mar 2015 17:28:30 +0000
Message-ID: <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>
In-Reply-To: <D13990E2.4EE3E%pbrisset@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [166.173.61.40]
authentication-results: cisco.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BLUPR05MB564;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(36304003)(69234005)(24454002)(164054003)(50854003)(377454003)(46102003)(82746002)(110136001)(2950100001)(19580405001)(19580395003)(2656002)(15975445007)(102836002)(87936001)(19622755009)(77096005)(83716003)(54356999)(99286002)(66066001)(2900100001)(33656002)(76176999)(50986999)(16236675004)(40100003)(93886004)(18206015028)(106116001)(19617315012)(19273905006)(77156002)(122556002)(62966003)(19300405004)(92566002)(36756003)(86362001)(561944003)(104396002); DIR:OUT; SFP:1102; SCL:1; SRVR:BLUPR05MB564; H:BLUPR05MB562.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BLUPR05MB564CA71F82A8BF91070BA9BC7080@BLUPR05MB564.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BLUPR05MB564; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB564; 
x-forefront-prvs: 0527DFA348
Content-Type: multipart/alternative; boundary="_000_FB7DD2A6C2054F9DA70BA73A564E1F99junipernet_"
MIME-Version: 1.0
X-OriginatorOrg: juniper.net
X-MS-Exchange-CrossTenant-originalarrivaltime: 26 Mar 2015 17:28:30.5894 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BLUPR05MB564
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/jiF-Nv4B89SpFMSXS_UDoro1Ujw>
Cc: Haoweiguo <haoweiguo@huawei.com>, "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 17:28:52 -0000

--_000_FB7DD2A6C2054F9DA70BA73A564E1F99junipernet_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:

Weiguo,

I=92m not sure I=92m following here.
Don=92t you have the same issue with your handshaking mechanism?
If you don=92t know your peer, how can you handshake?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_FB7DD2A6C2054F9DA70BA73A564E1F99junipernet_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body dir=3D"auto">
<div>I think Patrice is correct. &nbsp;Your proposal doesn't solve the prob=
lem and it does so at huge cost.<br>
<br>
Sent from my iPhone</div>
<div><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com">pbrisset@cisco.com</a>&gt; wrote:<br>
<br>
</div>
<blockquote type=3D"cite">
<div>
<div>
<div>Weiguo,</div>
<div><br>
</div>
<div>I=92m not sure I=92m following here.</div>
<div>Don=92t you have the same issue with your handshaking mechanism?&nbsp;=
</div>
<div>If you don=92t know your peer, how can you handshake?</div>
<div>
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"padding-left: 24px; padding-bottom: 0;" align=3D"left" nowrap=
=3D"nowrap" valign=3D"top">
<p style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 11px; col=
or: rgb(102, 102, 102);">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"color:#666666; text-decoration: none;" href=3D"mailto:pbrisset@=
cisco.com">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"padding-left: 20px; padding-bottom: 10px; padding-top: 15px;" =
valign=3D"top" width=3D"50%">
<p style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 11px; col=
or: rgb(102, 102, 102);">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"color:#666666; text-decoration: none;" href=3D"http://www.cisco=
.com/global/CA/">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 10px; co=
lor: #999999; padding-left: 24px; padding-right: 20px; padding-top: 5px; pa=
dding-bottom: 0px; color:#009900; ">
<img src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif" al=
t=3D""> Think before you print.</td>
</tr>
<tr>
<td style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 10px; co=
lor: #999999; padding-left: 24px; padding-right: 20px; padding-top: 7px; pa=
dding-bottom: 6px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a href=3D"http://www.cisco.com/web/about/doing_business/legal/cr=
i/index.html" title=3D"Legal Information" style=3D"color: #0E58A0;">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Calibri; font-size:11pt; text-align:left; color:b=
lack; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM:=
 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid;=
 BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Haoweiguo &lt;<a href=3D"mail=
to:haoweiguo@huawei.com">haoweiguo@huawei.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Thursday, March 26, 2015 at 1=
0:24 AM<br>
<span style=3D"font-weight:bold">To: </span>Patrice Brissette &lt;<a href=
=3D"mailto:pbrisset@cisco.com">pbrisset@cisco.com</a>&gt;, John E Drake &lt=
;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net</a>&gt;, Ali Saja=
ssi &lt;<a href=3D"mailto:sajassi@cisco.com">sajassi@cisco.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org">bess@ietf.org</a>&quot; &lt;<a href=
=3D"mailto:bess@ietf.org">bess@ietf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>RE: [bess] Handshaking amo=
ng PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE" style=3D"BORDER-LEFT:=
 #b5c4df 5 solid; PADDING:0 0 0 5; MARGIN:0 0 0 5;">
<div dir=3D"ltr">
<div style=3D"WORD-WRAP: break-word" fpstyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3" face=3D"Calibri">Hi Patrice,</font></p>
<p><font size=3D"3" face=3D"Calibri">Up to reception timer traffic disrupti=
on in transient phase is one of the issues.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF422254"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> Patrice Brissette (pbrisset) [<a hre=
f=3D"mailto:pbrisset@cisco.com">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>
<div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
Weiguo,</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<div><font face=3D"Calibri,sans-serif">You mention &quot;</font>But if your=
 draft have not solved all issues=94,</div>
<div>Can you explain what Satya=92s draft is not solving?</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<span style=3D"font-family: Calibri, sans-serif; color: rgb(0, 0, 0); font-=
size: 14px;" id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Wednesday, March 25, 2015 at=
 10:38 PM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>John E Drake &lt;<a href=3D"ma=
ilto:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali =
Sajassi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank"=
>bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>Re: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr"><style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.MsoChpDefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Hi John,<=
/font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">&nbsp;Fir=
stly i think EVPN community should reach consensus on the issues of current=
 DF election mechanism. All these issues should be resolved in a single new=
 DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3"></font></=
font></font>&nbsp;</p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Thanks,</=
font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">weiguo</f=
ont></font></font></p>
<font face=3D"MingLiU"><font face=3D"Calibri"></font></font>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF271558"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net" target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Your proposal introduces a control =
plane processing load that is O(#EVIs * PEs) per DF election and given that=
 there can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">you can=92t &nbsp;use the ES route =
to co-ordinate DF election because you would need to carry your new extende=
d community for each EVI and they would not
 all fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because =
that is processed by all PEs in the EVI.</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">I think that from a practical persp=
ective the new DF election proposed in Satya=92s draft is sufficiently stab=
le that it renders your draft moot, even
 if it could be made to work.</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
font-size: 11pt;">From:</span></b><span style=3D"font-family: Calibri, sans=
-serif; font-size: 11pt;"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" t=
arget=3D"_blank">mailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Hi Ali,</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Thanks fo=
r your information. I scanned through this draft, it really introduces inte=
r-chassis message for DF election handshaking, the requirements in this dra=
ft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"fon=
t-family: Tahoma, sans-serif; color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Also for =
accuracy, i think handshaking state machine on each multi-homed PE is also =
needed. From solution perspective, in my draft, no inter-chassis message is=
 introduced, only one new extended
 community is introduced, i think the process is comparatively simple than =
your following draft.</span><span style=3D"font-family: Tahoma, sans-serif;=
 color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Current E=
VPN DF election has some drawbacks, so there are three new drafts about DF =
election emerged. I think BESS WG can consider these three drafts in global=
 view,&nbsp; a single,comprehensive new
 DF election draft is hoped.</span><span style=3D"font-family: Tahoma, sans=
-serif; color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">thanks.</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Thanks,</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">weiguo</s=
pan><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size=
: 10pt;"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">be=
ss-bounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</blockquote>
</body>
</html>

--_000_FB7DD2A6C2054F9DA70BA73A564E1F99junipernet_--


From nobody Thu Mar 26 12:24:12 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B56611B2D99 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 12:24:10 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.61
X-Spam-Level: 
X-Spam-Status: No, score=-3.61 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JHc-OzztGB2D for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 12:24:07 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9859B1B2B12 for <bess@ietf.org>; Thu, 26 Mar 2015 12:24:06 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml405-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BUE29718; Thu, 26 Mar 2015 19:24:04 +0000 (GMT)
Received: from nkgeml405-hub.china.huawei.com (10.98.56.36) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.3.158.1; Thu, 26 Mar 2015 19:24:03 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml405-hub.china.huawei.com ([10.98.56.36]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 03:23:58 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>, John E Drake <jdrake@juniper.net>, "Ali Sajassi (sajassi)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAMR4Vg==
Date: Thu, 26 Mar 2015 19:23:57 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857476@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>
In-Reply-To: <D13990E2.4EE3E%pbrisset@cisco.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.152.182]
Content-Type: multipart/alternative; boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F857476nkgeml501mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/npOPWApEHsDQ97W9txe2ijhLeFE>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 19:24:10 -0000

--_000_DD5FC8DE455C3348B94340C0AB5517334F857476nkgeml501mbschi_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Handshaking process happens after PEs auto-discovery using ES route,  it ca=
n be used to avoid traffic disruption. Handshaking mechanism should be comb=
ined with PEs auto-discovery defined in current EVPN base protocol.



weiguo

________________________________
From: BESS [bess-bounces@ietf.org] on behalf of Patrice Brissette (pbrisset=
) [pbrisset@cisco.com]
Sent: Thursday, March 26, 2015 23:34
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

I=92m not sure I=92m following here.
Don=92t you have the same issue with your handshaking mechanism?
If you don=92t know your peer, how can you handshake?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_DD5FC8DE455C3348B94340C0AB5517334F857476nkgeml501mbschi_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"FONT-FAMILY: Calibri,sans-serif; WORD-WRAP: break-word; COLO=
R: rgb(0,0,0); FONT-SIZE: 14px" fPStyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3">Handshaking process happens after PEs auto-discovery us=
ing ES route,&nbsp;
</font><font size=3D"3">it can be used to avoid traffic disruption. Handsha=
king mechanism should be combined with PEs auto-discovery defined in curren=
t EVPN base protocol.</font></p>
<p><font size=3D"3"></font>&nbsp;</p>
<p><font size=3D"3">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF958960"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> BESS [bess-bounces@ietf.org] on beha=
lf of Patrice Brissette (pbrisset) [pbrisset@cisco.com]<br>
<b>Sent:</b> Thursday, March 26, 2015 23:34<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<br=
>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>
<div>Weiguo,</div>
<div><br>
</div>
<div>I=92m not sure I=92m following here.</div>
<div>Don=92t you have the same issue with your handshaking mechanism?&nbsp;=
</div>
<div>If you don=92t know your peer, how can you handshake?</div>
<div>
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Thursday, March 26, 2015 at =
10:24 AM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>Patrice Brissette &lt;<a href=
=3D"mailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt;=
, John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D"_blank">=
jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:sajassi@cisco=
.com" target=3D"_blank">sajassi@cisco.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>RE: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr">
<div style=3D"WORD-WRAP: break-word">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font size=3D"3" face=3D"Calibri">Hi Patrice,</font></p>
<p><font size=3D"3" face=3D"Calibri">Up to reception timer traffic disrupti=
on in transient phase is one of the issues.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF422254"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> Patrice Brissette (pbrisset) [<a hre=
f=3D"mailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>]<b=
r>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>
<div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
Weiguo,</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<div><font face=3D"Calibri,sans-serif">You mention &quot;</font>But if your=
 draft have not solved all issues=94,</div>
<div>Can you explain what Satya=92s draft is not solving?</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZ=
E: 14px" id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Wednesday, March 25, 2015 at=
 10:38 PM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>John E Drake &lt;<a href=3D"ma=
ilto:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali =
Sajassi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank"=
>bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>Re: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr"><style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.MsoChpDefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Hi John,<=
/font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">&nbsp;Fir=
stly i think EVPN community should reach consensus on the issues of current=
 DF election mechanism. All these issues should be resolved in a single new=
 DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3"></font></=
font></font>&nbsp;</p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Thanks,</=
font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">weiguo</f=
ont></font></font></p>
<font face=3D"MingLiU"><font face=3D"Calibri"></font></font>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF271558"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net" target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Your proposal introduces a control plan=
e processing load that is O(#EVIs * PEs) per DF election and given that the=
re can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">you can=92t &nbsp;use the ES route to c=
o-ordinate DF election because you would need to carry your new extended co=
mmunity for each EVI and they would not all
 fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because that=
 is processed by all PEs in the EVI.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">I think that from a practical perspecti=
ve the new DF election proposed in Satya=92s draft is sufficiently stable t=
hat it renders your draft moot, even if
 it could be made to work.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: Calibri,sans-serif; F=
ONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY: Calibri,sans-se=
rif; FONT-SIZE: 11pt"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" targe=
t=3D"_blank">mailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Hi Ali,</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks for =
your information. I scanned through this draft, it really introduces inter-=
chassis message for DF election handshaking, the requirements in this draft=
 is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"FON=
T-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Also for ac=
curacy, i think handshaking state machine on each multi-homed PE is also ne=
eded. From solution perspective, in my draft, no inter-chassis message is i=
ntroduced, only one new extended community
 is introduced, i think the process is comparatively simple than your follo=
wing draft.</span><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: bla=
ck; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Current EVP=
N DF election has some drawbacks, so there are three new drafts about DF el=
ection emerged. I think BESS WG can consider these three drafts in global v=
iew,&nbsp; a single,comprehensive new DF
 election draft is hoped.</span><span style=3D"FONT-FAMILY: Tahoma,sans-ser=
if; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">thanks.</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks,</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">weiguo</spa=
n><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 1=
0pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></b>=
<span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10p=
t"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">bess-b=
ounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F857476nkgeml501mbschi_--


From nobody Thu Mar 26 12:25:49 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 33E2B1B2DB5 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 12:25:47 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.61
X-Spam-Level: 
X-Spam-Status: No, score=-3.61 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iNPkHbogF80C for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 12:25:44 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 15ADE1B2D86 for <bess@ietf.org>; Thu, 26 Mar 2015 12:25:42 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml405-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BUE29809; Thu, 26 Mar 2015 19:25:41 +0000 (GMT)
Received: from NKGEML406-HUB.china.huawei.com (10.98.56.37) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.3.158.1; Thu, 26 Mar 2015 19:25:40 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml406-hub.china.huawei.com ([10.98.56.37]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 03:25:33 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: John E Drake <jdrake@juniper.net>, "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0
Date: Thu, 26 Mar 2015 19:25:32 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net>
In-Reply-To: <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.152.182]
Content-Type: multipart/alternative; boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F857483nkgeml501mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/QQ8mH4YY27C6PbKvMIQwLeuYhGk>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 19:25:47 -0000

--_000_DD5FC8DE455C3348B94340C0AB5517334F857483nkgeml501mbschi_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Pls read my detail replies to Satya. If you still can't catch it, pls read =
my draft and EVPN base protocol,  thanks

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 1:28
To: Patrice Brissette (pbrisset)
Cc: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:

Weiguo,

I=92m not sure I=92m following here.
Don=92t you have the same issue with your handshaking mechanism?
If you don=92t know your peer, how can you handshake?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_DD5FC8DE455C3348B94340C0AB5517334F857483nkgeml501mbschi_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body dir=3D"auto" fPStyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3" face=3D"Calibri">Pls read my detail replies to Satya. I=
f you still can't catch it, pls read my draft and EVPN base protocol, &nbsp=
;thanks</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF478649"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [jdrake@juniper.net]<br=
>
<b>Sent:</b> Friday, March 27, 2015 1:28<br>
<b>To:</b> Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>I think Patrice is correct. &nbsp;Your proposal doesn't solve the prob=
lem and it does so at huge cost.<br>
<br>
Sent from my iPhone</div>
<div><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt; wrot=
e:<br>
<br>
</div>
<blockquote type=3D"cite">
<div>
<div>
<div>Weiguo,</div>
<div><br>
</div>
<div>I=92m not sure I=92m following here.</div>
<div>Don=92t you have the same issue with your handshaking mechanism?&nbsp;=
</div>
<div>If you don=92t know your peer, how can you handshake?</div>
<div>
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Thursday, March 26, 2015 at =
10:24 AM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>Patrice Brissette &lt;<a href=
=3D"mailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt;=
, John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D"_blank">=
jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:sajassi@cisco=
.com" target=3D"_blank">sajassi@cisco.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>RE: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr">
<div style=3D"WORD-WRAP: break-word">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font size=3D"3" face=3D"Calibri">Hi Patrice,</font></p>
<p><font size=3D"3" face=3D"Calibri">Up to reception timer traffic disrupti=
on in transient phase is one of the issues.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF422254"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> Patrice Brissette (pbrisset) [<a hre=
f=3D"mailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>]<b=
r>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>
<div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
Weiguo,</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<div><font face=3D"Calibri,sans-serif">You mention &quot;</font>But if your=
 draft have not solved all issues=94,</div>
<div>Can you explain what Satya=92s draft is not solving?</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZ=
E: 14px" id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Wednesday, March 25, 2015 at=
 10:38 PM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>John E Drake &lt;<a href=3D"ma=
ilto:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali =
Sajassi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank"=
>bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>Re: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr"><style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.MsoChpDefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Hi John,<=
/font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">&nbsp;Fir=
stly i think EVPN community should reach consensus on the issues of current=
 DF election mechanism. All these issues should be resolved in a single new=
 DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3"></font></=
font></font>&nbsp;</p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Thanks,</=
font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">weiguo</f=
ont></font></font></p>
<font face=3D"MingLiU"><font face=3D"Calibri"></font></font>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF271558"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net" target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Your proposal introduces a control plan=
e processing load that is O(#EVIs * PEs) per DF election and given that the=
re can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">you can=92t &nbsp;use the ES route to c=
o-ordinate DF election because you would need to carry your new extended co=
mmunity for each EVI and they would not all
 fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because that=
 is processed by all PEs in the EVI.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">I think that from a practical perspecti=
ve the new DF election proposed in Satya=92s draft is sufficiently stable t=
hat it renders your draft moot, even if
 it could be made to work.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: Calibri,sans-serif; F=
ONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY: Calibri,sans-se=
rif; FONT-SIZE: 11pt"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" targe=
t=3D"_blank">mailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Hi Ali,</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks for =
your information. I scanned through this draft, it really introduces inter-=
chassis message for DF election handshaking, the requirements in this draft=
 is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"FON=
T-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Also for ac=
curacy, i think handshaking state machine on each multi-homed PE is also ne=
eded. From solution perspective, in my draft, no inter-chassis message is i=
ntroduced, only one new extended community
 is introduced, i think the process is comparatively simple than your follo=
wing draft.</span><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: bla=
ck; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Current EVP=
N DF election has some drawbacks, so there are three new drafts about DF el=
ection emerged. I think BESS WG can consider these three drafts in global v=
iew,&nbsp; a single,comprehensive new DF
 election draft is hoped.</span><span style=3D"FONT-FAMILY: Tahoma,sans-ser=
if; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">thanks.</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks,</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">weiguo</spa=
n><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 1=
0pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></b>=
<span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10p=
t"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">bess-b=
ounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</blockquote>
</div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F857483nkgeml501mbschi_--


From nobody Thu Mar 26 13:02:56 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A9A8C1B29FD for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 13:02:51 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.309
X-Spam-Level: 
X-Spam-Status: No, score=-6.309 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NWEoy4U8LOMT for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 13:02:46 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpgre-esg-01.alcatel-lucent.com [135.245.210.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 164D51B2E2E for <bess@ietf.org>; Thu, 26 Mar 2015 13:02:18 -0700 (PDT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (unknown [135.239.2.122]) by Websense Email Security Gateway with ESMTPS id 60939438EF294; Thu, 26 Mar 2015 20:02:12 +0000 (GMT)
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id t2QK2FLa028754 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 26 Mar 2015 21:02:15 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR711WXCHHUB02.zeu.alcatel-lucent.com ([135.239.2.112]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 21:02:15 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Haoweiguo <haoweiguo@huawei.com>, John E Drake <jdrake@juniper.net>, "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ1HeqN8Ih9YQqE+v8/YVcLeCH50uDT8AgACbTwCAACnuAIAAAsqAgAAf4ACAACCzAP//lOMA
Date: Thu, 26 Mar 2015 20:02:13 +0000
Message-ID: <D139B0BC.6C723%jorge.rabadan@alcatel-lucent.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.41]
Content-Type: multipart/alternative; boundary="_000_D139B0BC6C723jorgerabadanalcatellucentcom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/93R3pLzoKTMGeSnJvs8fzCTzRVQ>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 20:02:51 -0000

--_000_D139B0BC6C723jorgerabadanalcatellucentcom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_D139B0BC6C723jorgerabadanalcatellucentcom_
Content-Type: text/html; charset="utf-8"
Content-ID: <EA3B607720AB9348BF91B55ACCB83A92@exchange.lucent.com>
Content-Transfer-Encoding: base64
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=

--_000_D139B0BC6C723jorgerabadanalcatellucentcom_--


From nobody Thu Mar 26 13:41:47 2015
Return-Path: <jdrake@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4E0B21A8762 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 13:41:46 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.301
X-Spam-Level: 
X-Spam-Status: No, score=-1.301 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OKo7dbZ9kiML for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 13:41:40 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1bon0778.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::1:778]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BC0271B2EC7 for <bess@ietf.org>; Thu, 26 Mar 2015 13:41:39 -0700 (PDT)
Received: from BLUPR05MB561.namprd05.prod.outlook.com (10.141.202.139) by BLUPR05MB435.namprd05.prod.outlook.com (10.141.27.150) with Microsoft SMTP Server (TLS) id 15.1.118.21; Thu, 26 Mar 2015 20:41:19 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) by BLUPR05MB561.namprd05.prod.outlook.com (10.141.202.139) with Microsoft SMTP Server (TLS) id 15.1.118.21; Thu, 26 Mar 2015 20:41:11 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) by BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) with mapi id 15.01.0118.022; Thu, 26 Mar 2015 20:41:11 +0000
From: John E Drake <jdrake@juniper.net>
To: Haoweiguo <haoweiguo@huawei.com>, "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8P/o146fJTTukSNC2GolBinQZ0u4lsAgAACyoCAAB/gjYAAILMAgAAMEtA=
Date: Thu, 26 Mar 2015 20:41:11 +0000
Message-ID: <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [66.129.239.11]
authentication-results: huawei.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB561; UriScan:; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB435; 
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(24454002)(50854003)(36304003)(69234005)(51444003)(377454003)(164054003)(19609705001)(77096005)(15975445007)(86362001)(102836002)(16236675004)(17760045003)(19625215002)(561944003)(99936001)(50986999)(66066001)(74316001)(76176999)(54356999)(46102003)(33656002)(93886004)(62966003)(77156002)(2900100001)(92566002)(40100003)(2950100001)(87936001)(19580395003)(122556002)(19580405001)(99286002)(19617315012)(19627405001)(19627595001)(106116001)(76576001)(18206015028)(2656002)(19300405004); DIR:OUT; SFP:1102; SCL:1; SRVR:BLUPR05MB561; H:BLUPR05MB562.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BLUPR05MB561906AC98AF0F285CDFB6FC7080@BLUPR05MB561.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BLUPR05MB561; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB561; 
x-forefront-prvs: 0527DFA348
Content-Type: multipart/related; boundary="_005_BLUPR05MB56203C9BEDD75197CAE5489C7080BLUPR05MB562namprd_"; type="multipart/alternative"
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-originalarrivaltime: 26 Mar 2015 20:41:11.1187 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BLUPR05MB561
X-OriginatorOrg: juniper.net
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/d7UbZnfJDMNxWkYKml3-Uq00jVA>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 20:41:46 -0000

--_005_BLUPR05MB56203C9BEDD75197CAE5489C7080BLUPR05MB562namprd_
Content-Type: multipart/alternative;
	boundary="_000_BLUPR05MB56203C9BEDD75197CAE5489C7080BLUPR05MB562namprd_"

--_000_BLUPR05MB56203C9BEDD75197CAE5489C7080BLUPR05MB562namprd_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Weiguo,

We have already established that your proposal is untenable because of its =
control plane load.

What we are now discussing is that your proposal is based upon a misunderst=
anding of the algorithm in RFC 7432.  You are assuming that PE1 will advert=
ise an ES route and then wait for the configured interval before performing=
 the DF election while PE2 and PE3 will perform the DF election as soon as =
they receive the ES route from PE1.  This is not what RFC 7432 says.

Rather, what is says is that the advertisement of the ES route by PE1 and i=
ts receipt by PE2 and PE3 causes all three PEs to start the configured inte=
rval timer  - "3. When the timer expires, each PE builds an ordered list of=
 the IP addresses of all the PE nodes connected to the Ethernet segment (in=
cluding itself), in increasing numeric value."

Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 3:26 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Pls read my detail replies to Satya. If you still can't catch it, pls read =
my draft and EVPN base protocol,  thanks

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 1:28
To: Patrice Brissette (pbrisset)
Cc: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:
Weiguo,

I'm not sure I'm following here.
Don't you have the same issue with your handshaking mechanism?
If you don't know your peer, how can you handshake?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

You mention "But if your draft have not solved all issues",
Can you explain what Satya's draft is not solving?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can't  use the ES route to co-ordinate DF election because you would ne=
ed to carry your new extended community for each EVI and they would not all=
 fit.  You also can't use the Per EVI Ethernet AD route because that is pro=
cessed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya's draft is sufficiently stable that it renders your draft moot, even i=
f it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_BLUPR05MB56203C9BEDD75197CAE5489C7080BLUPR05MB562namprd_
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=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<!--[if !mso]><style id=3DowaParaStyle>v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style><![endif]--><style><!--
/* Font Definitions */
@font-face
	{font-family:MingLiU;
	panose-1:2 2 5 9 0 0 0 0 0 0;}
@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:Consolas;
	panose-1:2 11 6 9 2 2 4 3 2 4;}
@font-face
	{font-family:"\@MingLiU";
	panose-1:2 2 5 9 0 0 0 0 0 0;}
/* 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;
	margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman",serif;}
pre
	{mso-style-priority:99;
	mso-style-link:"HTML Preformatted Char";
	margin:0in;
	margin-bottom:.0001pt;
	font-size:10.0pt;
	font-family:Consolas;}
p.msochpdefault, li.msochpdefault, div.msochpdefault
	{mso-style-name:msochpdefault;
	margin:0in;
	margin-bottom:.0001pt;
	font-size:10.0pt;
	font-family:"Times New Roman",serif;}
span.emailstyle18
	{mso-style-name:emailstyle18;
	font-family:"Calibri",sans-serif;
	color:#1F497D;}
span.EmailStyle21
	{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:Consolas;}
.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=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;,sans-serif;color:#1F497D">Weiguo,<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">We have already established that your=
 proposal is untenable because of its control plane load.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">What we are now discussing is that yo=
ur proposal is based upon a misunderstanding of the algorithm in RFC 7432.&=
nbsp; You are assuming that PE1 will advertise an ES
 route and then wait for the configured interval before performing the DF e=
lection while PE2 and PE3 will perform the DF election as soon as they rece=
ive the ES route from PE1.&nbsp; This is not what RFC 7432 says.<o:p></o:p>=
</span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">Rather, what is says is that the adve=
rtisement of the ES route by PE1 and its receipt by PE2 and PE3 causes all =
three PEs to start the configured interval timer&nbsp;
 - &#8220;3. When the timer expires, each PE builds an ordered list of the =
IP addresses of all the PE nodes connected to the Ethernet segment (includi=
ng itself), in increasing numeric value.&#8221;<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D"><o:p>&nbsp;</o:p></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Yours Irrespectively,<o:p></o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">John<o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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 #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif">From:</span></b><span style=3D"font-size:11.0pt;=
font-family:&quot;Calibri&quot;,sans-serif"> Haoweiguo [mailto:haoweiguo@hu=
awei.com]
<br>
<b>Sent:</b> Thursday, March 26, 2015 3:26 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">P=
ls read my detail replies to Satya. If you still can't catch it, pls read m=
y draft and EVPN base protocol, &nbsp;thanks</span><span style=3D"font-size=
:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black"><o:p></o:p><=
/span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
sans-serif;color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF478649">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> John E Drake [jdrake@juniper.net]<br>
<b>Sent:</b> Friday, March 27, 2015 1:28<br>
<b>To:</b> Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">I think Patrice is corre=
ct. &nbsp;Your proposal doesn't solve the problem and it does so at huge co=
st.<br>
<br>
Sent from my iPhone<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><span style=3D"color:=
black"><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt; wrot=
e:<o:p></o:p></span></p>
</div>
<blockquote style=3D"margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Weiguo,<o:p></o:p></span=
></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><o:p>&nbsp;</o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">I&#8217;m not sure I&#82=
17;m following here.<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Don&#8217;t you have the=
 same issue with your handshaking mechanism?&nbsp;<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">If you don&#8217;t know =
your peer, how can you handshake?<o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><o:p>&nbsp;</o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><o:p>&nbsp;</o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Regards,<o:p></o:p></spa=
n></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><o:p>&nbsp;</o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><span style=3D"color:=
black">Patrice<o:p></o:p></span></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td style=3D"padding:0in 0in 0in 0in">
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"border:solid windowtext =
1.0pt;padding:0in"><img border=3D"0" width=3D"110" height=3D"73" id=3D"_x00=
00_i1026" src=3D"cid:image001.jpg@01D067DF.BB3C17F0" alt=3D"Image removed b=
y sender."></span><o:p></o:p></p>
</td>
</tr>
<tr>
<td nowrap=3D"" valign=3D"top" style=3D"padding:0in 0in 0in .25in">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Patrice Brissette</span></strong><span style=3D"font=
-size:8.5pt;font-family:&quot;Arial&quot;,sans-serif;color:#666666"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"colo=
r:#666666;text-decoration:none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"font-family:&quot;Arial&quot;,sans-serif">&#4=
3;1 613 254 3336</span></strong><o:p></o:p></span></p>
</td>
<td width=3D"50%" valign=3D"top" style=3D"width:50.0%;padding:11.25pt 0in 7=
.5pt 15.0pt">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Cisco Systems Canada Co. / Les Systemes Cisco Canada=
 CIE</span></strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&=
quot;,sans-serif;color:#666666"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"color:#666666;text-decoration:none">Cisco.com</span></a><o:p></o:p></sp=
an></p>
</td>
<td style=3D"padding:0in 0in 0in 0in"></td>
</tr>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in"></td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"display:none"><o:p>&nbsp;</o:p></span=
></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"400" style=3D"width:300.0pt">
<tbody>
<tr>
<td style=3D"padding:3.75pt 15.0pt 0in .25in">
<p class=3D"MsoNormal"><span style=3D"font-size:7.5pt;font-family:&quot;Ari=
al&quot;,sans-serif;color:#009900;border:solid windowtext 1.0pt;padding:0in=
"><img border=3D"0" width=3D"100" height=3D"100" id=3D"_x0000_i1027" src=3D=
"cid:~WRD000.jpg" alt=3D"Image removed by sender."></span><span style=3D"fo=
nt-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;color:#009900">Think
 before you print.<o:p></o:p></span></p>
</td>
</tr>
<tr>
<td style=3D"padding:5.25pt 15.0pt 4.5pt .25in">
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">This email may contain confidential and privileged material =
for the sole use of the intended recipient. Any review, use, distribution o=
r disclosure by others is strictly prohibited.
 If you are not the intended recipient (or authorized to receive for the re=
cipient), please contact the sender by reply email and delete all copies of=
 this message.<o:p></o:p></span></p>
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">Please
<a href=3D"http://www.cisco.com/web/about/doing_business/legal/cri/index.ht=
ml" target=3D"_blank" title=3D"Legal Information">
<span style=3D"color:#0E58A0">click here</span></a> for Company Registratio=
n Information.<o:p></o:p></span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"color:black"><br clear=3D"all">
<o:p></o:p></span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><o:p>&nbsp;</o:p></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:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:
</span></b><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,=
sans-serif;color:black">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.co=
m" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Thursday, March 26, 2015 at 10:24 AM<br>
<b>To: </b>Patrice Brissette &lt;<a href=3D"mailto:pbrisset@cisco.com" targ=
et=3D"_blank">pbrisset@cisco.com</a>&gt;, John E Drake &lt;<a href=3D"mailt=
o:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali Saj=
assi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@cis=
co.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<b>Subject: </b>RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><o:p>&nbsp;</o:p></span>=
</p>
</div>
<blockquote style=3D"border:none;border-left:solid #B5C4DF 4.5pt;padding:0i=
n 0in 0in 4.0pt;margin-left:3.75pt;margin-right:0in" id=3D"MAC_OUTLOOK_ATTR=
IBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">H=
i Patrice,</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&q=
uot;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">U=
p to reception timer traffic disruption in transient phase is one of the is=
sues.</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-seri=
f;color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">T=
hanks,</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;=
,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
sans-serif;color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF422254">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> Patrice Brissette (pbrisset) [<a href=3D"mailto:pbri=
sset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black">Weiguo,<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black"><o:p>&nbsp;</o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family:&quot;Calibri&quot;,sans-=
serif;color:black">You mention &quot;</span><span style=3D"color:black">But=
 if your draft have not solved all issues&#8221;,<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Can you explain what Sat=
ya&#8217;s draft is not solving?<o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black"><o:p>&nbsp;</o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black"><o:p>&nbsp;</o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black">Regards,<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black"><o:p>&nbsp;</o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><span style=3D"font-s=
ize:10.5pt;font-family:&quot;Calibri&quot;,sans-serif;color:black">Patrice<=
o:p></o:p></span></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td style=3D"padding:0in 0in 0in 0in">
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"border:solid windowtext =
1.0pt;padding:0in"><img border=3D"0" width=3D"110" height=3D"73" id=3D"_x00=
00_i1029" src=3D"cid:image001.jpg@01D067DF.BB3C17F0" alt=3D"Image removed b=
y sender."></span><o:p></o:p></p>
</td>
</tr>
<tr>
<td nowrap=3D"" valign=3D"top" style=3D"padding:0in 0in 0in .25in">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Patrice Brissette</span></strong><span style=3D"font=
-size:8.5pt;font-family:&quot;Arial&quot;,sans-serif;color:#666666"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"colo=
r:#666666;text-decoration:none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"font-family:&quot;Arial&quot;,sans-serif">&#4=
3;1 613 254 3336</span></strong><o:p></o:p></span></p>
</td>
<td width=3D"50%" valign=3D"top" style=3D"width:50.0%;padding:11.25pt 0in 7=
.5pt 15.0pt">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Cisco Systems Canada Co. / Les Systemes Cisco Canada=
 CIE</span></strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&=
quot;,sans-serif;color:#666666"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"color:#666666;text-decoration:none">Cisco.com</span></a><o:p></o:p></sp=
an></p>
</td>
<td style=3D"padding:0in 0in 0in 0in"></td>
</tr>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in"></td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"display:none"><o:p>&nbsp;</o:p></span=
></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"400" style=3D"width:300.0pt">
<tbody>
<tr>
<td style=3D"padding:3.75pt 15.0pt 0in .25in">
<p class=3D"MsoNormal"><span style=3D"font-size:7.5pt;font-family:&quot;Ari=
al&quot;,sans-serif;color:#009900;border:solid windowtext 1.0pt;padding:0in=
"><img border=3D"0" width=3D"100" height=3D"100" id=3D"_x0000_i1030" src=3D=
"cid:~WRD000.jpg" alt=3D"Image removed by sender."></span><span style=3D"fo=
nt-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;color:#009900">Think
 before you print.<o:p></o:p></span></p>
</td>
</tr>
<tr>
<td style=3D"padding:5.25pt 15.0pt 4.5pt .25in">
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">This email may contain confidential and privileged material =
for the sole use of the intended recipient. Any review, use, distribution o=
r disclosure by others is strictly prohibited.
 If you are not the intended recipient (or authorized to receive for the re=
cipient), please contact the sender by reply email and delete all copies of=
 this message.<o:p></o:p></span></p>
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">Please
<a href=3D"http://www.cisco.com/web/about/doing_business/legal/cri/index.ht=
ml" target=3D"_blank" title=3D"Legal Information">
<span style=3D"color:#0E58A0">click here</span></a> for Company Registratio=
n Information.<o:p></o:p></span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black"><br clear=3D"all">
<o:p></o:p></span></p>
</div>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black"><o:p>&nbsp;</o:p></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:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:
</span></b><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,=
sans-serif;color:black">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.co=
m" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Wednesday, March 25, 2015 at 10:38 PM<br>
<b>To: </b>John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D=
"_blank">jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:saja=
ssi@cisco.com" target=3D"_blank">sajassi@cisco.com</a>&gt;, &quot;<a href=
=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<b>Subject: </b>Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black"><o:p>&nbsp;</o:p></span></p>
</div>
<blockquote style=3D"border:none;border-left:solid #B5C4DF 4.5pt;padding:0i=
n 0in 0in 4.0pt;margin-left:3.75pt;margin-right:0in" id=3D"MAC_OUTLOOK_ATTR=
IBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">H=
i John,</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot=
;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">&=
nbsp;Firstly i think EVPN community should reach consensus on the issues of=
 current DF election mechanism. All these issues should be resolved in a si=
ngle new DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</span><span style=3D"font-size:10.0pt;font-family:&quot;Tah=
oma&quot;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-seri=
f;color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">T=
hanks,</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;=
,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
sans-serif;color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF271558">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> John E Drake [<a href=3D"mailto:jdrake@juniper.net" =
target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Weiguo,</span><span style=3D"color:bl=
ack"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Your proposal introduces a control pl=
ane processing load that is O(#EVIs * PEs) per DF election and given that t=
here can be 4K EVIs per ES, this looks like a
 *<b>substantial</b>* load.&nbsp; Furthermore,</span><span style=3D"color:b=
lack"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">you can&#8217;t &nbsp;use the ES rout=
e to co-ordinate DF election because you would need to carry your new exten=
ded community for each EVI and they would not all fit.&nbsp; You
 also can&#8217;t use the Per EVI Ethernet AD route because that is process=
ed by all PEs in the EVI.</span><span style=3D"color:black"><o:p></o:p></sp=
an></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">I think that from a practical perspec=
tive the new DF election proposed in Satya&#8217;s draft is sufficiently st=
able that it renders your draft moot, even if it could
 be made to work.</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Yours Irrespectively,</span><span sty=
le=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">John</span><span style=3D"color:black=
"><o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></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 #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:</span></b><span style=3D"font-=
size:11.0pt;font-family:&quot;Calibri&quot;,sans-serif;color:black"> BESS [=
<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">mailto:bess-boun=
ces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<div>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Hi=
 Ali,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Th=
anks for your information. I scanned through this draft, it really introduc=
es inter-chassis message for DF election handshaking, the requirements in t=
his draft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"col=
or:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Al=
so for accuracy, i think handshaking state machine on each multi-homed PE i=
s also needed. From solution perspective, in my draft, no inter-chassis mes=
sage is introduced, only one new extended community
 is introduced, i think the process is comparatively simple than your follo=
wing draft.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Cu=
rrent EVPN DF election has some drawbacks, so there are three new drafts ab=
out DF election emerged. I think BESS WG can consider these three drafts in=
 global view,&nbsp; a single,comprehensive new DF
 election draft is hoped.</span><span style=3D"color:black"><o:p></o:p></sp=
an></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">th=
anks.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Th=
anks,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">we=
iguo</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF484031">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" targe=
t=3D"_blank">bess-bounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">FYI- First published Jul=
y 4, 2011<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><a href=3D"https://datat=
racker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_blan=
k">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route<=
/a>/<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">-Ali<o:p></o:p></span></=
p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_BLUPR05MB56203C9BEDD75197CAE5489C7080BLUPR05MB562namprd_--

--_005_BLUPR05MB56203C9BEDD75197CAE5489C7080BLUPR05MB562namprd_
Content-Type: image/jpeg; name="~WRD000.jpg"
Content-Description: ~WRD000.jpg
Content-Disposition: inline; filename="~WRD000.jpg"; size=823;
	creation-date="Thu, 26 Mar 2015 20:08:48 GMT";
	modification-date="Thu, 26 Mar 2015 20:08:48 GMT"
Content-ID: <~WRD000.jpg>
Content-Transfer-Encoding: base64
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--_005_BLUPR05MB56203C9BEDD75197CAE5489C7080BLUPR05MB562namprd_
Content-Type: image/jpeg; name="image001.jpg"
Content-Description: image001.jpg
Content-Disposition: inline; filename="image001.jpg"; size=437;
	creation-date="Thu, 26 Mar 2015 20:41:10 GMT";
	modification-date="Thu, 26 Mar 2015 20:41:10 GMT"
Content-ID: <image001.jpg@01D067DF.BB3C17F0>
Content-Transfer-Encoding: base64

/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAoHBwkHBgoJCAkLCwoMDxkQDw4ODx4WFxIZJCAmJSMg
IyIoLTkwKCo2KyIjMkQyNjs9QEBAJjBGS0U+Sjk/QD3/wAALCABJAG4BAREA/8QAHwAAAQUBAQEB
AQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1Fh
ByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZ
WmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXG
x8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/9oACAEBAAA/APZqKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK//9k=

--_005_BLUPR05MB56203C9BEDD75197CAE5489C7080BLUPR05MB562namprd_--


From nobody Thu Mar 26 14:17:28 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5A9841B2F2F for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 14:17:26 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.61
X-Spam-Level: 
X-Spam-Status: No, score=-3.61 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gEq9BjlHryoQ for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 14:17:22 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1E5DA1B2F27 for <bess@ietf.org>; Thu, 26 Mar 2015 14:17:20 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml405-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT13769; Thu, 26 Mar 2015 21:17:19 +0000 (GMT)
Received: from NKGEML403-HUB.china.huawei.com (10.98.56.34) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.3.158.1; Thu, 26 Mar 2015 21:17:11 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml403-hub.china.huawei.com ([10.98.56.34]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 05:17:06 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, John E Drake <jdrake@juniper.net>, "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+EjICAAJSQRw==
Date: Thu, 26 Mar 2015 21:17:06 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F8574D0@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>, <D139B0BC.6C723%jorge.rabadan@alcatel-lucent.com>
In-Reply-To: <D139B0BC.6C723%jorge.rabadan@alcatel-lucent.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.156.0]
Content-Type: multipart/alternative; boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F8574D0nkgeml501mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/B7z0pnealR4Kym_HKngR2MON4dE>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 21:17:26 -0000

--_000_DD5FC8DE455C3348B94340C0AB5517334F8574D0nkgeml501mbschi_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Hi Jorge,

I understood your draft. Pls see below other replies with [weiguo].



Thanks,

weiguo

________________________________
From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
Sent: Friday, March 27, 2015 4:02
To: Haoweiguo; John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Since you mention https://tools.ietf.org/html/draft-rabadan-bess-evpn-ac-df=
-00 let me point at a few differences with your draft:

- In our draft we are NOT using AD routes for the DF election. The DF elect=
ion is still based on RFC7432.
[weiguo]: I know your draft don't using AD routes for DF election, but your=
 DF election needs to consider AD-route not just ES route.

I fully agree with Patrice/John/Satya that doing this handshaking is an unn=
ecessary control plane overhead. It also slows down significantly the DF el=
ection process, at the expense of maybe avoiding a few duplicate packets (l=
oops are resolved in the RFC7432) if the timers in RFC7432 were not configu=
red properly.
[weiguo]: It's not slowing down DF election, it is to make DF PE and non-DF=
 take effect at the same time. If you think it is slowing down, it just slo=
wing down one handshaking message transmission timer, the transmission time=
r is very short comparing to reception timer value. And i think the slowing=
 down have no side effect to network forwarding behaviour, traffic disrupti=
on timer is the matter.

- In our draft, for EVPN, we are not adding any new attribute or modifying =
the DF election. We are only using the existing information from the AD per=
-EVI routes to modify the candidate list for a given EVI. But note that tho=
se AD routes per-EVI are part of RFC7432 and are sent any way. So our draft=
 is fully compatible with RFC7432 and Satya=92s draft.

- In your draft, it is not clear if you use AD routes per-ES or per-EVI.
a) If it is per-ES, it simply does not work - as John says you need the DF =
indication per EVI.
b) if it is per-EVI then it is a huge overhead as was mentioned below, and =
those routes get to every PE in the network.
[weiguo]: I think the per-EVI handshaking message overhead is acceptable, i=
t is transmitted between multi-homed PE belonging to same ESI. It doesn't n=
eed to be transmitted to other remote PEs.


- Also in your slides, you also seem to use a timer to allow time to discov=
er the ESI=85 plus the additional time of sending the DF indication and rec=
eiving the DF notification. This is BGP, so assuming you have RRs and the m=
in-route-advertisement-intervals recommended, plus the local additional pro=
cessing you are introducing, the whole DF election process is significantly=
 slowed down.
[weiguo]: I think the message transmission timer normally is very short. Th=
e handshaking mechanism is a common mechanism used in native ethernet world=
, all handshaking mechanism has message transmission timer traffic loss, i =
think it is acceptable.
Thank you.
Jorge

From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 12:25 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, "Patrice =
Brissette (pbrisset)" <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>
Cc: "Ali Sajassi (sajassi)" <sajassi@cisco.com<mailto:sajassi@cisco.com>>, =
"bess@ietf.org<mailto:bess@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Pls read my detail replies to Satya. If you still can't catch it, pls read =
my draft and EVPN base protocol,  thanks

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Friday, March 27, 2015 1:28
To: Patrice Brissette (pbrisset)
Cc: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:

Weiguo,

I=92m not sure I=92m following here.
Don=92t you have the same issue with your handshaking mechanism?
If you don=92t know your peer, how can you handshake?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_DD5FC8DE455C3348B94340C0AB5517334F8574D0nkgeml501mbschi_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"FONT-FAMILY: Avenir,sans-serif; WORD-WRAP: break-word; COLOR=
: rgb(0,0,0); FONT-SIZE: 15px" fPStyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3" face=3D"Calibri">Hi Jorge,</font></p>
<p><font size=3D"3" face=3D"Calibri">I understood your draft. Pls see below=
 other replies with [weiguo].</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF403042"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> Rabadan, Jorge (Jorge) [jorge.rabada=
n@alcatel-lucent.com]<br>
<b>Sent:</b> Friday, March 27, 2015 4:02<br>
<b>To:</b> Haoweiguo; John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>Weiguo,</div>
<div><br>
</div>
<div>Since you mention&nbsp;<a href=3D"https://tools.ietf.org/html/draft-ra=
badan-bess-evpn-ac-df-00" target=3D"_blank">https://tools.ietf.org/html/dra=
ft-rabadan-bess-evpn-ac-df-00</a>&nbsp;let me point at a few differences wi=
th your draft:</div>
<div><br>
</div>
<div>- In our draft we are NOT using AD routes for the DF election. The DF =
election is still based on RFC7432.
</div>
<div>[weiguo]: I know your draft&nbsp;don't using AD routes for DF election=
, but your DF election needs to consider AD-route not just ES route.</div>
<div>&nbsp;</div>
<div>I fully agree with Patrice/John/Satya that doing this handshaking is a=
n unnecessary control plane overhead. It also slows down significantly the =
DF election process, at the expense of maybe avoiding a few duplicate packe=
ts (loops are resolved in the RFC7432)
 if the timers in RFC7432 were not configured properly.<br>
[weiguo]: It's not slowing down DF election, it is to make DF PE and non-DF=
 take effect at the same time. If you think it is slowing down, it just slo=
wing down one handshaking message transmission timer, the transmission time=
r is very short comparing to reception
 timer value. And i think the slowing down have no side effect to network f=
orwarding behaviour, traffic disruption timer is the matter.</div>
<div>&nbsp;</div>
<div>- In our draft, for EVPN, we are not adding any new attribute or modif=
ying the DF election. We are only using the existing information from the A=
D per-EVI routes to modify the candidate list for a given EVI. But note tha=
t those AD routes per-EVI are part
 of RFC7432 and are sent any way. So our draft is fully compatible with RFC=
7432 and Satya=92s draft.</div>
<div><br>
</div>
<div>- In your draft, it is not clear if you use AD routes per-ES or per-EV=
I.&nbsp;</div>
<div>a) If it is per-ES, it simply does not work - as John says you need th=
e DF indication per EVI.&nbsp;</div>
<div>b) if it is per-EVI then it is a huge overhead as was mentioned below,=
 and those routes get to every PE in the network.</div>
<div>[weiguo]: I think the per-EVI handshaking message overhead is acceptab=
le, it is transmitted between multi-homed PE belonging to same ESI. It does=
n't need to be transmitted to other remote PEs.</div>
<div><br>
&nbsp;</div>
<div>- Also in your slides, you also seem to use a timer to allow time to d=
iscover the ESI=85 plus the additional time of sending the DF indication an=
d receiving the DF notification. This is BGP, so assuming you have RRs and =
the min-route-advertisement-intervals
 recommended, plus the local additional processing you are introducing, the=
 whole DF election process is significantly slowed down.</div>
<div>[weiguo]: I think the message transmission timer normally is very shor=
t. The handshaking mechanism is a common mechanism used in native ethernet =
world, all handshaking mechanism has message transmission timer traffic los=
s, i think it is acceptable.<br>
</div>
<div>Thank you.</div>
<div>Jorge</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Thursday, March 26, 2015 at =
12:25 PM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>John E Drake &lt;<a href=3D"ma=
ilto:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, &quo=
t;Patrice Brissette (pbrisset)&quot; &lt;<a href=3D"mailto:pbrisset@cisco.c=
om" target=3D"_blank">pbrisset@cisco.com</a>&gt;<br>
<span style=3D"FONT-WEIGHT: bold">Cc: </span>&quot;Ali Sajassi (sajassi)&qu=
ot; &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@cisc=
o.com</a>&gt;, &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bes=
s@ietf.org</a>&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank"=
>bess@ietf.org</a>&gt;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>Re: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr">
<div dir=3D"auto">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font size=3D"3" face=3D"Calibri">Pls read my detail replies to Satya. I=
f you still can't catch it, pls read my draft and EVPN base protocol, &nbsp=
;thanks</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF478649"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net" target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Friday, March 27, 2015 1:28<br>
<b>To:</b> Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>I think Patrice is correct. &nbsp;Your proposal doesn't solve the prob=
lem and it does so at huge cost.<br>
<br>
Sent from my iPhone</div>
<div><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt; wrot=
e:<br>
<br>
</div>
<blockquote type=3D"cite">
<div>
<div>
<div>Weiguo,</div>
<div><br>
</div>
<div>I=92m not sure I=92m following here.</div>
<div>Don=92t you have the same issue with your handshaking mechanism?&nbsp;=
</div>
<div>If you don=92t know your peer, how can you handshake?</div>
<div>
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Thursday, March 26, 2015 at =
10:24 AM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>Patrice Brissette &lt;<a href=
=3D"mailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt;=
, John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D"_blank">=
jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:sajassi@cisco=
.com" target=3D"_blank">sajassi@cisco.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>RE: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr">
<div style=3D"WORD-WRAP: break-word">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font size=3D"3" face=3D"Calibri">Hi Patrice,</font></p>
<p><font size=3D"3" face=3D"Calibri">Up to reception timer traffic disrupti=
on in transient phase is one of the issues.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF422254"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> Patrice Brissette (pbrisset) [<a hre=
f=3D"mailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>]<b=
r>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>
<div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
Weiguo,</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<div><font face=3D"Calibri,sans-serif">You mention &quot;</font>But if your=
 draft have not solved all issues=94,</div>
<div>Can you explain what Satya=92s draft is not solving?</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<div><br>
</div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
</div>
<div style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZE=
: 14px">
<br>
</div>
<span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: rgb(0,0,0); FONT-SIZ=
E: 14px" id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Wednesday, March 25, 2015 at=
 10:38 PM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>John E Drake &lt;<a href=3D"ma=
ilto:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali =
Sajassi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@=
cisco.com</a>&gt;, &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank"=
>bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>Re: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr"><style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.EmailStyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.MsoChpDefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Hi John,<=
/font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">&nbsp;Fir=
stly i think EVPN community should reach consensus on the issues of current=
 DF election mechanism. All these issues should be resolved in a single new=
 DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3"></font></=
font></font>&nbsp;</p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">Thanks,</=
font></font></font></p>
<p><font face=3D"MingLiU"><font face=3D"Calibri"><font size=3D"3">weiguo</f=
ont></font></font></p>
<font face=3D"MingLiU"><font face=3D"Calibri"></font></font>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF271558"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net" target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Your proposal introduces a control plan=
e processing load that is O(#EVIs * PEs) per DF election and given that the=
re can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">you can=92t &nbsp;use the ES route to c=
o-ordinate DF election because you would need to carry your new extended co=
mmunity for each EVI and they would not all
 fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because that=
 is processed by all PEs in the EVI.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">I think that from a practical perspecti=
ve the new DF election proposed in Satya=92s draft is sufficiently stable t=
hat it renders your draft moot, even if
 it could be made to work.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: Calibri,sans-serif; F=
ONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY: Calibri,sans-se=
rif; FONT-SIZE: 11pt"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" targe=
t=3D"_blank">mailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Hi Ali,</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks for =
your information. I scanned through this draft, it really introduces inter-=
chassis message for DF election handshaking, the requirements in this draft=
 is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"FON=
T-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Also for ac=
curacy, i think handshaking state machine on each multi-homed PE is also ne=
eded. From solution perspective, in my draft, no inter-chassis message is i=
ntroduced, only one new extended community
 is introduced, i think the process is comparatively simple than your follo=
wing draft.</span><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: bla=
ck; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Current EVP=
N DF election has some drawbacks, so there are three new drafts about DF el=
ection emerged. I think BESS WG can consider these three drafts in global v=
iew,&nbsp; a single,comprehensive new DF
 election draft is hoped.</span><span style=3D"FONT-FAMILY: Tahoma,sans-ser=
if; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">thanks.</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks,</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">weiguo</spa=
n><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 1=
0pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></b>=
<span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10p=
t"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">bess-b=
ounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F8574D0nkgeml501mbschi_--


From nobody Thu Mar 26 14:37:07 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CC42F1B2B56 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 14:37:04 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.61
X-Spam-Level: 
X-Spam-Status: No, score=-3.61 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id b_gsxFlT-Cly for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 14:37:00 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D5F681A89B4 for <bess@ietf.org>; Thu, 26 Mar 2015 14:36:59 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml402-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT14681; Thu, 26 Mar 2015 21:36:58 +0000 (GMT)
Received: from NKGEML404-HUB.china.huawei.com (10.98.56.35) by lhreml402-hub.china.huawei.com (10.201.5.241) with Microsoft SMTP Server (TLS) id 14.3.158.1; Thu, 26 Mar 2015 21:36:57 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml404-hub.china.huawei.com ([10.98.56.35]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 05:36:51 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: John E Drake <jdrake@juniper.net>, "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuow==
Date: Thu, 26 Mar 2015 21:36:50 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com>
In-Reply-To: <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.156.0]
Content-Type: multipart/related; boundary="_005_DD5FC8DE455C3348B94340C0AB5517334F8574F0nkgeml501mbschi_"; type="multipart/alternative"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/cQo-24BGN5vA01PjwWefJs9sQSw>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 21:37:05 -0000

--_005_DD5FC8DE455C3348B94340C0AB5517334F8574F0nkgeml501mbschi_
Content-Type: multipart/alternative;
	boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F8574F0nkgeml501mbschi_"

--_000_DD5FC8DE455C3348B94340C0AB5517334F8574F0nkgeml501mbschi_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

John,

As your understanding of the EVPN draft, the DF election mechanism has more=
 serious side effect, it will have short time traffic loop,i.e., dual DF PE=
s will exist for a short time. I think dual DF PEs is absolutely not tolera=
ted, because native ethernet header has no TTL, up to several hundred ms tr=
affic loop normally not tolerated in commertial networks.



As your understanding, the PEs should do as following:

1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have rarely=
 deployment.

Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins ESI an=
d trigger DF re-election. When reception timer expires:

PE1 upgrades to DF PE.

After reception timer+ ES route transmission timer:

PE2 downloads to non-DF PE.

So in timer clock sync case, dual DF PEs will exist at least transmission t=
imer.



If NTP is used for timer sync, because it has bad accuracy, dual DF PEs wil=
l exist more longer timer.



So as your understanding for DF election, the drawback is more clear.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 4:41
To: Haoweiguo; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

We have already established that your proposal is untenable because of its =
control plane load.

What we are now discussing is that your proposal is based upon a misunderst=
anding of the algorithm in RFC 7432.  You are assuming that PE1 will advert=
ise an ES route and then wait for the configured interval before performing=
 the DF election while PE2 and PE3 will perform the DF election as soon as =
they receive the ES route from PE1.  This is not what RFC 7432 says.

Rather, what is says is that the advertisement of the ES route by PE1 and i=
ts receipt by PE2 and PE3 causes all three PEs to start the configured inte=
rval timer  - =933. When the timer expires, each PE builds an ordered list =
of the IP addresses of all the PE nodes connected to the Ethernet segment (=
including itself), in increasing numeric value.=94

Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 3:26 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Pls read my detail replies to Satya. If you still can't catch it, pls read =
my draft and EVPN base protocol,  thanks

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 1:28
To: Patrice Brissette (pbrisset)
Cc: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:
Weiguo,

I=92m not sure I=92m following here.
Don=92t you have the same issue with your handshaking mechanism?
If you don=92t know your peer, how can you handshake?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_DD5FC8DE455C3348B94340C0AB5517334F8574F0nkgeml501mbschi_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
<style>@font-face {
	font-family: MingLiU;
}
@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@font-face {
	font-family: Consolas;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
PRE {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: Consolas; FONT-SIZE: 10pt
}
P.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
LI.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
DIV.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
SPAN.emailstyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
SPAN.EmailStyle21 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
SPAN.HTMLPreformattedChar {
	FONT-FAMILY: Consolas
}
.msochpdefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple" fPStyle=3D"1" ocsi=3D"0=
">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3" face=3D"Calibri">John,</font></p>
<p><font size=3D"3" face=3D"Calibri">As your understanding of the EVPN draf=
t, the DF election mechanism has more serious side effect, it will have sho=
rt time traffic loop,i.e., dual DF PEs will exist for a short time. I think=
 dual DF PEs is absolutely not tolerated,
 because native ethernet header has no TTL, up to several hundred ms traffi=
c loop normally not tolerated in commertial networks.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">As your understanding, the PEs should =
do as following:</font></p>
<p><font size=3D"3" face=3D"Calibri">1. Accurate timer sync. NTP accuracy i=
s bad, 1588v2 is good but have rarely deployment.</font></p>
<p><font size=3D"3" face=3D"Calibri">Assuming PE1,PE2 and PE3 have consiste=
nt timer clock, when PE3 joins ESI and trigger DF re-election. When recepti=
on timer expires:</font></p>
<p><font size=3D"3" face=3D"Calibri">PE1 upgrades to DF PE.</font></p>
<p><font size=3D"3" face=3D"Calibri">After reception timer&#43; ES route tr=
ansmission timer:</font></p>
<p><font size=3D"3" face=3D"Calibri">PE2 downloads to non-DF PE.</font></p>
<p><font size=3D"3" face=3D"Calibri">So in timer clock sync case, dual DF P=
Es will exist at least transmission timer.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">If NTP is used for timer sync, because=
 it has bad accuracy, dual DF PEs will exist more longer timer.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">So as your understanding for DF electi=
on, the drawback is more clear.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF234411"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [jdrake@juniper.net]<br=
>
<b>Sent:</b> Friday, March 27, 2015 4:41<br>
<b>To:</b> Haoweiguo; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">We have already established that your propos=
al is untenable because of its control plane load.
</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">What we are now discussing is that your prop=
osal is based upon a misunderstanding of the algorithm in RFC 7432.&nbsp; Y=
ou are assuming that PE1 will advertise an
 ES route and then wait for the configured interval before performing the D=
F election while PE2 and PE3 will perform the DF election as soon as they r=
eceive the ES route from PE1.&nbsp; This is not what RFC 7432 says.</span><=
/p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Rather, what is says is that the advertiseme=
nt of the ES route by PE1 and its receipt by PE2 and PE3 causes all three P=
Es to start the configured interval
 timer&nbsp; - =933. When the timer expires, each PE builds an ordered list=
 of the IP addresses of all the PE nodes connected to the Ethernet segment =
(including itself), in increasing numeric value.=94</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif;=
 FONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY: 'Calibri',san=
s-serif; FONT-SIZE: 11pt"> Haoweiguo [mailto:haoweiguo@huawei.com]
<br>
<b>Sent:</b> Thursday, March 26, 2015 3:26 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Pls read=
 my detail replies to Satya. If you still can't catch it, pls read my draft=
 and EVPN base protocol, &nbsp;thanks</span><span style=3D"FONT-FAMILY: 'Ta=
homa',sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">weiguo</=
span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SI=
ZE: 10pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF478649">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></=
b><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE:=
 10pt"> John E Drake [jdrake@juniper.net]<br>
<b>Sent:</b> Friday, March 27, 2015 1:28<br>
<b>To:</b> Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">I think Patrice is corr=
ect. &nbsp;Your proposal doesn't solve the problem and it does so at huge c=
ost.<br>
<br>
Sent from my iPhone</span></p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"COLOR: =
black"><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt; wrot=
e:</span></p>
</div>
<blockquote style=3D"MARGIN-TOP: 5pt; MARGIN-BOTTOM: 5pt">
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Weiguo,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">I=92m not sure I=92m fo=
llowing here.</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Don=92t you have the sa=
me issue with your handshaking mechanism?&nbsp;</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">If you don=92t know you=
r peer, how can you handshake?</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Regards,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"COLOR: =
black">Patrice</span></p>
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"BORDER-BOTTOM: windowtex=
t 1pt solid; BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDIN=
G-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: windowtext 1pt solid; BORDER-R=
IGHT: windowtext 1pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1026" bord=
er=3D"0" alt=3D"Image removed by sender." src=3D"cid:image001.jpg@01D067DF.=
BB3C17F0" width=3D"110" height=3D"73"></span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 0in;=
 PADDING-TOP: 0in" valign=3D"top" nowrap=3D"">
<p><strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #666666; =
FONT-SIZE: 8.5pt">Patrice Brissette</span></strong><span style=3D"FONT-FAMI=
LY: 'Arial',sans-serif; COLOR: #666666; FONT-SIZE: 8.5pt"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"COLO=
R: #666666; TEXT-DECORATION: none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif">&#43;1 613 2=
54 3336</span></strong></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 7.5pt; PADDING-LEFT: 15pt; WIDTH: 50%; PADDING=
-RIGHT: 0in; PADDING-TOP: 11.25pt" valign=3D"top" width=3D"50%">
<p><strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #666666; =
FONT-SIZE: 8.5pt">Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE<=
/span></strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #6666=
66; FONT-SIZE: 8.5pt"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"COLOR: #666666; TEXT-DECORATION: none">Cisco.com</span></a></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span></span>&nbsp;</p>
<table style=3D"WIDTH: 300pt" class=3D"MsoNormalTable" border=3D"0" cellspa=
cing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15pt=
; PADDING-TOP: 3.75pt">
<p class=3D"MsoNormal"><span style=3D"BORDER-BOTTOM: windowtext 1pt solid; =
BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; =
PADDING-RIGHT: 0in; FONT-FAMILY: 'Arial',sans-serif; COLOR: #009900; FONT-S=
IZE: 7.5pt; BORDER-TOP: windowtext 1pt solid; BORDER-RIGHT: windowtext 1pt =
solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1027" border=3D"0" alt=3D"Image=
 removed by sender." src=3D"cid:~WRD000.jpg" width=3D"100" height=3D"100"><=
/span><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #009900; FONT-=
SIZE: 7.5pt">Think
 before you print.</span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 4.5pt; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15=
pt; PADDING-TOP: 5.25pt">
<p><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #999999; FONT-SIZ=
E: 7.5pt">This email may contain confidential and privileged material for t=
he sole use of the intended recipient. Any review, use, distribution or dis=
closure by others is strictly prohibited.
 If you are not the intended recipient (or authorized to receive for the re=
cipient), please contact the sender by reply email and delete all copies of=
 this message.</span></p>
<p><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #999999; FONT-SIZ=
E: 7.5pt">Please
<a title=3D"Legal Information" href=3D"http://www.cisco.com/web/about/doing=
_business/legal/cri/index.html" target=3D"_blank">
<span style=3D"COLOR: #0e58a0">click here</span></a> for Company Registrati=
on Information.</span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><br clear=3D"all">
</span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif;=
 COLOR: black; FONT-SIZE: 11pt">From:
</span></b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black; =
FONT-SIZE: 11pt">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.com" targ=
et=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Thursday, March 26, 2015 at 10:24 AM<br>
<b>To: </b>Patrice Brissette &lt;<a href=3D"mailto:pbrisset@cisco.com" targ=
et=3D"_blank">pbrisset@cisco.com</a>&gt;, John E Drake &lt;<a href=3D"mailt=
o:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali Saj=
assi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@cis=
co.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<b>Subject: </b>RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<blockquote style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: #b5c4df 4.5pt=
 solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; MARGIN-=
LEFT: 3.75pt; BORDER-TOP: medium none; MARGIN-RIGHT: 0in; BORDER-RIGHT: med=
ium none; PADDING-TOP: 0in" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Hi Patri=
ce,</span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FO=
NT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Up to re=
ception timer traffic disruption in transient phase is one of the issues.</=
span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SI=
ZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE=
: 10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Thanks,<=
/span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-S=
IZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">weiguo</=
span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SI=
ZE: 10pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF422254">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></=
b><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE:=
 10pt"> Patrice Brissette (pbrisset) [<a href=3D"mailto:pbrisset@cisco.com"=
 target=3D"_blank">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt">Weiguo,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black">You mention &quot;</span><span style=3D"COLOR: black">But if yo=
ur draft have not solved all issues=94,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Can you explain what Sa=
tya=92s draft is not solving?</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt">Regards,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"FONT-FA=
MILY: 'Calibri',sans-serif; COLOR: black; FONT-SIZE: 10.5pt">Patrice</span>=
</p>
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"BORDER-BOTTOM: windowtex=
t 1pt solid; BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDIN=
G-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: windowtext 1pt solid; BORDER-R=
IGHT: windowtext 1pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1029" bord=
er=3D"0" alt=3D"Image removed by sender." src=3D"cid:image001.jpg@01D067DF.=
BB3C17F0" width=3D"110" height=3D"73"></span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 0in;=
 PADDING-TOP: 0in" valign=3D"top" nowrap=3D"">
<p><strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #666666; =
FONT-SIZE: 8.5pt">Patrice Brissette</span></strong><span style=3D"FONT-FAMI=
LY: 'Arial',sans-serif; COLOR: #666666; FONT-SIZE: 8.5pt"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"COLO=
R: #666666; TEXT-DECORATION: none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif">&#43;1 613 2=
54 3336</span></strong></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 7.5pt; PADDING-LEFT: 15pt; WIDTH: 50%; PADDING=
-RIGHT: 0in; PADDING-TOP: 11.25pt" valign=3D"top" width=3D"50%">
<p><strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #666666; =
FONT-SIZE: 8.5pt">Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE<=
/span></strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #6666=
66; FONT-SIZE: 8.5pt"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"COLOR: #666666; TEXT-DECORATION: none">Cisco.com</span></a></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span></span>&nbsp;</p>
<table style=3D"WIDTH: 300pt" class=3D"MsoNormalTable" border=3D"0" cellspa=
cing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15pt=
; PADDING-TOP: 3.75pt">
<p class=3D"MsoNormal"><span style=3D"BORDER-BOTTOM: windowtext 1pt solid; =
BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; =
PADDING-RIGHT: 0in; FONT-FAMILY: 'Arial',sans-serif; COLOR: #009900; FONT-S=
IZE: 7.5pt; BORDER-TOP: windowtext 1pt solid; BORDER-RIGHT: windowtext 1pt =
solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1030" border=3D"0" alt=3D"Image=
 removed by sender." src=3D"cid:~WRD000.jpg" width=3D"100" height=3D"100"><=
/span><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #009900; FONT-=
SIZE: 7.5pt">Think
 before you print.</span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 4.5pt; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15=
pt; PADDING-TOP: 5.25pt">
<p><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #999999; FONT-SIZ=
E: 7.5pt">This email may contain confidential and privileged material for t=
he sole use of the intended recipient. Any review, use, distribution or dis=
closure by others is strictly prohibited.
 If you are not the intended recipient (or authorized to receive for the re=
cipient), please contact the sender by reply email and delete all copies of=
 this message.</span></p>
<p><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #999999; FONT-SIZ=
E: 7.5pt">Please
<a title=3D"Legal Information" href=3D"http://www.cisco.com/web/about/doing=
_business/legal/cri/index.html" target=3D"_blank">
<span style=3D"COLOR: #0e58a0">click here</span></a> for Company Registrati=
on Information.</span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt"><br clear=3D"all">
</span></p>
</div>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif;=
 COLOR: black; FONT-SIZE: 11pt">From:
</span></b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black; =
FONT-SIZE: 11pt">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.com" targ=
et=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Wednesday, March 25, 2015 at 10:38 PM<br>
<b>To: </b>John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D=
"_blank">jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:saja=
ssi@cisco.com" target=3D"_blank">sajassi@cisco.com</a>&gt;, &quot;<a href=
=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<b>Subject: </b>Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<blockquote style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: #b5c4df 4.5pt=
 solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; MARGIN-=
LEFT: 3.75pt; BORDER-TOP: medium none; MARGIN-RIGHT: 0in; BORDER-RIGHT: med=
ium none; PADDING-TOP: 0in" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Hi John,=
</span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-=
SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">&nbsp;Fi=
rstly i think EVPN community should reach consensus on the issues of curren=
t DF election mechanism. All these issues should be resolved in a single ne=
w DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLO=
R: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE=
: 10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Thanks,<=
/span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-S=
IZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">weiguo</=
span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SI=
ZE: 10pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF271558">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></=
b><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE:=
 10pt"> John E Drake [<a href=3D"mailto:jdrake@juniper.net" target=3D"_blan=
k">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Weiguo,</span><span style=3D"COLOR: black"><=
/span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Your proposal introduces a control plane pro=
cessing load that is O(#EVIs * PEs) per DF election and given that there ca=
n be 4K EVIs per ES, this looks like
 a *<b>substantial</b>* load.&nbsp; Furthermore,</span><span style=3D"COLOR=
: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">you can=92t &nbsp;use the ES route to co-ord=
inate DF election because you would need to carry your new extended communi=
ty for each EVI and they would not all fit.&nbsp;
 You also can=92t use the Per EVI Ethernet AD route because that is process=
ed by all PEs in the EVI.</span><span style=3D"COLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">I think that from a practical perspective th=
e new DF election proposed in Satya=92s draft is sufficiently stable that i=
t renders your draft moot, even if it
 could be made to work.</span><span style=3D"COLOR: black"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Yours Irrespectively,</span><span style=3D"C=
OLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">John</span><span style=3D"COLOR: black"></sp=
an></p>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif;=
 COLOR: black; FONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY:=
 'Calibri',sans-serif; COLOR: black; FONT-SIZE: 11pt"> BESS [<a href=3D"mai=
lto:bess-bounces@ietf.org" target=3D"_blank">mailto:bess-bounces@ietf.org</=
a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Hi Ali,</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Thanks fo=
r your information. I scanned through this draft, it really introduces inte=
r-chassis message for DF election handshaking, the requirements in this dra=
ft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"COL=
OR: black"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Also for =
accuracy, i think handshaking state machine on each multi-homed PE is also =
needed. From solution perspective, in my draft, no inter-chassis message is=
 introduced, only one new extended
 community is introduced, i think the process is comparatively simple than =
your following draft.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Current E=
VPN DF election has some drawbacks, so there are three new drafts about DF =
election emerged. I think BESS WG can consider these three drafts in global=
 view,&nbsp; a single,comprehensive new
 DF election draft is hoped.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">thanks.</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Thanks,</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">weiguo</s=
pan><span style=3D"COLOR: black"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></=
b><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE:=
 10pt"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">be=
ss-bounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F8574F0nkgeml501mbschi_--

--_005_DD5FC8DE455C3348B94340C0AB5517334F8574F0nkgeml501mbschi_
Content-Type: image/jpeg; name="~WRD000.jpg"
Content-Description: ~WRD000.jpg
Content-Disposition: inline; filename="~WRD000.jpg"; size=823;
	creation-date="Thu, 26 Mar 2015 20:08:48 GMT";
	modification-date="Thu, 26 Mar 2015 20:08:48 GMT"
Content-ID: <~WRD000.jpg>
Content-Transfer-Encoding: base64
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--_005_DD5FC8DE455C3348B94340C0AB5517334F8574F0nkgeml501mbschi_
Content-Type: image/jpeg; name="image001.jpg"
Content-Description: image001.jpg
Content-Disposition: inline; filename="image001.jpg"; size=437;
	creation-date="Thu, 26 Mar 2015 20:41:10 GMT";
	modification-date="Thu, 26 Mar 2015 20:41:10 GMT"
Content-ID: <image001.jpg@01D067DF.BB3C17F0>
Content-Transfer-Encoding: base64

/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAoHBwkHBgoJCAkLCwoMDxkQDw4ODx4WFxIZJCAmJSMg
IyIoLTkwKCo2KyIjMkQyNjs9QEBAJjBGS0U+Sjk/QD3/wAALCABJAG4BAREA/8QAHwAAAQUBAQEB
AQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1Fh
ByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZ
WmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXG
x8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/9oACAEBAAA/APZqKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK//9k=

--_005_DD5FC8DE455C3348B94340C0AB5517334F8574F0nkgeml501mbschi_--


From nobody Thu Mar 26 14:56:46 2015
Return-Path: <pbrisset@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EC8561A902A for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 14:56:44 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -13.91
X-Spam-Level: 
X-Spam-Status: No, score=-13.91 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HnZ2HYpUW01f for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 14:56:40 -0700 (PDT)
Received: from alln-iport-1.cisco.com (alln-iport-1.cisco.com [173.37.142.88]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0A3961B2EFB for <bess@ietf.org>; Thu, 26 Mar 2015 14:56:39 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=55856; q=dns/txt; s=iport; t=1427407000; x=1428616600; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=7jpbW5eX2IpBlTzvIlnn2UQl0pXKilnMgeGYEZhJlos=; b=NbDKdmUmzpRVDxX9ONJHZ5jo1hfzoTDzUryyZ6Ydvlz3S6vJ1U4gmCWx 7KInlQeTTrSUd4r9LhSBOJsuh/3EzOka+VqSe/SMRkmisdWHON3FomEAH mpFRir3TgJNkywc/eaUuprsrp/Kwx0zkhTC/MblyQN1VH3APd4jhXWLDK I=;
X-Files: ~WRD000.jpg, image001.jpg : 823, 437
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0CDBgADgBRV/4YNJK1ZAxaCLUNSWgTCcoI9eYR6AoFKTAEBAQEBAX2EFAEBAQQFFQsJOAMBDw4CAgEIEQMBAQEGAQEBCg4BBgcCFQQJAgwUCQgCBAENBAEOBgeIFA2zKJh1AQEBAQEBAQEBAQEBAQEBAQEBAQEBFwSBMYlzhBUBEQEOFgYLCQEBBgYEBgEGAwiCaYEzBY5Cgg6BaYEyAVMfhWGBGzqCdgWCVYlBg0ciggEdgVBvAYEDBxcifwEBAQ
X-IronPort-AV: E=Sophos;i="5.11,475,1422921600";  d="jpg'145?scan'145,208,217,145";a="135724830"
Received: from alln-core-12.cisco.com ([173.36.13.134]) by alln-iport-1.cisco.com with ESMTP; 26 Mar 2015 21:56:39 +0000
Received: from xhc-rcd-x09.cisco.com (xhc-rcd-x09.cisco.com [173.37.183.83]) by alln-core-12.cisco.com (8.14.5/8.14.5) with ESMTP id t2QLucRo012382 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 26 Mar 2015 21:56:38 GMT
Received: from xmb-aln-x09.cisco.com ([169.254.4.43]) by xhc-rcd-x09.cisco.com ([173.37.183.83]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 16:56:38 -0500
From: "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
To: Haoweiguo <haoweiguo@huawei.com>, John E Drake <jdrake@juniper.net>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQaA+7Ws9VLcZ0TUWO8xI+wRZimQ==
Date: Thu, 26 Mar 2015 21:56:37 +0000
Message-ID: <D139E9F1.4EFAC%pbrisset@cisco.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [10.89.7.222]
Content-Type: multipart/mixed; boundary="_005_D139E9F14EFACpbrissetciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/9wOhk2MY5uylRDqrIR5DhuL8WEE>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 21:56:45 -0000

--_005_D139E9F14EFACpbrissetciscocom_
Content-Type: multipart/alternative;
	boundary="_000_D139E9F14EFACpbrissetciscocom_"

--_000_D139E9F14EFACpbrissetciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Weiguo,

I sent already too much time of this so I hope this is the last email.
I read again one more time your draft and it is NOT addressing the issue th=
at you keep coming back to.
See, if you don=92t want any loop, simply bring your interface in blocked s=
tate until DF election converge.

Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 4:36 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Patrice B=
rissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>
Cc: Ali Sajassi <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.o=
rg<mailto:bess@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


John,

As your understanding of the EVPN draft, the DF election mechanism has more=
 serious side effect, it will have short time traffic loop,i.e., dual DF PE=
s will exist for a short time. I think dual DF PEs is absolutely not tolera=
ted, because native ethernet header has no TTL, up to several hundred ms tr=
affic loop normally not tolerated in commertial networks.



As your understanding, the PEs should do as following:

1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have rarely=
 deployment.

Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins ESI an=
d trigger DF re-election. When reception timer expires:

PE1 upgrades to DF PE.

After reception timer+ ES route transmission timer:

PE2 downloads to non-DF PE.

So in timer clock sync case, dual DF PEs will exist at least transmission t=
imer.



If NTP is used for timer sync, because it has bad accuracy, dual DF PEs wil=
l exist more longer timer.



So as your understanding for DF election, the drawback is more clear.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Friday, March 27, 2015 4:41
To: Haoweiguo; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

We have already established that your proposal is untenable because of its =
control plane load.

What we are now discussing is that your proposal is based upon a misunderst=
anding of the algorithm in RFC 7432.  You are assuming that PE1 will advert=
ise an ES route and then wait for the configured interval before performing=
 the DF election while PE2 and PE3 will perform the DF election as soon as =
they receive the ES route from PE1.  This is not what RFC 7432 says.

Rather, what is says is that the advertisement of the ES route by PE1 and i=
ts receipt by PE2 and PE3 causes all three PEs to start the configured inte=
rval timer  - =933. When the timer expires, each PE builds an ordered list =
of the IP addresses of all the PE nodes connected to the Ethernet segment (=
including itself), in increasing numeric value.=94

Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 3:26 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Pls read my detail replies to Satya. If you still can't catch it, pls read =
my draft and EVPN base protocol,  thanks

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Friday, March 27, 2015 1:28
To: Patrice Brissette (pbrisset)
Cc: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:
Weiguo,

I=92m not sure I=92m following here.
Don=92t you have the same issue with your handshaking mechanism?
If you don=92t know your peer, how can you handshake?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_D139E9F14EFACpbrissetciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <6250FE1C800FE94CB2826746FF5B88BD@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-fami=
ly: Calibri, sans-serif;">
<div>
<div>Weiguo,</div>
<div><br>
</div>
<div>I sent already too much time of this so I hope this is the last email.=
</div>
<div>I read again one more time your draft and it is NOT addressing the iss=
ue that you keep coming back to.</div>
<div>See, if you don=92t want any loop, simply bring your interface in bloc=
ked state until DF election converge.</div>
<div>
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"padding-left: 24px; padding-bottom: 0;" align=3D"left" nowrap=
=3D"nowrap" valign=3D"top">
<p style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 11px; col=
or: rgb(102, 102, 102);">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"color:#666666; text-decoration: none;" href=3D"mailto:pbrisset@=
cisco.com">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"padding-left: 20px; padding-bottom: 10px; padding-top: 15px;" =
valign=3D"top" width=3D"50%">
<p style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 11px; col=
or: rgb(102, 102, 102);">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"color:#666666; text-decoration: none;" href=3D"http://www.cisco=
.com/global/CA/">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 10px; co=
lor: #999999; padding-left: 24px; padding-right: 20px; padding-top: 5px; pa=
dding-bottom: 0px; color:#009900; ">
<img src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif" al=
t=3D""> Think before you print.</td>
</tr>
<tr>
<td style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 10px; co=
lor: #999999; padding-left: 24px; padding-right: 20px; padding-top: 7px; pa=
dding-bottom: 6px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a href=3D"http://www.cisco.com/web/about/doing_business/legal/cr=
i/index.html" title=3D"Legal Information" style=3D"color: #0E58A0;">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"font-family:Calibri; font-size:11pt; text-align:left; color:b=
lack; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM:=
 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid;=
 BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Haoweiguo &lt;<a href=3D"mail=
to:haoweiguo@huawei.com">haoweiguo@huawei.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Thursday, March 26, 2015 at 4=
:36 PM<br>
<span style=3D"font-weight:bold">To: </span>John E Drake &lt;<a href=3D"mai=
lto:jdrake@juniper.net">jdrake@juniper.net</a>&gt;, Patrice Brissette &lt;<=
a href=3D"mailto:pbrisset@cisco.com">pbrisset@cisco.com</a>&gt;<br>
<span style=3D"font-weight:bold">Cc: </span>Ali Sajassi &lt;<a href=3D"mail=
to:sajassi@cisco.com">sajassi@cisco.com</a>&gt;, &quot;<a href=3D"mailto:be=
ss@ietf.org">bess@ietf.org</a>&quot; &lt;<a href=3D"mailto:bess@ietf.org">b=
ess@ietf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>RE: [bess] Handshaking amo=
ng PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE" style=3D"BORDER-LEFT:=
 #b5c4df 5 solid; PADDING:0 0 0 5; MARGIN:0 0 0 5;">
<div dir=3D"ltr"><style>@font-face {
	font-family: MingLiU;
}
@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@font-face {
	font-family: Consolas;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
PRE {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: Consolas; FONT-SIZE: 10pt
}
P.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
LI.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
DIV.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
SPAN.emailstyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
SPAN.EmailStyle21 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
SPAN.HTMLPreformattedChar {
	FONT-FAMILY: Consolas
}
.msochpdefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple" fpstyle=3D"1" ocsi=3D"0"=
>
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3" face=3D"Calibri">John,</font></p>
<p><font size=3D"3" face=3D"Calibri">As your understanding of the EVPN draf=
t, the DF election mechanism has more serious side effect, it will have sho=
rt time traffic loop,i.e., dual DF PEs will exist for a short time. I think=
 dual DF PEs is absolutely not tolerated,
 because native ethernet header has no TTL, up to several hundred ms traffi=
c loop normally not tolerated in commertial networks.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">As your understanding, the PEs should =
do as following:</font></p>
<p><font size=3D"3" face=3D"Calibri">1. Accurate timer sync. NTP accuracy i=
s bad, 1588v2 is good but have rarely deployment.</font></p>
<p><font size=3D"3" face=3D"Calibri">Assuming PE1,PE2 and PE3 have consiste=
nt timer clock, when PE3 joins ESI and trigger DF re-election. When recepti=
on timer expires:</font></p>
<p><font size=3D"3" face=3D"Calibri">PE1 upgrades to DF PE.</font></p>
<p><font size=3D"3" face=3D"Calibri">After reception timer&#43; ES route tr=
ansmission timer:</font></p>
<p><font size=3D"3" face=3D"Calibri">PE2 downloads to non-DF PE.</font></p>
<p><font size=3D"3" face=3D"Calibri">So in timer clock sync case, dual DF P=
Es will exist at least transmission timer.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">If NTP is used for timer sync, because=
 it has bad accuracy, dual DF PEs will exist more longer timer.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">So as your understanding for DF electi=
on, the drawback is more clear.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF234411"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Friday, March 27, 2015 4:41<br>
<b>To:</b> Haoweiguo; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org">bess@iet=
f.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">We have already established that yo=
ur proposal is untenable because of its control plane load.
</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">What we are now discussing is that =
your proposal is based upon a misunderstanding of the algorithm in RFC 7432=
.&nbsp; You are assuming that PE1 will advertise
 an ES route and then wait for the configured interval before performing th=
e DF election while PE2 and PE3 will perform the DF election as soon as the=
y receive the ES route from PE1.&nbsp; This is not what RFC 7432 says.</spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Rather, what is says is that the ad=
vertisement of the ES route by PE1 and its receipt by PE2 and PE3 causes al=
l three PEs to start the configured
 interval timer&nbsp; - =933. When the timer expires, each PE builds an ord=
ered list of the IP addresses of all the PE nodes connected to the Ethernet=
 segment (including itself), in increasing numeric value.=94</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
font-size: 11pt;">From:</span></b><span style=3D"font-family: Calibri, sans=
-serif; font-size: 11pt;"> Haoweiguo [<a href=3D"mailto:haoweiguo@huawei.co=
m">mailto:haoweiguo@huawei.com</a>]
<br>
<b>Sent:</b> Thursday, March 26, 2015 3:26 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org">bess@iet=
f.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Pls read=
 my detail replies to Satya. If you still can't catch it, pls read my draft=
 and EVPN base protocol, &nbsp;thanks</span><span style=3D"font-family: Tah=
oma, sans-serif; color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">weiguo</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF478649">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> John E Drake [<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.=
net</a>]<br>
<b>Sent:</b> Friday, March 27, 2015 1:28<br>
<b>To:</b> Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g">bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">I think Patrice is corr=
ect. &nbsp;Your proposal doesn't solve the problem and it does so at huge c=
ost.<br>
<br>
Sent from my iPhone</span></p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"COLOR: =
black"><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt; wrot=
e:</span></p>
</div>
<blockquote style=3D"MARGIN-TOP: 5pt; MARGIN-BOTTOM: 5pt">
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Weiguo,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">I=92m not sure I=92m fo=
llowing here.</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Don=92t you have the sa=
me issue with your handshaking mechanism?&nbsp;</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">If you don=92t know you=
r peer, how can you handshake?</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Regards,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"COLOR: =
black">Patrice</span></p>
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"BORDER-BOTTOM: windowtex=
t 1pt solid; BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDIN=
G-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: windowtext 1pt solid; BORDER-R=
IGHT: windowtext 1pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1026" bord=
er=3D"0" alt=3D"Image removed by sender." src=3D"cid:image001.jpg@01D067DF.=
BB3C17F0" width=3D"110" height=3D"73"></span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 0in;=
 PADDING-TOP: 0in" valign=3D"top" nowrap=3D"">
<p><strong><span style=3D"font-family: Arial, sans-serif; color: rgb(102, 1=
02, 102); font-size: 8.5pt;">Patrice Brissette</span></strong><span style=
=3D"font-family: Arial, sans-serif; color: rgb(102, 102, 102); font-size: 8=
.5pt;"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"COLO=
R: #666666; TEXT-DECORATION: none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"font-family: Arial, sans-serif;">&#43;1 613 2=
54 3336</span></strong></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 7.5pt; PADDING-LEFT: 15pt; WIDTH: 50%; PADDING=
-RIGHT: 0in; PADDING-TOP: 11.25pt" valign=3D"top" width=3D"50%">
<p><strong><span style=3D"font-family: Arial, sans-serif; color: rgb(102, 1=
02, 102); font-size: 8.5pt;">Cisco Systems Canada Co. / Les Systemes Cisco =
Canada CIE</span></strong><span style=3D"font-family: Arial, sans-serif; co=
lor: rgb(102, 102, 102); font-size: 8.5pt;"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"COLOR: #666666; TEXT-DECORATION: none">Cisco.com</span></a></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span></span>&nbsp;</p>
<table style=3D"WIDTH: 300pt" class=3D"MsoNormalTable" border=3D"0" cellspa=
cing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15pt=
; PADDING-TOP: 3.75pt">
<p class=3D"MsoNormal"><span style=3D"border: 1pt solid windowtext; padding=
: 0in; font-family: Arial, sans-serif; color: rgb(0, 153, 0); font-size: 7.=
5pt;"><img id=3D"_x0000_i1027" border=3D"0" alt=3D"Image removed by sender.=
" src=3D"cid:~WRD000.jpg" width=3D"100" height=3D"100"></span><span style=
=3D"font-family: Arial, sans-serif; color: rgb(0, 153, 0); font-size: 7.5pt=
;">Think
 before you print.</span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 4.5pt; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15=
pt; PADDING-TOP: 5.25pt">
<p><span style=3D"font-family: Arial, sans-serif; color: rgb(153, 153, 153)=
; font-size: 7.5pt;">This email may contain confidential and privileged mat=
erial for the sole use of the intended recipient. Any review, use, distribu=
tion or disclosure by others is strictly
 prohibited. If you are not the intended recipient (or authorized to receiv=
e for the recipient), please contact the sender by reply email and delete a=
ll copies of this message.</span></p>
<p><span style=3D"font-family: Arial, sans-serif; color: rgb(153, 153, 153)=
; font-size: 7.5pt;">Please
<a title=3D"Legal Information" href=3D"http://www.cisco.com/web/about/doing=
_business/legal/cri/index.html" target=3D"_blank">
<span style=3D"COLOR: #0e58a0">click here</span></a> for Company Registrati=
on Information.</span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><br clear=3D"all">
</span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
color: black; font-size: 11pt;">From:
</span></b><span style=3D"font-family: Calibri, sans-serif; color: black; f=
ont-size: 11pt;">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.com" targ=
et=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Thursday, March 26, 2015 at 10:24 AM<br>
<b>To: </b>Patrice Brissette &lt;<a href=3D"mailto:pbrisset@cisco.com" targ=
et=3D"_blank">pbrisset@cisco.com</a>&gt;, John E Drake &lt;<a href=3D"mailt=
o:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali Saj=
assi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@cis=
co.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<b>Subject: </b>RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<blockquote style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: #b5c4df 4.5pt=
 solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; MARGIN-=
LEFT: 3.75pt; BORDER-TOP: medium none; MARGIN-RIGHT: 0in; BORDER-RIGHT: med=
ium none; PADDING-TOP: 0in" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Hi Patri=
ce,</span><span style=3D"font-family: Tahoma, sans-serif; color: black; fon=
t-size: 10pt;"></span></p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Up to re=
ception timer traffic disruption in transient phase is one of the issues.</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Thanks,<=
/span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-si=
ze: 10pt;"></span></p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">weiguo</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF422254">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> Patrice Brissette (pbrisset) [<a href=3D"mailto:pbrisset@cisco.com"=
 target=3D"_blank">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;">Weiguo,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black;">You mention &quot;</span><span style=3D"COLOR: black">But if yo=
ur draft have not solved all issues=94,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Can you explain what Sa=
tya=92s draft is not solving?</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;">Regards,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"font-fa=
mily: Calibri, sans-serif; color: black; font-size: 10.5pt;">Patrice</span>=
</p>
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"BORDER-BOTTOM: windowtex=
t 1pt solid; BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDIN=
G-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: windowtext 1pt solid; BORDER-R=
IGHT: windowtext 1pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1029" bord=
er=3D"0" alt=3D"Image removed by sender." src=3D"cid:image001.jpg@01D067DF.=
BB3C17F0" width=3D"110" height=3D"73"></span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 0in;=
 PADDING-TOP: 0in" valign=3D"top" nowrap=3D"">
<p><strong><span style=3D"font-family: Arial, sans-serif; color: rgb(102, 1=
02, 102); font-size: 8.5pt;">Patrice Brissette</span></strong><span style=
=3D"font-family: Arial, sans-serif; color: rgb(102, 102, 102); font-size: 8=
.5pt;"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"COLO=
R: #666666; TEXT-DECORATION: none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"font-family: Arial, sans-serif;">&#43;1 613 2=
54 3336</span></strong></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 7.5pt; PADDING-LEFT: 15pt; WIDTH: 50%; PADDING=
-RIGHT: 0in; PADDING-TOP: 11.25pt" valign=3D"top" width=3D"50%">
<p><strong><span style=3D"font-family: Arial, sans-serif; color: rgb(102, 1=
02, 102); font-size: 8.5pt;">Cisco Systems Canada Co. / Les Systemes Cisco =
Canada CIE</span></strong><span style=3D"font-family: Arial, sans-serif; co=
lor: rgb(102, 102, 102); font-size: 8.5pt;"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"COLOR: #666666; TEXT-DECORATION: none">Cisco.com</span></a></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span></span>&nbsp;</p>
<table style=3D"WIDTH: 300pt" class=3D"MsoNormalTable" border=3D"0" cellspa=
cing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15pt=
; PADDING-TOP: 3.75pt">
<p class=3D"MsoNormal"><span style=3D"border: 1pt solid windowtext; padding=
: 0in; font-family: Arial, sans-serif; color: rgb(0, 153, 0); font-size: 7.=
5pt;"><img id=3D"_x0000_i1030" border=3D"0" alt=3D"Image removed by sender.=
" src=3D"cid:~WRD000.jpg" width=3D"100" height=3D"100"></span><span style=
=3D"font-family: Arial, sans-serif; color: rgb(0, 153, 0); font-size: 7.5pt=
;">Think
 before you print.</span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 4.5pt; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15=
pt; PADDING-TOP: 5.25pt">
<p><span style=3D"font-family: Arial, sans-serif; color: rgb(153, 153, 153)=
; font-size: 7.5pt;">This email may contain confidential and privileged mat=
erial for the sole use of the intended recipient. Any review, use, distribu=
tion or disclosure by others is strictly
 prohibited. If you are not the intended recipient (or authorized to receiv=
e for the recipient), please contact the sender by reply email and delete a=
ll copies of this message.</span></p>
<p><span style=3D"font-family: Arial, sans-serif; color: rgb(153, 153, 153)=
; font-size: 7.5pt;">Please
<a title=3D"Legal Information" href=3D"http://www.cisco.com/web/about/doing=
_business/legal/cri/index.html" target=3D"_blank">
<span style=3D"COLOR: #0e58a0">click here</span></a> for Company Registrati=
on Information.</span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"><br clear=3D"all">
</span></p>
</div>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
color: black; font-size: 11pt;">From:
</span></b><span style=3D"font-family: Calibri, sans-serif; color: black; f=
ont-size: 11pt;">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.com" targ=
et=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Wednesday, March 25, 2015 at 10:38 PM<br>
<b>To: </b>John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D=
"_blank">jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:saja=
ssi@cisco.com" target=3D"_blank">sajassi@cisco.com</a>&gt;, &quot;<a href=
=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<b>Subject: </b>Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<blockquote style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: #b5c4df 4.5pt=
 solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; MARGIN-=
LEFT: 3.75pt; BORDER-TOP: medium none; MARGIN-RIGHT: 0in; BORDER-RIGHT: med=
ium none; PADDING-TOP: 0in" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Hi John,=
</span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-s=
ize: 10pt;"></span></p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">&nbsp;Fi=
rstly i think EVPN community should reach consensus on the issues of curren=
t DF election mechanism. All these issues should be resolved in a single ne=
w DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</span><span style=3D"font-family: Tahoma, sans-serif; color=
: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Thanks,<=
/span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-si=
ze: 10pt;"></span></p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">weiguo</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF271558">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> John E Drake [<a href=3D"mailto:jdrake@juniper.net" target=3D"_blan=
k">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Weiguo,</span><span style=3D"COLOR:=
 black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Your proposal introduces a control =
plane processing load that is O(#EVIs * PEs) per DF election and given that=
 there can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span><span style=3D"=
COLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">you can=92t &nbsp;use the ES route =
to co-ordinate DF election because you would need to carry your new extende=
d community for each EVI and they would not
 all fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because =
that is processed by all PEs in the EVI.</span><span style=3D"COLOR: black"=
></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">I think that from a practical persp=
ective the new DF election proposed in Satya=92s draft is sufficiently stab=
le that it renders your draft moot, even
 if it could be made to work.</span><span style=3D"COLOR: black"></span></p=
>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Yours Irrespectively,</span><span s=
tyle=3D"COLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">John</span><span style=3D"COLOR: bl=
ack"></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
color: black; font-size: 11pt;">From:</span></b><span style=3D"font-family:=
 Calibri, sans-serif; color: black; font-size: 11pt;"> BESS [<a href=3D"mai=
lto:bess-bounces@ietf.org" target=3D"_blank">mailto:bess-bounces@ietf.org</=
a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Hi Ali,</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Thanks fo=
r your information. I scanned through this draft, it really introduces inte=
r-chassis message for DF election handshaking, the requirements in this dra=
ft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"COL=
OR: black"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Also for =
accuracy, i think handshaking state machine on each multi-homed PE is also =
needed. From solution perspective, in my draft, no inter-chassis message is=
 introduced, only one new extended
 community is introduced, i think the process is comparatively simple than =
your following draft.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Current E=
VPN DF election has some drawbacks, so there are three new drafts about DF =
election emerged. I think BESS WG can consider these three drafts in global=
 view,&nbsp; a single,comprehensive new
 DF election draft is hoped.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">thanks.</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Thanks,</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">weiguo</s=
pan><span style=3D"COLOR: black"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">be=
ss-bounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span>
</body>
</html>

--_000_D139E9F14EFACpbrissetciscocom_--

--_005_D139E9F14EFACpbrissetciscocom_
Content-Type: image/jpeg; name="~WRD000.jpg"
Content-Description: ~WRD000.jpg
Content-Disposition: attachment; filename="~WRD000.jpg"; size=823;
	creation-date="Thu, 26 Mar 2015 21:56:37 GMT";
	modification-date="Thu, 26 Mar 2015 21:56:37 GMT"
Content-ID: <~WRD000.jpg>
Content-Transfer-Encoding: base64
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--_005_D139E9F14EFACpbrissetciscocom_
Content-Type: image/jpeg; name="image001.jpg"
Content-Description: image001.jpg
Content-Disposition: attachment; filename="image001.jpg"; size=437;
	creation-date="Thu, 26 Mar 2015 21:56:37 GMT";
	modification-date="Thu, 26 Mar 2015 21:56:37 GMT"
Content-ID: <image001.jpg@01D067DF.BB3C17F0>
Content-Transfer-Encoding: base64

/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAoHBwkHBgoJCAkLCwoMDxkQDw4ODx4WFxIZJCAmJSMg
IyIoLTkwKCo2KyIjMkQyNjs9QEBAJjBGS0U+Sjk/QD3/wAALCABJAG4BAREA/8QAHwAAAQUBAQEB
AQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1Fh
ByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZ
WmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXG
x8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/9oACAEBAAA/APZqKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK//9k=

--_005_D139E9F14EFACpbrissetciscocom_--


From nobody Thu Mar 26 15:00:33 2015
Return-Path: <jdrake@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id ADAF51B2FCE for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 15:00:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.301
X-Spam-Level: 
X-Spam-Status: No, score=-1.301 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3wms2D2CYnNj for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 15:00:22 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1bon0742.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::1:742]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 188941B2FCD for <bess@ietf.org>; Thu, 26 Mar 2015 15:00:22 -0700 (PDT)
Received: from BLUPR05MB561.namprd05.prod.outlook.com (10.141.202.139) by BLUPR05MB769.namprd05.prod.outlook.com (10.141.209.19) with Microsoft SMTP Server (TLS) id 15.1.118.21; Thu, 26 Mar 2015 22:00:06 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) by BLUPR05MB561.namprd05.prod.outlook.com (10.141.202.139) with Microsoft SMTP Server (TLS) id 15.1.118.21; Thu, 26 Mar 2015 22:00:04 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) by BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) with mapi id 15.01.0118.022; Thu, 26 Mar 2015 22:00:04 +0000
From: John E Drake <jdrake@juniper.net>
To: Haoweiguo <haoweiguo@huawei.com>, "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8P/o146fJTTukSNC2GolBinQZ0u4lsAgAACyoCAAB/gjYAAILMAgAAMEtCAABidAIAAA48A
Date: Thu, 26 Mar 2015 22:00:04 +0000
Message-ID: <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [66.129.239.11]
authentication-results: huawei.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB561; UriScan:; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB769; 
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(51444003)(164054003)(377454003)(50854003)(24454002)(69234005)(36304003)(2900100001)(2950100001)(92566002)(40100003)(62966003)(77156002)(19627595001)(106116001)(99286002)(19617315012)(2656002)(18206015028)(19300405004)(76576001)(19580395003)(87936001)(19580405001)(122556002)(561944003)(19625215002)(17760045003)(16236675004)(66066001)(50986999)(99936001)(19609705001)(86362001)(102836002)(77096005)(15975445007)(93886004)(33656002)(74316001)(46102003)(76176999)(54356999); DIR:OUT; SFP:1102; SCL:1; SRVR:BLUPR05MB561; H:BLUPR05MB562.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BLUPR05MB561FEA54AEB60E7F61EF8A9C7080@BLUPR05MB561.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BLUPR05MB561; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB561; 
x-forefront-prvs: 0527DFA348
Content-Type: multipart/related; boundary="_005_BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080BLUPR05MB562namprd_"; type="multipart/alternative"
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-originalarrivaltime: 26 Mar 2015 22:00:04.1969 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BLUPR05MB561
X-OriginatorOrg: juniper.net
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/J8YoHIbP-Xcs2J7EbekMXvTS1Fw>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 22:00:30 -0000

--_005_BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080BLUPR05MB562namprd_
Content-Type: multipart/alternative;
	boundary="_000_BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080BLUPR05MB562namprd_"

--_000_BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080BLUPR05MB562namprd_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

To recap,

We have established that your proposal is untenable because of its control =
plane load.

We have established that your proposal is based upon a flawed understanding=
 of the DF election in RFC 7432.

What you are now arguing is that your draft prevents two or more PEs from b=
eing DF simultaneously.  This is clearly nonsense.

Furthermore, we have established that having two or more DFs for what even =
you admit is a brief transient leads to duplicate traffic, which is accepta=
ble, but not loops, your assertion to the contrary.

Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 5:37 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


John,

As your understanding of the EVPN draft, the DF election mechanism has more=
 serious side effect, it will have short time traffic loop,i.e., dual DF PE=
s will exist for a short time. I think dual DF PEs is absolutely not tolera=
ted, because native ethernet header has no TTL, up to several hundred ms tr=
affic loop normally not tolerated in commertial networks.



As your understanding, the PEs should do as following:

1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have rarely=
 deployment.

Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins ESI an=
d trigger DF re-election. When reception timer expires:

PE1 upgrades to DF PE.

After reception timer+ ES route transmission timer:

PE2 downloads to non-DF PE.

So in timer clock sync case, dual DF PEs will exist at least transmission t=
imer.



If NTP is used for timer sync, because it has bad accuracy, dual DF PEs wil=
l exist more longer timer.



So as your understanding for DF election, the drawback is more clear.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 4:41
To: Haoweiguo; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

We have already established that your proposal is untenable because of its =
control plane load.

What we are now discussing is that your proposal is based upon a misunderst=
anding of the algorithm in RFC 7432.  You are assuming that PE1 will advert=
ise an ES route and then wait for the configured interval before performing=
 the DF election while PE2 and PE3 will perform the DF election as soon as =
they receive the ES route from PE1.  This is not what RFC 7432 says.

Rather, what is says is that the advertisement of the ES route by PE1 and i=
ts receipt by PE2 and PE3 causes all three PEs to start the configured inte=
rval timer  - "3. When the timer expires, each PE builds an ordered list of=
 the IP addresses of all the PE nodes connected to the Ethernet segment (in=
cluding itself), in increasing numeric value."

Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 3:26 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Pls read my detail replies to Satya. If you still can't catch it, pls read =
my draft and EVPN base protocol,  thanks

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 1:28
To: Patrice Brissette (pbrisset)
Cc: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:
Weiguo,

I'm not sure I'm following here.
Don't you have the same issue with your handshaking mechanism?
If you don't know your peer, how can you handshake?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

You mention "But if your draft have not solved all issues",
Can you explain what Satya's draft is not solving?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can't  use the ES route to co-ordinate DF election because you would ne=
ed to carry your new extended community for each EVI and they would not all=
 fit.  You also can't use the Per EVI Ethernet AD route because that is pro=
cessed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya's draft is sufficiently stable that it renders your draft moot, even i=
f it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080BLUPR05MB562namprd_
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=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<!--[if !mso]><style>v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style><![endif]--><style><!--
/* Font Definitions */
@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:Consolas;
	panose-1:2 11 6 9 2 2 4 3 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;
	margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman",serif;}
pre
	{mso-style-priority:99;
	mso-style-link:"HTML Preformatted Char";
	margin:0in;
	margin-bottom:.0001pt;
	font-size:10.0pt;
	font-family:Consolas;}
span.HTMLPreformattedChar
	{mso-style-name:"HTML Preformatted Char";
	mso-style-priority:99;
	mso-style-link:"HTML Preformatted";
	font-family:Consolas;}
p.msochpdefault, li.msochpdefault, div.msochpdefault
	{mso-style-name:msochpdefault;
	margin:0in;
	margin-bottom:.0001pt;
	font-size:10.0pt;
	font-family:"Times New Roman",serif;}
span.emailstyle18
	{mso-style-name:emailstyle18;
	font-family:"Calibri",sans-serif;
	color:#1F497D;}
span.emailstyle21
	{mso-style-name:emailstyle21;
	font-family:"Calibri",sans-serif;
	color:#1F497D;}
span.htmlpreformattedchar0
	{mso-style-name:htmlpreformattedchar;
	font-family:Consolas;}
span.EmailStyle25
	{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=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;,sans-serif;color:#1F497D">To recap,<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">We have established that your proposa=
l is untenable because of its control plane load.
<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">&nbsp;<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">We have established that your proposa=
l is based upon a flawed understanding of the DF election in RFC 7432.<o:p>=
</o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">What you are now arguing is that your=
 draft prevents two or more PEs from being DF simultaneously.&nbsp; This is=
 clearly nonsense.<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">Furthermore, we have established that=
 having two or more DFs for what even you admit is a brief transient leads =
to duplicate traffic, which is acceptable, but
 not loops, your assertion to the contrary.<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D"><o:p>&nbsp;</o:p></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Yours Irrespectively,<o:p></o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">John<o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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 #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif">From:</span></b><span style=3D"font-size:11.0pt;=
font-family:&quot;Calibri&quot;,sans-serif"> Haoweiguo [mailto:haoweiguo@hu=
awei.com]
<br>
<b>Sent:</b> Thursday, March 26, 2015 5:37 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">J=
ohn,</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,s=
ans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">A=
s your understanding of the EVPN draft, the DF election mechanism has more =
serious side effect, it will have short time traffic loop,i.e., dual DF PEs=
 will exist for a short time. I think dual DF
 PEs is absolutely not tolerated, because native ethernet header has no TTL=
, up to several hundred ms traffic loop normally not tolerated in commertia=
l networks.</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&=
quot;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-seri=
f;color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">A=
s your understanding, the PEs should do as following:</span><span style=3D"=
font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black"><o:=
p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">1=
. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have rarely =
deployment.</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&=
quot;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">A=
ssuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins ESI and=
 trigger DF re-election. When reception timer expires:</span><span style=3D=
"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black"><o=
:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">P=
E1 upgrades to DF PE.</span><span style=3D"font-size:10.0pt;font-family:&qu=
ot;Tahoma&quot;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">A=
fter reception timer&#43; ES route transmission timer:</span><span style=3D=
"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black"><o=
:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">P=
E2 downloads to non-DF PE.</span><span style=3D"font-size:10.0pt;font-famil=
y:&quot;Tahoma&quot;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">S=
o in timer clock sync case, dual DF PEs will exist at least transmission ti=
mer.</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,s=
ans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-seri=
f;color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">I=
f NTP is used for timer sync, because it has bad accuracy, dual DF PEs will=
 exist more longer timer.</span><span style=3D"font-size:10.0pt;font-family=
:&quot;Tahoma&quot;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-seri=
f;color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">S=
o as your understanding for DF election, the drawback is more clear.</span>=
<span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;c=
olor:black"><o:p></o:p></span></p>
<p><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-seri=
f;color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">T=
hanks,</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;=
,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
sans-serif;color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF234411">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> John E Drake [jdrake@juniper.net]<br>
<b>Sent:</b> Friday, March 27, 2015 4:41<br>
<b>To:</b> Haoweiguo; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org">bess@iet=
f.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Weiguo,</span><span style=3D"color:bl=
ack"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">We have already established that your=
 proposal is untenable because of its control plane load.
</span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">What we are now discussing is that yo=
ur proposal is based upon a misunderstanding of the algorithm in RFC 7432.&=
nbsp; You are assuming that PE1 will advertise an ES
 route and then wait for the configured interval before performing the DF e=
lection while PE2 and PE3 will perform the DF election as soon as they rece=
ive the ES route from PE1.&nbsp; This is not what RFC 7432 says.</span><spa=
n style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Rather, what is says is that the adve=
rtisement of the ES route by PE1 and its receipt by PE2 and PE3 causes all =
three PEs to start the configured interval timer&nbsp;
 - &#8220;3. When the timer expires, each PE builds an ordered list of the =
IP addresses of all the PE nodes connected to the Ethernet segment (includi=
ng itself), in increasing numeric value.&#8221;</span><span style=3D"color:=
black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Yours Irrespectively,</span><span sty=
le=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">John</span><span style=3D"color:black=
"><o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></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 #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:</span></b><span style=3D"font-=
size:11.0pt;font-family:&quot;Calibri&quot;,sans-serif;color:black"> Haowei=
guo [<a href=3D"mailto:haoweiguo@huawei.com">mailto:haoweiguo@huawei.com</a=
>]
<br>
<b>Sent:</b> Thursday, March 26, 2015 3:26 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org">bess@iet=
f.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">P=
ls read my detail replies to Satya. If you still can't catch it, pls read m=
y draft and EVPN base protocol, &nbsp;thanks</span><span style=3D"color:bla=
ck"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF478649">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> John E Drake [jdrake@juniper.net]<br>
<b>Sent:</b> Friday, March 27, 2015 1:28<br>
<b>To:</b> Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g">bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">I think Patrice is corre=
ct. &nbsp;Your proposal doesn't solve the problem and it does so at huge co=
st.<br>
<br>
Sent from my iPhone<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><span style=3D"color:=
black"><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt; wrot=
e:<o:p></o:p></span></p>
</div>
<blockquote style=3D"margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Weiguo,<o:p></o:p></span=
></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">I&#8217;m not sure I&#82=
17;m following here.<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Don&#8217;t you have the=
 same issue with your handshaking mechanism?&nbsp;<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">If you don&#8217;t know =
your peer, how can you handshake?<o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Regards,<o:p></o:p></spa=
n></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><span style=3D"color:=
black">Patrice<o:p></o:p></span></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td style=3D"padding:0in 0in 0in 0in">
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"border:solid windowtext =
1.0pt;padding:0in"><img border=3D"0" width=3D"110" height=3D"73" id=3D"_x00=
00_i1027" src=3D"cid:image001.jpg@01D067EE.7BA7B450" alt=3D"Image removed b=
y sender."></span><o:p></o:p></p>
</td>
</tr>
<tr>
<td nowrap=3D"" valign=3D"top" style=3D"padding:0in 0in 0in .25in">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Patrice Brissette</span></strong><span style=3D"font=
-size:8.5pt;font-family:&quot;Arial&quot;,sans-serif;color:#666666"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"colo=
r:#666666;text-decoration:none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"font-family:&quot;Arial&quot;,sans-serif">&#4=
3;1 613 254 3336</span></strong></span><o:p></o:p></p>
</td>
<td width=3D"50%" valign=3D"top" style=3D"width:50.0%;padding:11.25pt 0in 7=
.5pt 15.0pt">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Cisco Systems Canada Co. / Les Systemes Cisco Canada=
 CIE</span></strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&=
quot;,sans-serif;color:#666666"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"color:#666666;text-decoration:none">Cisco.com</span></a></span><o:p></o=
:p></p>
</td>
<td style=3D"padding:0in 0in 0in 0in"></td>
</tr>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in"></td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal">&nbsp;<o:p></o:p></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"400" style=3D"width:300.0pt">
<tbody>
<tr>
<td style=3D"padding:3.75pt 15.0pt 0in .25in">
<p class=3D"MsoNormal"><span style=3D"font-size:7.5pt;font-family:&quot;Ari=
al&quot;,sans-serif;color:#009900;border:solid windowtext 1.0pt;padding:0in=
"><img border=3D"0" width=3D"100" height=3D"100" id=3D"_x0000_i1028" src=3D=
"cid:image002.jpg@01D067EE.7BA7B450" alt=3D"Image removed by sender."></spa=
n><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;c=
olor:#009900">Think
 before you print.</span><o:p></o:p></p>
</td>
</tr>
<tr>
<td style=3D"padding:5.25pt 15.0pt 4.5pt .25in">
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">This email may contain confidential and privileged material =
for the sole use of the intended recipient. Any review, use, distribution o=
r disclosure by others is strictly prohibited.
 If you are not the intended recipient (or authorized to receive for the re=
cipient), please contact the sender by reply email and delete all copies of=
 this message.</span><o:p></o:p></p>
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">Please
<a href=3D"http://www.cisco.com/web/about/doing_business/legal/cri/index.ht=
ml" target=3D"_blank" title=3D"Legal Information">
<span style=3D"color:#0E58A0">click here</span></a> for Company Registratio=
n Information.</span><o:p></o:p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"color:black"><br clear=3D"all">
<o:p></o:p></span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></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:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:
</span></b><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,=
sans-serif;color:black">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.co=
m" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Thursday, March 26, 2015 at 10:24 AM<br>
<b>To: </b>Patrice Brissette &lt;<a href=3D"mailto:pbrisset@cisco.com" targ=
et=3D"_blank">pbrisset@cisco.com</a>&gt;, John E Drake &lt;<a href=3D"mailt=
o:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali Saj=
assi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@cis=
co.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<b>Subject: </b>RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<blockquote style=3D"border:none;border-left:solid #B5C4DF 4.5pt;padding:0i=
n 0in 0in 4.0pt;margin-left:3.75pt;margin-top:5.0pt;margin-right:0in;margin=
-bottom:5.0pt" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">H=
i Patrice,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">U=
p to reception timer traffic disruption in transient phase is one of the is=
sues.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">T=
hanks,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF422254">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> Patrice Brissette (pbrisset) [<a href=3D"mailto:pbri=
sset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black">Weiguo,</span><span style=3D"color:blac=
k"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family:&quot;Calibri&quot;,sans-=
serif;color:black">You mention &quot;</span><span style=3D"color:black">But=
 if your draft have not solved all issues&#8221;,<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Can you explain what Sat=
ya&#8217;s draft is not solving?<o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black">Regards,</span><span style=3D"color:bla=
ck"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><span style=3D"font-s=
ize:10.5pt;font-family:&quot;Calibri&quot;,sans-serif;color:black">Patrice<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td style=3D"padding:0in 0in 0in 0in">
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"border:solid windowtext =
1.0pt;padding:0in"><img border=3D"0" width=3D"110" height=3D"73" id=3D"_x00=
00_i1030" src=3D"cid:image001.jpg@01D067EE.7BA7B450" alt=3D"Image removed b=
y sender."></span><o:p></o:p></p>
</td>
</tr>
<tr>
<td nowrap=3D"" valign=3D"top" style=3D"padding:0in 0in 0in .25in">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Patrice Brissette</span></strong><span style=3D"font=
-size:8.5pt;font-family:&quot;Arial&quot;,sans-serif;color:#666666"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"colo=
r:#666666;text-decoration:none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"font-family:&quot;Arial&quot;,sans-serif">&#4=
3;1 613 254 3336</span></strong></span><o:p></o:p></p>
</td>
<td width=3D"50%" valign=3D"top" style=3D"width:50.0%;padding:11.25pt 0in 7=
.5pt 15.0pt">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Cisco Systems Canada Co. / Les Systemes Cisco Canada=
 CIE</span></strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&=
quot;,sans-serif;color:#666666"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"color:#666666;text-decoration:none">Cisco.com</span></a></span><o:p></o=
:p></p>
</td>
<td style=3D"padding:0in 0in 0in 0in"></td>
</tr>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in"></td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal">&nbsp;<o:p></o:p></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"400" style=3D"width:300.0pt">
<tbody>
<tr>
<td style=3D"padding:3.75pt 15.0pt 0in .25in">
<p class=3D"MsoNormal"><span style=3D"font-size:7.5pt;font-family:&quot;Ari=
al&quot;,sans-serif;color:#009900;border:solid windowtext 1.0pt;padding:0in=
"><img border=3D"0" width=3D"100" height=3D"100" id=3D"_x0000_i1031" src=3D=
"cid:image002.jpg@01D067EE.7BA7B450" alt=3D"Image removed by sender."></spa=
n><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;c=
olor:#009900">Think
 before you print.</span><o:p></o:p></p>
</td>
</tr>
<tr>
<td style=3D"padding:5.25pt 15.0pt 4.5pt .25in">
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">This email may contain confidential and privileged material =
for the sole use of the intended recipient. Any review, use, distribution o=
r disclosure by others is strictly prohibited.
 If you are not the intended recipient (or authorized to receive for the re=
cipient), please contact the sender by reply email and delete all copies of=
 this message.</span><o:p></o:p></p>
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">Please
<a href=3D"http://www.cisco.com/web/about/doing_business/legal/cri/index.ht=
ml" target=3D"_blank" title=3D"Legal Information">
<span style=3D"color:#0E58A0">click here</span></a> for Company Registratio=
n Information.</span><o:p></o:p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black"><br clear=3D"all">
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></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:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:
</span></b><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,=
sans-serif;color:black">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.co=
m" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Wednesday, March 25, 2015 at 10:38 PM<br>
<b>To: </b>John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D=
"_blank">jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:saja=
ssi@cisco.com" target=3D"_blank">sajassi@cisco.com</a>&gt;, &quot;<a href=
=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<b>Subject: </b>Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<blockquote style=3D"border:none;border-left:solid #B5C4DF 4.5pt;padding:0i=
n 0in 0in 4.0pt;margin-left:3.75pt;margin-top:5.0pt;margin-right:0in;margin=
-bottom:5.0pt" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">H=
i John,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">&=
nbsp;Firstly i think EVPN community should reach consensus on the issues of=
 current DF election mechanism. All these issues should be resolved in a si=
ngle new DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">T=
hanks,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF271558">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> John E Drake [<a href=3D"mailto:jdrake@juniper.net" =
target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Weiguo,</span><span style=3D"color:bl=
ack"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Your proposal introduces a control pl=
ane processing load that is O(#EVIs * PEs) per DF election and given that t=
here can be 4K EVIs per ES, this looks like a
 *<b>substantial</b>* load.&nbsp; Furthermore,</span><span style=3D"color:b=
lack"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">you can&#8217;t &nbsp;use the ES rout=
e to co-ordinate DF election because you would need to carry your new exten=
ded community for each EVI and they would not all fit.&nbsp; You
 also can&#8217;t use the Per EVI Ethernet AD route because that is process=
ed by all PEs in the EVI.</span><span style=3D"color:black"><o:p></o:p></sp=
an></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">I think that from a practical perspec=
tive the new DF election proposed in Satya&#8217;s draft is sufficiently st=
able that it renders your draft moot, even if it could
 be made to work.</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Yours Irrespectively,</span><span sty=
le=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">John</span><span style=3D"color:black=
"><o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></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 #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:</span></b><span style=3D"font-=
size:11.0pt;font-family:&quot;Calibri&quot;,sans-serif;color:black"> BESS [=
<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">mailto:bess-boun=
ces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<div>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Hi=
 Ali,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Th=
anks for your information. I scanned through this draft, it really introduc=
es inter-chassis message for DF election handshaking, the requirements in t=
his draft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"col=
or:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Al=
so for accuracy, i think handshaking state machine on each multi-homed PE i=
s also needed. From solution perspective, in my draft, no inter-chassis mes=
sage is introduced, only one new extended community
 is introduced, i think the process is comparatively simple than your follo=
wing draft.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Cu=
rrent EVPN DF election has some drawbacks, so there are three new drafts ab=
out DF election emerged. I think BESS WG can consider these three drafts in=
 global view,&nbsp; a single,comprehensive new DF
 election draft is hoped.</span><span style=3D"color:black"><o:p></o:p></sp=
an></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">th=
anks.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Th=
anks,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">we=
iguo</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF484031">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" targe=
t=3D"_blank">bess-bounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">FYI- First published Jul=
y 4, 2011<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><a href=3D"https://datat=
racker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_blan=
k">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route<=
/a>/<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">-Ali<o:p></o:p></span></=
p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080BLUPR05MB562namprd_--

--_005_BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080BLUPR05MB562namprd_
Content-Type: image/jpeg; name="image001.jpg"
Content-Description: image001.jpg
Content-Disposition: inline; filename="image001.jpg"; size=437;
	creation-date="Thu, 26 Mar 2015 22:00:03 GMT";
	modification-date="Thu, 26 Mar 2015 22:00:03 GMT"
Content-ID: <image001.jpg@01D067EE.7BA7B450>
Content-Transfer-Encoding: base64

/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAoHBwkHBgoJCAkLCwoMDxkQDw4ODx4WFxIZJCAmJSMg
IyIoLTkwKCo2KyIjMkQyNjs9QEBAJjBGS0U+Sjk/QD3/wAALCABJAG4BAREA/8QAHwAAAQUBAQEB
AQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1Fh
ByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZ
WmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXG
x8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/9oACAEBAAA/APZqKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK//9k=

--_005_BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080BLUPR05MB562namprd_
Content-Type: image/jpeg; name="image002.jpg"
Content-Description: image002.jpg
Content-Disposition: inline; filename="image002.jpg"; size=823;
	creation-date="Thu, 26 Mar 2015 22:00:03 GMT";
	modification-date="Thu, 26 Mar 2015 22:00:03 GMT"
Content-ID: <image002.jpg@01D067EE.7BA7B450>
Content-Transfer-Encoding: base64

/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a
HBwgJC4nICIsIxwcKDcpLDAxNDQ0Hyc5PTgyPC4zNDL/2wBDAQkJCQwLDBgNDRgyIRwhMjIyMjIy
MjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjIyMjL/wAARCABkAGQDASIA
AhEBAxEB/8QAHwAAAQUBAQEBAQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQA
AAF9AQIDAAQRBRIhMUEGE1FhByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3
ODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWm
p6ipqrKztLW2t7i5usLDxMXGx8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/8QAHwEA
AwEBAQEBAQEBAQAAAAAAAAECAwQFBgcICQoL/8QAtREAAgECBAQDBAcFBAQAAQJ3AAECAxEEBSEx
BhJBUQdhcRMiMoEIFEKRobHBCSMzUvAVYnLRChYkNOEl8RcYGRomJygpKjU2Nzg5OkNERUZHSElK
U1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6goOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3
uLm6wsPExcbHyMnK0tPU1dbX2Nna4uPk5ebn6Onq8vP09fb3+Pn6/9oADAMBAAIRAxEAPwD3+iii
gAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKA
CiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAK
KKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAoo
ooAKKKKACiiigAooooAKKKKACiiigD//2Q==

--_005_BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080BLUPR05MB562namprd_--


From nobody Thu Mar 26 15:08:11 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0246A1A006D for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 15:08:10 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.61
X-Spam-Level: 
X-Spam-Status: No, score=-3.61 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wxuIAJ7yUraj for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 15:08:03 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 963661A007B for <bess@ietf.org>; Thu, 26 Mar 2015 15:08:01 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml401-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BUE37398; Thu, 26 Mar 2015 22:08:00 +0000 (GMT)
Received: from NKGEML404-HUB.china.huawei.com (10.98.56.35) by lhreml401-hub.china.huawei.com (10.201.5.240) with Microsoft SMTP Server (TLS) id 14.3.158.1; Thu, 26 Mar 2015 22:07:58 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml404-hub.china.huawei.com ([10.98.56.35]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 06:07:46 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>, John E Drake <jdrake@juniper.net>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///gyaAABDJIj8=
Date: Thu, 26 Mar 2015 22:07:45 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857524@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>, <D139E9F1.4EFAC%pbrisset@cisco.com>
In-Reply-To: <D139E9F1.4EFAC%pbrisset@cisco.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.156.0]
Content-Type: multipart/related; boundary="_005_DD5FC8DE455C3348B94340C0AB5517334F857524nkgeml501mbschi_"; type="multipart/alternative"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/9x2G__Xqo4Fzk30Cf6hRMthRT4o>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 22:08:10 -0000

--_005_DD5FC8DE455C3348B94340C0AB5517334F857524nkgeml501mbschi_
Content-Type: multipart/alternative;
	boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F857524nkgeml501mbschi_"

--_000_DD5FC8DE455C3348B94340C0AB5517334F857524nkgeml501mbschi_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Patrice,

Pls see below with [weiguo].

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com]
Sent: Friday, March 27, 2015 5:56
To: Haoweiguo; John E Drake
Cc: Ali Sajassi (sajassi); bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

I sent already too much time of this so I hope this is the last email.

I read again one more time your draft and it is NOT addressing the issue th=
at you keep coming back to.
[weiguo]: Sorry, maybe more time is needed to reach consensus.

See, if you don=92t want any loop, simply bring your interface in blocked s=
tate until DF election converge.
[weiguo]: When EVPN PE boots up, the access interface should be set blocked=
 state, this is common sense and i have doubt. But in DF re-election case, =
PE1 and PE2 are in steady state, PE1 is non-DF and PE2
is DF. After some time, PE3 joins the same ESI, PE3 is elected as DF PE, on=
ly when PE3 upgrades to DF PE, PE2 can degrade to non-DF PE to avoid traffi=
c disruption. This is node failure and recovery scenario(PE3 fails and reco=
vers again). Traffic disruption and loop should be avoided in this scenario=
.

Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 4:36 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Patrice B=
rissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>
Cc: Ali Sajassi <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.o=
rg<mailto:bess@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


John,

As your understanding of the EVPN draft, the DF election mechanism has more=
 serious side effect, it will have short time traffic loop,i.e., dual DF PE=
s will exist for a short time. I think dual DF PEs is absolutely not tolera=
ted, because native ethernet header has no TTL, up to several hundred ms tr=
affic loop normally not tolerated in commertial networks.



As your understanding, the PEs should do as following:

1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have rarely=
 deployment.

Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins ESI an=
d trigger DF re-election. When reception timer expires:

PE1 upgrades to DF PE.

After reception timer+ ES route transmission timer:

PE2 downloads to non-DF PE.

So in timer clock sync case, dual DF PEs will exist at least transmission t=
imer.



If NTP is used for timer sync, because it has bad accuracy, dual DF PEs wil=
l exist more longer timer.



So as your understanding for DF election, the drawback is more clear.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Friday, March 27, 2015 4:41
To: Haoweiguo; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

We have already established that your proposal is untenable because of its =
control plane load.

What we are now discussing is that your proposal is based upon a misunderst=
anding of the algorithm in RFC 7432.  You are assuming that PE1 will advert=
ise an ES route and then wait for the configured interval before performing=
 the DF election while PE2 and PE3 will perform the DF election as soon as =
they receive the ES route from PE1.  This is not what RFC 7432 says.

Rather, what is says is that the advertisement of the ES route by PE1 and i=
ts receipt by PE2 and PE3 causes all three PEs to start the configured inte=
rval timer  - =933. When the timer expires, each PE builds an ordered list =
of the IP addresses of all the PE nodes connected to the Ethernet segment (=
including itself), in increasing numeric value.=94

Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 3:26 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Pls read my detail replies to Satya. If you still can't catch it, pls read =
my draft and EVPN base protocol,  thanks

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Friday, March 27, 2015 1:28
To: Patrice Brissette (pbrisset)
Cc: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:
Weiguo,

I=92m not sure I=92m following here.
Don=92t you have the same issue with your handshaking mechanism?
If you don=92t know your peer, how can you handshake?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_DD5FC8DE455C3348B94340C0AB5517334F857524nkgeml501mbschi_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"FONT-FAMILY: Calibri,sans-serif; WORD-WRAP: break-word; COLO=
R: rgb(0,0,0); FONT-SIZE: 14px" fPStyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3">Patrice,</font></p>
<p><font size=3D"3">Pls see below with [weiguo].</font></p>
<p><font size=3D"3">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF535825"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> Patrice Brissette (pbrisset) [pbriss=
et@cisco.com]<br>
<b>Sent:</b> Friday, March 27, 2015 5:56<br>
<b>To:</b> Haoweiguo; John E Drake<br>
<b>Cc:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>
<div>Weiguo,</div>
<div><br>
</div>
<div>I sent already too much time of this so I hope this is the last email.=
</div>
<div>&nbsp;</div>
<div>I read again one more time your draft and it is NOT addressing the iss=
ue that you keep coming back to.</div>
<div>[weiguo]: Sorry, maybe more time is needed to reach consensus.</div>
<div>&nbsp;</div>
<div>See, if you don=92t want any loop, simply bring your interface in bloc=
ked state until DF election converge.</div>
<div>
<div>[weiguo]: When EVPN PE boots up, the access interface should be set bl=
ocked state, this is common sense and i have doubt. But in DF re-election c=
ase, PE1 and PE2 are in steady state, PE1 is non-DF and PE2
</div>
<div>is DF. After some time, PE3 joins the same ESI, PE3 is elected as DF P=
E, only when PE3 upgrades to DF PE, PE2 can degrade to non-DF PE to avoid t=
raffic disruption. This is node failure and recovery scenario(PE3 fails and=
 recovers again). Traffic disruption
 and loop should be avoided in this scenario.</div>
<div>&nbsp;</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Thursday, March 26, 2015 at =
4:36 PM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>John E Drake &lt;<a href=3D"ma=
ilto:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Patr=
ice Brissette &lt;<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank">p=
brisset@cisco.com</a>&gt;<br>
<span style=3D"FONT-WEIGHT: bold">Cc: </span>Ali Sajassi &lt;<a href=3D"mai=
lto:sajassi@cisco.com" target=3D"_blank">sajassi@cisco.com</a>&gt;, &quot;<=
a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&quot; &=
lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&gt;=
<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>RE: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr"><style>@font-face {
	font-family: MingLiU;
}
@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@font-face {
	font-family: Consolas;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
PRE {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: Consolas; FONT-SIZE: 10pt
}
P.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
LI.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
DIV.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
SPAN.emailstyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
SPAN.EmailStyle21 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
SPAN.HTMLPreformattedChar {
	FONT-FAMILY: Consolas
}
.msochpdefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font size=3D"3" face=3D"Calibri">John,</font></p>
<p><font size=3D"3" face=3D"Calibri">As your understanding of the EVPN draf=
t, the DF election mechanism has more serious side effect, it will have sho=
rt time traffic loop,i.e., dual DF PEs will exist for a short time. I think=
 dual DF PEs is absolutely not tolerated,
 because native ethernet header has no TTL, up to several hundred ms traffi=
c loop normally not tolerated in commertial networks.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">As your understanding, the PEs should =
do as following:</font></p>
<p><font size=3D"3" face=3D"Calibri">1. Accurate timer sync. NTP accuracy i=
s bad, 1588v2 is good but have rarely deployment.</font></p>
<p><font size=3D"3" face=3D"Calibri">Assuming PE1,PE2 and PE3 have consiste=
nt timer clock, when PE3 joins ESI and trigger DF re-election. When recepti=
on timer expires:</font></p>
<p><font size=3D"3" face=3D"Calibri">PE1 upgrades to DF PE.</font></p>
<p><font size=3D"3" face=3D"Calibri">After reception timer&#43; ES route tr=
ansmission timer:</font></p>
<p><font size=3D"3" face=3D"Calibri">PE2 downloads to non-DF PE.</font></p>
<p><font size=3D"3" face=3D"Calibri">So in timer clock sync case, dual DF P=
Es will exist at least transmission timer.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">If NTP is used for timer sync, because=
 it has bad accuracy, dual DF PEs will exist more longer timer.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">So as your understanding for DF electi=
on, the drawback is more clear.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF234411"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net" target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Friday, March 27, 2015 4:41<br>
<b>To:</b> Haoweiguo; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">We have already established that your p=
roposal is untenable because of its control plane load.
</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">What we are now discussing is that your=
 proposal is based upon a misunderstanding of the algorithm in RFC 7432.&nb=
sp; You are assuming that PE1 will advertise
 an ES route and then wait for the configured interval before performing th=
e DF election while PE2 and PE3 will perform the DF election as soon as the=
y receive the ES route from PE1.&nbsp; This is not what RFC 7432 says.</spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Rather, what is says is that the advert=
isement of the ES route by PE1 and its receipt by PE2 and PE3 causes all th=
ree PEs to start the configured interval
 timer&nbsp; - =933. When the timer expires, each PE builds an ordered list=
 of the IP addresses of all the PE nodes connected to the Ethernet segment =
(including itself), in increasing numeric value.=94</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: Calibri,sans-serif; F=
ONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY: Calibri,sans-se=
rif; FONT-SIZE: 11pt"> Haoweiguo [<a href=3D"mailto:haoweiguo@huawei.com" t=
arget=3D"_blank">mailto:haoweiguo@huawei.com</a>]
<br>
<b>Sent:</b> Thursday, March 26, 2015 3:26 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black">Pls read m=
y detail replies to Satya. If you still can't catch it, pls read my draft a=
nd EVPN base protocol, &nbsp;thanks</span><span style=3D"FONT-FAMILY: Tahom=
a,sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black">weiguo</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF478649">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></b>=
<span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10p=
t"> John E Drake [<a href=3D"mailto:jdrake@juniper.net" target=3D"_blank">j=
drake@juniper.net</a>]<br>
<b>Sent:</b> Friday, March 27, 2015 1:28<br>
<b>To:</b> Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">I think Patrice is corr=
ect. &nbsp;Your proposal doesn't solve the problem and it does so at huge c=
ost.<br>
<br>
Sent from my iPhone</span></p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"COLOR: =
black"><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt; wrot=
e:</span></p>
</div>
<blockquote style=3D"MARGIN-TOP: 5pt; MARGIN-BOTTOM: 5pt">
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Weiguo,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">I=92m not sure I=92m fo=
llowing here.</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Don=92t you have the sa=
me issue with your handshaking mechanism?&nbsp;</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">If you don=92t know you=
r peer, how can you handshake?</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Regards,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"COLOR: =
black">Patrice</span></p>
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"BORDER-BOTTOM: windowtex=
t 1pt solid; BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDIN=
G-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: windowtext 1pt solid; BORDER-R=
IGHT: windowtext 1pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1026" bord=
er=3D"0" alt=3D"Image removed by sender." src=3D"cid:image001.jpg@01D067DF.=
BB3C17F0" width=3D"110" height=3D"73"></span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 0in;=
 PADDING-TOP: 0in" valign=3D"top" nowrap=3D"">
<p><strong><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: rgb(102,102=
,102); FONT-SIZE: 8.5pt">Patrice Brissette</span></strong><span style=3D"FO=
NT-FAMILY: Arial,sans-serif; COLOR: rgb(102,102,102); FONT-SIZE: 8.5pt"><br=
>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"COLO=
R: #666666; TEXT-DECORATION: none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"FONT-FAMILY: Arial,sans-serif">&#43;1 613 254=
 3336</span></strong></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 7.5pt; PADDING-LEFT: 15pt; WIDTH: 50%; PADDING=
-RIGHT: 0in; PADDING-TOP: 11.25pt" valign=3D"top" width=3D"50%">
<p><strong><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: rgb(102,102=
,102); FONT-SIZE: 8.5pt">Cisco Systems Canada Co. / Les Systemes Cisco Cana=
da CIE</span></strong><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: =
rgb(102,102,102); FONT-SIZE: 8.5pt"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"COLOR: #666666; TEXT-DECORATION: none">Cisco.com</span></a></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span></span>&nbsp;</p>
<table style=3D"WIDTH: 300pt" class=3D"MsoNormalTable" border=3D"0" cellspa=
cing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15pt=
; PADDING-TOP: 3.75pt">
<p class=3D"MsoNormal"><span style=3D"BORDER-BOTTOM: windowtext 1pt solid; =
BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; =
PADDING-RIGHT: 0in; FONT-FAMILY: Arial,sans-serif; COLOR: rgb(0,153,0); FON=
T-SIZE: 7.5pt; BORDER-TOP: windowtext 1pt solid; BORDER-RIGHT: windowtext 1=
pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1027" border=3D"0" alt=3D"Im=
age removed by sender." src=3D"cid:~WRD000.jpg" width=3D"100" height=3D"100=
"></span><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: rgb(0,153,0);=
 FONT-SIZE: 7.5pt">Think
 before you print.</span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 4.5pt; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15=
pt; PADDING-TOP: 5.25pt">
<p><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: rgb(153,153,153); F=
ONT-SIZE: 7.5pt">This email may contain confidential and privileged materia=
l for the sole use of the intended recipient. Any review, use, distribution=
 or disclosure by others is strictly
 prohibited. If you are not the intended recipient (or authorized to receiv=
e for the recipient), please contact the sender by reply email and delete a=
ll copies of this message.</span></p>
<p><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: rgb(153,153,153); F=
ONT-SIZE: 7.5pt">Please
<a title=3D"Legal Information" href=3D"http://www.cisco.com/web/about/doing=
_business/legal/cri/index.html" target=3D"_blank">
<span style=3D"COLOR: #0e58a0">click here</span></a> for Company Registrati=
on Information.</span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><br clear=3D"all">
</span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: Calibri,sans-serif; C=
OLOR: black; FONT-SIZE: 11pt">From:
</span></b><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black; FO=
NT-SIZE: 11pt">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.com" target=
=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Thursday, March 26, 2015 at 10:24 AM<br>
<b>To: </b>Patrice Brissette &lt;<a href=3D"mailto:pbrisset@cisco.com" targ=
et=3D"_blank">pbrisset@cisco.com</a>&gt;, John E Drake &lt;<a href=3D"mailt=
o:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali Saj=
assi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@cis=
co.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<b>Subject: </b>RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<blockquote style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: #b5c4df 4.5pt=
 solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; MARGIN-=
LEFT: 3.75pt; BORDER-TOP: medium none; MARGIN-RIGHT: 0in; BORDER-RIGHT: med=
ium none; PADDING-TOP: 0in" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black">Hi Patrice=
,</span><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-S=
IZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black">Up to rece=
ption timer traffic disruption in transient phase is one of the issues.</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black">Thanks,</s=
pan><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE:=
 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black">weiguo</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF422254">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></b>=
<span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10p=
t"> Patrice Brissette (pbrisset) [<a href=3D"mailto:pbrisset@cisco.com" tar=
get=3D"_blank">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: black; FONT-SIZE: 10.5pt">Weiguo,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: black">You mention &quot;</span><span style=3D"COLOR: black">But if your=
 draft have not solved all issues=94,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Can you explain what Sa=
tya=92s draft is not solving?</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: black; FONT-SIZE: 10.5pt">Regards,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"FONT-FA=
MILY: Calibri,sans-serif; COLOR: black; FONT-SIZE: 10.5pt">Patrice</span></=
p>
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"BORDER-BOTTOM: windowtex=
t 1pt solid; BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDIN=
G-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: windowtext 1pt solid; BORDER-R=
IGHT: windowtext 1pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1029" bord=
er=3D"0" alt=3D"Image removed by sender." src=3D"cid:image001.jpg@01D067DF.=
BB3C17F0" width=3D"110" height=3D"73"></span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 0in;=
 PADDING-TOP: 0in" valign=3D"top" nowrap=3D"">
<p><strong><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: rgb(102,102=
,102); FONT-SIZE: 8.5pt">Patrice Brissette</span></strong><span style=3D"FO=
NT-FAMILY: Arial,sans-serif; COLOR: rgb(102,102,102); FONT-SIZE: 8.5pt"><br=
>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"COLO=
R: #666666; TEXT-DECORATION: none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"FONT-FAMILY: Arial,sans-serif">&#43;1 613 254=
 3336</span></strong></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 7.5pt; PADDING-LEFT: 15pt; WIDTH: 50%; PADDING=
-RIGHT: 0in; PADDING-TOP: 11.25pt" valign=3D"top" width=3D"50%">
<p><strong><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: rgb(102,102=
,102); FONT-SIZE: 8.5pt">Cisco Systems Canada Co. / Les Systemes Cisco Cana=
da CIE</span></strong><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: =
rgb(102,102,102); FONT-SIZE: 8.5pt"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"COLOR: #666666; TEXT-DECORATION: none">Cisco.com</span></a></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span></span>&nbsp;</p>
<table style=3D"WIDTH: 300pt" class=3D"MsoNormalTable" border=3D"0" cellspa=
cing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15pt=
; PADDING-TOP: 3.75pt">
<p class=3D"MsoNormal"><span style=3D"BORDER-BOTTOM: windowtext 1pt solid; =
BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; =
PADDING-RIGHT: 0in; FONT-FAMILY: Arial,sans-serif; COLOR: rgb(0,153,0); FON=
T-SIZE: 7.5pt; BORDER-TOP: windowtext 1pt solid; BORDER-RIGHT: windowtext 1=
pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1030" border=3D"0" alt=3D"Im=
age removed by sender." src=3D"cid:~WRD000.jpg" width=3D"100" height=3D"100=
"></span><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: rgb(0,153,0);=
 FONT-SIZE: 7.5pt">Think
 before you print.</span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 4.5pt; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15=
pt; PADDING-TOP: 5.25pt">
<p><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: rgb(153,153,153); F=
ONT-SIZE: 7.5pt">This email may contain confidential and privileged materia=
l for the sole use of the intended recipient. Any review, use, distribution=
 or disclosure by others is strictly
 prohibited. If you are not the intended recipient (or authorized to receiv=
e for the recipient), please contact the sender by reply email and delete a=
ll copies of this message.</span></p>
<p><span style=3D"FONT-FAMILY: Arial,sans-serif; COLOR: rgb(153,153,153); F=
ONT-SIZE: 7.5pt">Please
<a title=3D"Legal Information" href=3D"http://www.cisco.com/web/about/doing=
_business/legal/cri/index.html" target=3D"_blank">
<span style=3D"COLOR: #0e58a0">click here</span></a> for Company Registrati=
on Information.</span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: black; FONT-SIZE: 10.5pt"><br clear=3D"all">
</span></p>
</div>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: Calibri,sans-serif; C=
OLOR: black; FONT-SIZE: 11pt">From:
</span></b><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black; FO=
NT-SIZE: 11pt">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.com" target=
=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Wednesday, March 25, 2015 at 10:38 PM<br>
<b>To: </b>John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D=
"_blank">jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:saja=
ssi@cisco.com" target=3D"_blank">sajassi@cisco.com</a>&gt;, &quot;<a href=
=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<b>Subject: </b>Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: black; FONT-SIZE: 10.5pt"></span>&nbsp;</p>
</div>
<blockquote style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: #b5c4df 4.5pt=
 solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; MARGIN-=
LEFT: 3.75pt; BORDER-TOP: medium none; MARGIN-RIGHT: 0in; BORDER-RIGHT: med=
ium none; PADDING-TOP: 0in" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black">Hi John,</=
span><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE=
: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black">&nbsp;Firs=
tly i think EVPN community should reach consensus on the issues of current =
DF election mechanism. All these issues should be resolved in a single new =
DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</span><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR:=
 black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black">Thanks,</s=
pan><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE:=
 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLOR: black">weiguo</sp=
an><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: =
10pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF271558">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></b>=
<span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10p=
t"> John E Drake [<a href=3D"mailto:jdrake@juniper.net" target=3D"_blank">j=
drake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Weiguo,</span><span style=3D"COLOR: bla=
ck"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Your proposal introduces a control plan=
e processing load that is O(#EVIs * PEs) per DF election and given that the=
re can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span><span style=3D"=
COLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">you can=92t &nbsp;use the ES route to c=
o-ordinate DF election because you would need to carry your new extended co=
mmunity for each EVI and they would not all
 fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because that=
 is processed by all PEs in the EVI.</span><span style=3D"COLOR: black"></s=
pan></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">I think that from a practical perspecti=
ve the new DF election proposed in Satya=92s draft is sufficiently stable t=
hat it renders your draft moot, even if
 it could be made to work.</span><span style=3D"COLOR: black"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">Yours Irrespectively,</span><span style=
=3D"COLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: Calibri,sans-serif; COLO=
R: rgb(31,73,125); FONT-SIZE: 11pt">John</span><span style=3D"COLOR: black"=
></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: Calibri,sans-serif; C=
OLOR: black; FONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY: C=
alibri,sans-serif; COLOR: black; FONT-SIZE: 11pt"> BESS [<a href=3D"mailto:=
bess-bounces@ietf.org" target=3D"_blank">mailto:bess-bounces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Hi Ali,</sp=
an><span style=3D"COLOR: black"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks for =
your information. I scanned through this draft, it really introduces inter-=
chassis message for DF election handshaking, the requirements in this draft=
 is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"COL=
OR: black"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Also for ac=
curacy, i think handshaking state machine on each multi-homed PE is also ne=
eded. From solution perspective, in my draft, no inter-chassis message is i=
ntroduced, only one new extended community
 is introduced, i think the process is comparatively simple than your follo=
wing draft.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Current EVP=
N DF election has some drawbacks, so there are three new drafts about DF el=
ection emerged. I think BESS WG can consider these three drafts in global v=
iew,&nbsp; a single,comprehensive new DF
 election draft is hoped.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">thanks.</sp=
an><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">Thanks,</sp=
an><span style=3D"COLOR: black"></span></p>
<p><span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black">weiguo</spa=
n><span style=3D"COLOR: black"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></b>=
<span style=3D"FONT-FAMILY: Tahoma,sans-serif; COLOR: black; FONT-SIZE: 10p=
t"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">bess-b=
ounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F857524nkgeml501mbschi_--

--_005_DD5FC8DE455C3348B94340C0AB5517334F857524nkgeml501mbschi_
Content-Type: image/jpeg; name="~WRD000.jpg"
Content-Description: ~WRD000.jpg
Content-Disposition: inline; filename="~WRD000.jpg"; size=823;
	creation-date="Thu, 26 Mar 2015 21:56:37 GMT";
	modification-date="Thu, 26 Mar 2015 21:56:37 GMT"
Content-ID: <~WRD000.jpg>
Content-Transfer-Encoding: base64
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--_005_DD5FC8DE455C3348B94340C0AB5517334F857524nkgeml501mbschi_
Content-Type: image/jpeg; name="image001.jpg"
Content-Description: image001.jpg
Content-Disposition: inline; filename="image001.jpg"; size=437;
	creation-date="Thu, 26 Mar 2015 21:56:37 GMT";
	modification-date="Thu, 26 Mar 2015 21:56:37 GMT"
Content-ID: <image001.jpg@01D067DF.BB3C17F0>
Content-Transfer-Encoding: base64

/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAoHBwkHBgoJCAkLCwoMDxkQDw4ODx4WFxIZJCAmJSMg
IyIoLTkwKCo2KyIjMkQyNjs9QEBAJjBGS0U+Sjk/QD3/wAALCABJAG4BAREA/8QAHwAAAQUBAQEB
AQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1Fh
ByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZ
WmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXG
x8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/9oACAEBAAA/APZqKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK//9k=

--_005_DD5FC8DE455C3348B94340C0AB5517334F857524nkgeml501mbschi_--


From nobody Thu Mar 26 15:12:15 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D36931A00BD for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 15:12:13 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.309
X-Spam-Level: 
X-Spam-Status: No, score=-6.309 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UISA48FI_RJt for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 15:12:08 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 705A31A000D for <bess@ietf.org>; Thu, 26 Mar 2015 15:12:07 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id 7643952B3C463; Thu, 26 Mar 2015 22:12:01 +0000 (GMT)
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2QMC4WF004710 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 26 Mar 2015 23:12:05 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR712WXCHHUB03.zeu.alcatel-lucent.com ([135.239.2.74]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 23:12:04 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Haoweiguo <haoweiguo@huawei.com>, "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>, John E Drake <jdrake@juniper.net>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ1HeqN8Ih9YQqE+v8/YVcLeCH50uDT8AgACbTwCAACnuAIAAAsqAgAAf4ACAACCzAIAAFSKAgAAPjQCAAAWHgIAAAxyA//+L1oA=
Date: Thu, 26 Mar 2015 22:12:03 +0000
Message-ID: <D139D1D9.6C77E%jorge.rabadan@alcatel-lucent.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <D139E9F1.4EFAC%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857524@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857524@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.40]
Content-Type: multipart/mixed; boundary="_005_D139D1D96C77Ejorgerabadanalcatellucentcom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/TSPIziq40oIRpJWtcY_C0-y_tI8>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 22:12:14 -0000

--_005_D139D1D96C77Ejorgerabadanalcatellucentcom_
Content-Type: multipart/alternative;
	boundary="_000_D139D1D96C77Ejorgerabadanalcatellucentcom_"

--_000_D139D1D96C77Ejorgerabadanalcatellucentcom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_D139D1D96C77Ejorgerabadanalcatellucentcom_
Content-Type: text/html; charset="utf-8"
Content-ID: <B602963AE0E39E44B2A9F6AE2A968596@exchange.lucent.com>
Content-Transfer-Encoding: base64
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--_000_D139D1D96C77Ejorgerabadanalcatellucentcom_--

--_005_D139D1D96C77Ejorgerabadanalcatellucentcom_
Content-Type: image/jpeg; name="~WRD000.jpg"
Content-Description: ~WRD000.jpg
Content-Disposition: attachment; filename="~WRD000.jpg"; size=823;
	creation-date="Thu, 26 Mar 2015 22:12:03 GMT";
	modification-date="Thu, 26 Mar 2015 22:12:03 GMT"
Content-ID: <~WRD000.jpg>
Content-Transfer-Encoding: base64
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--_005_D139D1D96C77Ejorgerabadanalcatellucentcom_
Content-Type: image/jpeg; name="image001.jpg"
Content-Description: image001.jpg
Content-Disposition: attachment; filename="image001.jpg"; size=437;
	creation-date="Thu, 26 Mar 2015 22:12:03 GMT";
	modification-date="Thu, 26 Mar 2015 22:12:03 GMT"
Content-ID: <image001.jpg@01D067DF.BB3C17F0>
Content-Transfer-Encoding: base64

/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAoHBwkHBgoJCAkLCwoMDxkQDw4ODx4WFxIZJCAmJSMg
IyIoLTkwKCo2KyIjMkQyNjs9QEBAJjBGS0U+Sjk/QD3/wAALCABJAG4BAREA/8QAHwAAAQUBAQEB
AQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1Fh
ByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZ
WmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXG
x8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/9oACAEBAAA/APZqKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK//9k=

--_005_D139D1D96C77Ejorgerabadanalcatellucentcom_--


From nobody Thu Mar 26 15:13:30 2015
Return-Path: <pbrisset@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B067E1A00E8 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 15:13:29 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -13.91
X-Spam-Level: 
X-Spam-Status: No, score=-13.91 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mEquJSfbdv4n for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 15:13:24 -0700 (PDT)
Received: from alln-iport-5.cisco.com (alln-iport-5.cisco.com [173.37.142.92]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3AEB41A00EC for <bess@ietf.org>; Thu, 26 Mar 2015 15:13:24 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=65800; q=dns/txt; s=iport; t=1427408005; x=1428617605; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=F6yo3WsQ7NKrDr4Nrxi6NBF4eOOwqyrv8dS4c/aXTAA=; b=L9hVPTC5/nnmPvFnGo1NawnybeHcF75ONLENtUUuTvnFHbjzrmkKj14P AujJbrkn3dYww+7Tds31uAIb1YmKODF//XJ28NpGMrKOmGYwQ+E3Pkh22 ipWw/bZtkFppad7D4xWOUFPOzBH/dIQOTzMPBpwrXAbRqpxsvyMkijVOi Y=;
X-Files: ~WRD000.jpg, image001.jpg : 823, 437
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0A6CABGhBRV/4gNJK1ZAxaCLUNSWgS1OY05gj15hHoCgUZMAQEBAQEBfYQUAQEBBAUVCwk4AwEPDgICAQgRAwEBAQYBAQEKDgEGBwIVBAkCDBQJCAIEAQ0EAQ4GB4gUDbMfmHYBAQEBAQEBAQEBAQEBAQEBAQEBAQEXBIExiXOEFQERAQ4WBgsJAQEGBgQGAQYDCIJpgTMFjkKCDoFpgTIBUx+FYYEbOoJ2BYJViUGDRyKCAR2BUG8BgQMHFyJ/AQEB
X-IronPort-AV: E=Sophos;i="5.11,475,1422921600";  d="jpg'145?scan'145,208,217,145";a="135789328"
Received: from alln-core-3.cisco.com ([173.36.13.136]) by alln-iport-5.cisco.com with ESMTP; 26 Mar 2015 22:13:24 +0000
Received: from xhc-aln-x14.cisco.com (xhc-aln-x14.cisco.com [173.36.12.88]) by alln-core-3.cisco.com (8.14.5/8.14.5) with ESMTP id t2QMDMIo007536 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 26 Mar 2015 22:13:22 GMT
Received: from xmb-aln-x09.cisco.com ([169.254.4.43]) by xhc-aln-x14.cisco.com ([173.36.12.88]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 17:13:22 -0500
From: "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
To: Haoweiguo <haoweiguo@huawei.com>, John E Drake <jdrake@juniper.net>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQaBISWs9VLcZ0TUWO8xI+wRZimQ==
Date: Thu, 26 Mar 2015 22:13:22 +0000
Message-ID: <D139EE3A.4EFCF%pbrisset@cisco.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <D139E9F1.4EFAC%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857524@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857524@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [10.89.7.222]
Content-Type: multipart/mixed; boundary="_005_D139EE3A4EFCFpbrissetciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/i6kX3GlKWvIk5OoGuAsniV9H1VI>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 22:13:29 -0000

--_005_D139EE3A4EFCFpbrissetciscocom_
Content-Type: multipart/alternative;
	boundary="_000_D139EE3A4EFCFpbrissetciscocom_"

--_000_D139EE3A4EFCFpbrissetciscocom_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Weiguo,

"only when PE3 upgrades to DF PE, PE2 can degrade to non-DF PE to avoid tra=
ffic disruption=94

If you follow that, a loop is being created! For some time, PE3 and PE2 wil=
l be both DF.

Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 5:07 PM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>
Cc: Ali Sajassi <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.o=
rg<mailto:bess@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Patrice,

Pls see below with [weiguo].

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Friday, March 27, 2015 5:56
To: Haoweiguo; John E Drake
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

I sent already too much time of this so I hope this is the last email.

I read again one more time your draft and it is NOT addressing the issue th=
at you keep coming back to.
[weiguo]: Sorry, maybe more time is needed to reach consensus.

See, if you don=92t want any loop, simply bring your interface in blocked s=
tate until DF election converge.
[weiguo]: When EVPN PE boots up, the access interface should be set blocked=
 state, this is common sense and i have doubt. But in DF re-election case, =
PE1 and PE2 are in steady state, PE1 is non-DF and PE2
is DF. After some time, PE3 joins the same ESI, PE3 is elected as DF PE, on=
ly when PE3 upgrades to DF PE, PE2 can degrade to non-DF PE to avoid traffi=
c disruption. This is node failure and recovery scenario(PE3 fails and reco=
vers again). Traffic disruption and loop should be avoided in this scenario=
.

Regards,

Patrice

   [http://www.cisco.com/web/europe/images/email/signature/est2014/logo_06.=
png?ct=3D1406640631632]

Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif] Think before =
you print.

This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 4:36 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Patrice B=
rissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>
Cc: Ali Sajassi <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.o=
rg<mailto:bess@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


John,

As your understanding of the EVPN draft, the DF election mechanism has more=
 serious side effect, it will have short time traffic loop,i.e., dual DF PE=
s will exist for a short time. I think dual DF PEs is absolutely not tolera=
ted, because native ethernet header has no TTL, up to several hundred ms tr=
affic loop normally not tolerated in commertial networks.



As your understanding, the PEs should do as following:

1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have rarely=
 deployment.

Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins ESI an=
d trigger DF re-election. When reception timer expires:

PE1 upgrades to DF PE.

After reception timer+ ES route transmission timer:

PE2 downloads to non-DF PE.

So in timer clock sync case, dual DF PEs will exist at least transmission t=
imer.



If NTP is used for timer sync, because it has bad accuracy, dual DF PEs wil=
l exist more longer timer.



So as your understanding for DF election, the drawback is more clear.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Friday, March 27, 2015 4:41
To: Haoweiguo; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Weiguo,

We have already established that your proposal is untenable because of its =
control plane load.

What we are now discussing is that your proposal is based upon a misunderst=
anding of the algorithm in RFC 7432.  You are assuming that PE1 will advert=
ise an ES route and then wait for the configured interval before performing=
 the DF election while PE2 and PE3 will perform the DF election as soon as =
they receive the ES route from PE1.  This is not what RFC 7432 says.

Rather, what is says is that the advertisement of the ES route by PE1 and i=
ts receipt by PE2 and PE3 causes all three PEs to start the configured inte=
rval timer  - =933. When the timer expires, each PE builds an ordered list =
of the IP addresses of all the PE nodes connected to the Ethernet segment (=
including itself), in increasing numeric value.=94

Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 3:26 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Pls read my detail replies to Satya. If you still can't catch it, pls read =
my draft and EVPN base protocol,  thanks

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Friday, March 27, 2015 1:28
To: Patrice Brissette (pbrisset)
Cc: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:
Weiguo,

I=92m not sure I=92m following here.
Don=92t you have the same issue with your handshaking mechanism?
If you don=92t know your peer, how can you handshake?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_D139EE3A4EFCFpbrissetciscocom_
Content-Type: text/html; charset="Windows-1252"
Content-ID: <370E8E37A2EB3D41B528436F809F5D53@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
</head>
<body style=3D"word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-lin=
e-break: after-white-space;">
<div>
<div>
<div style=3D"color: rgb(0, 0, 0); font-family: Calibri, sans-serif; font-s=
ize: 14px;">
Weiguo,</div>
<div style=3D"color: rgb(0, 0, 0); font-family: Calibri, sans-serif; font-s=
ize: 14px;">
<br>
</div>
<div><font face=3D"Calibri,sans-serif">&quot;</font><span style=3D"color: r=
gb(0, 0, 0); font-family: 'Times New Roman'; font-size: 16px;">only when PE=
3 upgrades to DF PE, PE2 can degrade to non-DF PE to avoid traffic disrupti=
on</span><font face=3D"Times New Roman"><span style=3D"font-size: 16px;">=
=94</span></font></div>
<div style=3D"color: rgb(0, 0, 0); font-family: Calibri, sans-serif; font-s=
ize: 14px;">
<span style=3D"font-family: 'Times New Roman'; font-size: 16px;"><br>
</span></div>
<div style=3D"color: rgb(0, 0, 0); font-family: Calibri, sans-serif; font-s=
ize: 14px;">
<span style=3D"font-family: 'Times New Roman'; font-size: 16px;">If you fol=
low that, a loop is being created! For some time, PE3 and PE2 will be both =
DF.</span></div>
<div style=3D"color: rgb(0, 0, 0); font-family: Calibri, sans-serif; font-s=
ize: 14px;">
<div><br>
</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"padding-left: 24px; padding-bottom: 0;" align=3D"left" nowrap=
=3D"nowrap" valign=3D"top">
<p style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 11px; col=
or: rgb(102, 102, 102);">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"color:#666666; text-decoration: none;" href=3D"mailto:pbrisset@=
cisco.com">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"padding-left: 20px; padding-bottom: 10px; padding-top: 15px;" =
valign=3D"top" width=3D"50%">
<p style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 11px; col=
or: rgb(102, 102, 102);">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"color:#666666; text-decoration: none;" href=3D"http://www.cisco=
.com/global/CA/">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 10px; co=
lor: #999999; padding-left: 24px; padding-right: 20px; padding-top: 5px; pa=
dding-bottom: 0px; color:#009900; ">
<img src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif" al=
t=3D""> Think before you print.</td>
</tr>
<tr>
<td style=3D"font-family: Arial, Helvetica, sans-serif; font-size: 10px; co=
lor: #999999; padding-left: 24px; padding-right: 20px; padding-top: 7px; pa=
dding-bottom: 6px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a href=3D"http://www.cisco.com/web/about/doing_business/legal/cr=
i/index.html" title=3D"Legal Information" style=3D"color: #0E58A0;">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
</div>
<div style=3D"color: rgb(0, 0, 0); font-family: Calibri, sans-serif; font-s=
ize: 14px;">
<br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION" style=3D"color: rgb(0, 0, 0); font-family=
: Calibri, sans-serif; font-size: 14px;">
<div style=3D"font-family:Calibri; font-size:11pt; text-align:left; color:b=
lack; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM:=
 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid;=
 BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"font-weight:bold">From: </span>Haoweiguo &lt;<a href=3D"mail=
to:haoweiguo@huawei.com">haoweiguo@huawei.com</a>&gt;<br>
<span style=3D"font-weight:bold">Date: </span>Thursday, March 26, 2015 at 5=
:07 PM<br>
<span style=3D"font-weight:bold">To: </span>Patrice Brissette &lt;<a href=
=3D"mailto:pbrisset@cisco.com">pbrisset@cisco.com</a>&gt;, John E Drake &lt=
;<a href=3D"mailto:jdrake@juniper.net">jdrake@juniper.net</a>&gt;<br>
<span style=3D"font-weight:bold">Cc: </span>Ali Sajassi &lt;<a href=3D"mail=
to:sajassi@cisco.com">sajassi@cisco.com</a>&gt;, &quot;<a href=3D"mailto:be=
ss@ietf.org">bess@ietf.org</a>&quot; &lt;<a href=3D"mailto:bess@ietf.org">b=
ess@ietf.org</a>&gt;<br>
<span style=3D"font-weight:bold">Subject: </span>RE: [bess] Handshaking amo=
ng PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE" style=3D"BORDER-LEFT:=
 #b5c4df 5 solid; PADDING:0 0 0 5; MARGIN:0 0 0 5;">
<div dir=3D"ltr">
<div style=3D"FONT-FAMILY: Calibri,sans-serif; WORD-WRAP: break-word; COLOR=
: rgb(0,0,0); FONT-SIZE: 14px" fpstyle=3D"1" ocsi=3D"0">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3">Patrice,</font></p>
<p><font size=3D"3">Pls see below with [weiguo].</font></p>
<p><font size=3D"3">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF535825"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> Patrice Brissette (pbrisset) [<a hre=
f=3D"mailto:pbrisset@cisco.com">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Friday, March 27, 2015 5:56<br>
<b>To:</b> Haoweiguo; John E Drake<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org">bess@iet=
f.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div>
<div>Weiguo,</div>
<div><br>
</div>
<div>I sent already too much time of this so I hope this is the last email.=
</div>
<div>&nbsp;</div>
<div>I read again one more time your draft and it is NOT addressing the iss=
ue that you keep coming back to.</div>
<div>[weiguo]: Sorry, maybe more time is needed to reach consensus.</div>
<div>&nbsp;</div>
<div>See, if you don=92t want any loop, simply bring your interface in bloc=
ked state until DF election converge.</div>
<div>
<div>[weiguo]: When EVPN PE boots up, the access interface should be set bl=
ocked state, this is common sense and i have doubt. But in DF re-election c=
ase, PE1 and PE2 are in steady state, PE1 is non-DF and PE2
</div>
<div>is DF. After some time, PE3 joins the same ESI, PE3 is elected as DF P=
E, only when PE3 upgrades to DF PE, PE2 can degrade to non-DF PE to avoid t=
raffic disruption. This is node failure and recovery scenario(PE3 fails and=
 recovers again). Traffic disruption
 and loop should be avoided in this scenario.</div>
<div>&nbsp;</div>
<div>Regards,</div>
<div><br>
</div>
<div>Patrice<br>
<br>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td colspan=3D"3">&nbsp; &nbsp;<img src=3D"http://www.cisco.com/web/europe/=
images/email/signature/est2014/logo_06.png?ct=3D1406640631632" width=3D"110=
" height=3D"73"></td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px" valign=3D"top" nowrap=
=3D"" align=3D"left">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Patrice Brissette</strong><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"mailto:pbrisset@=
cisco.com" target=3D"_blank">pbrisset@cisco.com</a><br>
Phone: <strong>&#43;1 613 254 3336</strong><br>
</p>
</td>
<td style=3D"PADDING-BOTTOM: 10px; PADDING-LEFT: 20px; PADDING-TOP: 15px" v=
align=3D"top" width=3D"50%">
<p style=3D"FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: rgb(102,102,102=
); FONT-SIZE: 11px">
<strong>Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE</strong><b=
r>
Canada<br>
<a style=3D"COLOR: #666666; TEXT-DECORATION: none" href=3D"http://www.cisco=
.com/global/CA/" target=3D"_blank">Cisco.com</a><br>
</p>
</td>
</tr>
<tr>
<td colspan=3D"3"></td>
</tr>
</tbody>
</table>
<table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #009900; FONT-SIZE: 10px; P=
ADDING-TOP: 5px">
<img alt=3D"" src=3D"http://www.cisco.com/assets/swa/img/thinkbeforeyouprin=
t.gif"> Think before you print.</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 6px; PADDING-LEFT: 24px; PADDING-RIGHT: 20px; =
FONT-FAMILY: Arial,Helvetica,sans-serif; COLOR: #999999; FONT-SIZE: 10px; P=
ADDING-TOP: 7px">
<p>This email may contain confidential and privileged material for the sole=
 use of the intended recipient. Any review, use, distribution or disclosure=
 by others is strictly prohibited. If you are not the intended recipient (o=
r authorized to receive for the
 recipient), please contact the sender by reply email and delete all copies=
 of this message.</p>
<p>Please <a style=3D"COLOR: #0e58a0" title=3D"Legal Information" href=3D"h=
ttp://www.cisco.com/web/about/doing_business/legal/cri/index.html" target=
=3D"_blank">
click here</a> for Company Registration Information.</p>
<p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<br clear=3D"all">
</div>
</div>
</div>
<div><br>
</div>
<span id=3D"OLK_SRC_BODY_SECTION">
<div style=3D"BORDER-BOTTOM: medium none; TEXT-ALIGN: left; BORDER-LEFT: me=
dium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; FONT=
-FAMILY: Calibri; COLOR: black; FONT-SIZE: 11pt; BORDER-TOP: #b5c4df 1pt so=
lid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style=3D"FONT-WEIGHT: bold">From: </span>Haoweiguo &lt;<a href=3D"mai=
lto:haoweiguo@huawei.com" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br=
>
<span style=3D"FONT-WEIGHT: bold">Date: </span>Thursday, March 26, 2015 at =
4:36 PM<br>
<span style=3D"FONT-WEIGHT: bold">To: </span>John E Drake &lt;<a href=3D"ma=
ilto:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Patr=
ice Brissette &lt;<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank">p=
brisset@cisco.com</a>&gt;<br>
<span style=3D"FONT-WEIGHT: bold">Cc: </span>Ali Sajassi &lt;<a href=3D"mai=
lto:sajassi@cisco.com" target=3D"_blank">sajassi@cisco.com</a>&gt;, &quot;<=
a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&quot; &=
lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&gt;=
<br>
<span style=3D"FONT-WEIGHT: bold">Subject: </span>RE: [bess] Handshaking am=
ong PEs in an EVPN ES based on Paxos algorithm<br>
</div>
<div><br>
</div>
<blockquote style=3D"BORDER-LEFT: #b5c4df 5px solid; PADDING-BOTTOM: 0px; M=
ARGIN: 0px 0px 0px 5px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px; PADDING-TOP:=
 0px" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div dir=3D"ltr"><style>@font-face {
	font-family: MingLiU;
}
@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@font-face {
	font-family: Consolas;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
PRE {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: Consolas; FONT-SIZE: 10pt
}
P.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
LI.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
DIV.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
SPAN.emailstyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
SPAN.EmailStyle21 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
SPAN.HTMLPreformattedChar {
	FONT-FAMILY: Consolas
}
.msochpdefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
<div lang=3D"EN-US">
<div style=3D"FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZ=
E: 10pt">
<p><font size=3D"3" face=3D"Calibri">John,</font></p>
<p><font size=3D"3" face=3D"Calibri">As your understanding of the EVPN draf=
t, the DF election mechanism has more serious side effect, it will have sho=
rt time traffic loop,i.e., dual DF PEs will exist for a short time. I think=
 dual DF PEs is absolutely not tolerated,
 because native ethernet header has no TTL, up to several hundred ms traffi=
c loop normally not tolerated in commertial networks.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">As your understanding, the PEs should =
do as following:</font></p>
<p><font size=3D"3" face=3D"Calibri">1. Accurate timer sync. NTP accuracy i=
s bad, 1588v2 is good but have rarely deployment.</font></p>
<p><font size=3D"3" face=3D"Calibri">Assuming PE1,PE2 and PE3 have consiste=
nt timer clock, when PE3 joins ESI and trigger DF re-election. When recepti=
on timer expires:</font></p>
<p><font size=3D"3" face=3D"Calibri">PE1 upgrades to DF PE.</font></p>
<p><font size=3D"3" face=3D"Calibri">After reception timer&#43; ES route tr=
ansmission timer:</font></p>
<p><font size=3D"3" face=3D"Calibri">PE2 downloads to non-DF PE.</font></p>
<p><font size=3D"3" face=3D"Calibri">So in timer clock sync case, dual DF P=
Es will exist at least transmission timer.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">If NTP is used for timer sync, because=
 it has bad accuracy, dual DF PEs will exist more longer timer.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">So as your understanding for DF electi=
on, the drawback is more clear.</font></p>
<p>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF234411"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [<a href=3D"mailto:jdra=
ke@juniper.net" target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Friday, March 27, 2015 4:41<br>
<b>To:</b> Haoweiguo; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Weiguo,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">We have already established that yo=
ur proposal is untenable because of its control plane load.
</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">What we are now discussing is that =
your proposal is based upon a misunderstanding of the algorithm in RFC 7432=
.&nbsp; You are assuming that PE1 will advertise
 an ES route and then wait for the configured interval before performing th=
e DF election while PE2 and PE3 will perform the DF election as soon as the=
y receive the ES route from PE1.&nbsp; This is not what RFC 7432 says.</spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Rather, what is says is that the ad=
vertisement of the ES route by PE1 and its receipt by PE2 and PE3 causes al=
l three PEs to start the configured
 interval timer&nbsp; - =933. When the timer expires, each PE builds an ord=
ered list of the IP addresses of all the PE nodes connected to the Ethernet=
 segment (including itself), in increasing numeric value.=94</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
font-size: 11pt;">From:</span></b><span style=3D"font-family: Calibri, sans=
-serif; font-size: 11pt;"> Haoweiguo [<a href=3D"mailto:haoweiguo@huawei.co=
m" target=3D"_blank">mailto:haoweiguo@huawei.com</a>]
<br>
<b>Sent:</b> Thursday, March 26, 2015 3:26 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Pls read=
 my detail replies to Satya. If you still can't catch it, pls read my draft=
 and EVPN base protocol, &nbsp;thanks</span><span style=3D"font-family: Tah=
oma, sans-serif; color: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">weiguo</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF478649">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> John E Drake [<a href=3D"mailto:jdrake@juniper.net" target=3D"_blan=
k">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Friday, March 27, 2015 1:28<br>
<b>To:</b> Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">I think Patrice is corr=
ect. &nbsp;Your proposal doesn't solve the problem and it does so at huge c=
ost.<br>
<br>
Sent from my iPhone</span></p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"COLOR: =
black"><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt; wrot=
e:</span></p>
</div>
<blockquote style=3D"MARGIN-TOP: 5pt; MARGIN-BOTTOM: 5pt">
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Weiguo,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">I=92m not sure I=92m fo=
llowing here.</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Don=92t you have the sa=
me issue with your handshaking mechanism?&nbsp;</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">If you don=92t know you=
r peer, how can you handshake?</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Regards,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"COLOR: =
black">Patrice</span></p>
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"BORDER-BOTTOM: windowtex=
t 1pt solid; BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDIN=
G-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: windowtext 1pt solid; BORDER-R=
IGHT: windowtext 1pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1026" bord=
er=3D"0" alt=3D"Image removed by sender." src=3D"cid:image001.jpg@01D067DF.=
BB3C17F0" width=3D"110" height=3D"73"></span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 0in;=
 PADDING-TOP: 0in" valign=3D"top" nowrap=3D"">
<p><strong><span style=3D"font-family: Arial, sans-serif; color: rgb(102, 1=
02, 102); font-size: 8.5pt;">Patrice Brissette</span></strong><span style=
=3D"font-family: Arial, sans-serif; color: rgb(102, 102, 102); font-size: 8=
.5pt;"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"COLO=
R: #666666; TEXT-DECORATION: none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"font-family: Arial, sans-serif;">&#43;1 613 2=
54 3336</span></strong></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 7.5pt; PADDING-LEFT: 15pt; WIDTH: 50%; PADDING=
-RIGHT: 0in; PADDING-TOP: 11.25pt" valign=3D"top" width=3D"50%">
<p><strong><span style=3D"font-family: Arial, sans-serif; color: rgb(102, 1=
02, 102); font-size: 8.5pt;">Cisco Systems Canada Co. / Les Systemes Cisco =
Canada CIE</span></strong><span style=3D"font-family: Arial, sans-serif; co=
lor: rgb(102, 102, 102); font-size: 8.5pt;"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"COLOR: #666666; TEXT-DECORATION: none">Cisco.com</span></a></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span></span>&nbsp;</p>
<table style=3D"WIDTH: 300pt" class=3D"MsoNormalTable" border=3D"0" cellspa=
cing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15pt=
; PADDING-TOP: 3.75pt">
<p class=3D"MsoNormal"><span style=3D"border: 1pt solid windowtext; padding=
: 0in; font-family: Arial, sans-serif; color: rgb(0, 153, 0); font-size: 7.=
5pt;"><img id=3D"_x0000_i1027" border=3D"0" alt=3D"Image removed by sender.=
" src=3D"cid:~WRD000.jpg" width=3D"100" height=3D"100"></span><span style=
=3D"font-family: Arial, sans-serif; color: rgb(0, 153, 0); font-size: 7.5pt=
;">Think
 before you print.</span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 4.5pt; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15=
pt; PADDING-TOP: 5.25pt">
<p><span style=3D"font-family: Arial, sans-serif; color: rgb(153, 153, 153)=
; font-size: 7.5pt;">This email may contain confidential and privileged mat=
erial for the sole use of the intended recipient. Any review, use, distribu=
tion or disclosure by others is strictly
 prohibited. If you are not the intended recipient (or authorized to receiv=
e for the recipient), please contact the sender by reply email and delete a=
ll copies of this message.</span></p>
<p><span style=3D"font-family: Arial, sans-serif; color: rgb(153, 153, 153)=
; font-size: 7.5pt;">Please
<a title=3D"Legal Information" href=3D"http://www.cisco.com/web/about/doing=
_business/legal/cri/index.html" target=3D"_blank">
<span style=3D"COLOR: #0e58a0">click here</span></a> for Company Registrati=
on Information.</span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><br clear=3D"all">
</span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
color: black; font-size: 11pt;">From:
</span></b><span style=3D"font-family: Calibri, sans-serif; color: black; f=
ont-size: 11pt;">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.com" targ=
et=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Thursday, March 26, 2015 at 10:24 AM<br>
<b>To: </b>Patrice Brissette &lt;<a href=3D"mailto:pbrisset@cisco.com" targ=
et=3D"_blank">pbrisset@cisco.com</a>&gt;, John E Drake &lt;<a href=3D"mailt=
o:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali Saj=
assi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@cis=
co.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<b>Subject: </b>RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<blockquote style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: #b5c4df 4.5pt=
 solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; MARGIN-=
LEFT: 3.75pt; BORDER-TOP: medium none; MARGIN-RIGHT: 0in; BORDER-RIGHT: med=
ium none; PADDING-TOP: 0in" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Hi Patri=
ce,</span><span style=3D"font-family: Tahoma, sans-serif; color: black; fon=
t-size: 10pt;"></span></p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Up to re=
ception timer traffic disruption in transient phase is one of the issues.</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Thanks,<=
/span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-si=
ze: 10pt;"></span></p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">weiguo</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF422254">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> Patrice Brissette (pbrisset) [<a href=3D"mailto:pbrisset@cisco.com"=
 target=3D"_blank">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;">Weiguo,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black;">You mention &quot;</span><span style=3D"COLOR: black">But if yo=
ur draft have not solved all issues=94,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Can you explain what Sa=
tya=92s draft is not solving?</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;">Regards,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"font-fa=
mily: Calibri, sans-serif; color: black; font-size: 10.5pt;">Patrice</span>=
</p>
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"BORDER-BOTTOM: windowtex=
t 1pt solid; BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDIN=
G-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: windowtext 1pt solid; BORDER-R=
IGHT: windowtext 1pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1029" bord=
er=3D"0" alt=3D"Image removed by sender." src=3D"cid:image001.jpg@01D067DF.=
BB3C17F0" width=3D"110" height=3D"73"></span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 0in;=
 PADDING-TOP: 0in" valign=3D"top" nowrap=3D"">
<p><strong><span style=3D"font-family: Arial, sans-serif; color: rgb(102, 1=
02, 102); font-size: 8.5pt;">Patrice Brissette</span></strong><span style=
=3D"font-family: Arial, sans-serif; color: rgb(102, 102, 102); font-size: 8=
.5pt;"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"COLO=
R: #666666; TEXT-DECORATION: none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"font-family: Arial, sans-serif;">&#43;1 613 2=
54 3336</span></strong></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 7.5pt; PADDING-LEFT: 15pt; WIDTH: 50%; PADDING=
-RIGHT: 0in; PADDING-TOP: 11.25pt" valign=3D"top" width=3D"50%">
<p><strong><span style=3D"font-family: Arial, sans-serif; color: rgb(102, 1=
02, 102); font-size: 8.5pt;">Cisco Systems Canada Co. / Les Systemes Cisco =
Canada CIE</span></strong><span style=3D"font-family: Arial, sans-serif; co=
lor: rgb(102, 102, 102); font-size: 8.5pt;"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"COLOR: #666666; TEXT-DECORATION: none">Cisco.com</span></a></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span></span>&nbsp;</p>
<table style=3D"WIDTH: 300pt" class=3D"MsoNormalTable" border=3D"0" cellspa=
cing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15pt=
; PADDING-TOP: 3.75pt">
<p class=3D"MsoNormal"><span style=3D"border: 1pt solid windowtext; padding=
: 0in; font-family: Arial, sans-serif; color: rgb(0, 153, 0); font-size: 7.=
5pt;"><img id=3D"_x0000_i1030" border=3D"0" alt=3D"Image removed by sender.=
" src=3D"cid:~WRD000.jpg" width=3D"100" height=3D"100"></span><span style=
=3D"font-family: Arial, sans-serif; color: rgb(0, 153, 0); font-size: 7.5pt=
;">Think
 before you print.</span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 4.5pt; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15=
pt; PADDING-TOP: 5.25pt">
<p><span style=3D"font-family: Arial, sans-serif; color: rgb(153, 153, 153)=
; font-size: 7.5pt;">This email may contain confidential and privileged mat=
erial for the sole use of the intended recipient. Any review, use, distribu=
tion or disclosure by others is strictly
 prohibited. If you are not the intended recipient (or authorized to receiv=
e for the recipient), please contact the sender by reply email and delete a=
ll copies of this message.</span></p>
<p><span style=3D"font-family: Arial, sans-serif; color: rgb(153, 153, 153)=
; font-size: 7.5pt;">Please
<a title=3D"Legal Information" href=3D"http://www.cisco.com/web/about/doing=
_business/legal/cri/index.html" target=3D"_blank">
<span style=3D"COLOR: #0e58a0">click here</span></a> for Company Registrati=
on Information.</span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"><br clear=3D"all">
</span></p>
</div>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
color: black; font-size: 11pt;">From:
</span></b><span style=3D"font-family: Calibri, sans-serif; color: black; f=
ont-size: 11pt;">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.com" targ=
et=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Wednesday, March 25, 2015 at 10:38 PM<br>
<b>To: </b>John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D=
"_blank">jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:saja=
ssi@cisco.com" target=3D"_blank">sajassi@cisco.com</a>&gt;, &quot;<a href=
=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<b>Subject: </b>Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: black; font-size: 10.5pt;"></span>&nbsp;</p>
</div>
<blockquote style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: #b5c4df 4.5pt=
 solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; MARGIN-=
LEFT: 3.75pt; BORDER-TOP: medium none; MARGIN-RIGHT: 0in; BORDER-RIGHT: med=
ium none; PADDING-TOP: 0in" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Hi John,=
</span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-s=
ize: 10pt;"></span></p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">&nbsp;Fi=
rstly i think EVPN community should reach consensus on the issues of curren=
t DF election mechanism. All these issues should be resolved in a single ne=
w DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</span><span style=3D"font-family: Tahoma, sans-serif; color=
: black; font-size: 10pt;"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size:=
 10pt;"></span>&nbsp;</p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">Thanks,<=
/span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-si=
ze: 10pt;"></span></p>
<p><span style=3D"font-family: Calibri, sans-serif; color: black;">weiguo</=
span><span style=3D"font-family: Tahoma, sans-serif; color: black; font-siz=
e: 10pt;"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF271558">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> John E Drake [<a href=3D"mailto:jdrake@juniper.net" target=3D"_blan=
k">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Weiguo,</span><span style=3D"COLOR:=
 black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Your proposal introduces a control =
plane processing load that is O(#EVIs * PEs) per DF election and given that=
 there can be 4K EVIs per ES, this looks
 like a *<b>substantial</b>* load.&nbsp; Furthermore,</span><span style=3D"=
COLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">you can=92t &nbsp;use the ES route =
to co-ordinate DF election because you would need to carry your new extende=
d community for each EVI and they would not
 all fit.&nbsp; You also can=92t use the Per EVI Ethernet AD route because =
that is processed by all PEs in the EVI.</span><span style=3D"COLOR: black"=
></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">I think that from a practical persp=
ective the new DF election proposed in Satya=92s draft is sufficiently stab=
le that it renders your draft moot, even
 if it could be made to work.</span><span style=3D"COLOR: black"></span></p=
>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">Yours Irrespectively,</span><span s=
tyle=3D"COLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"font-family: Calibri, sans-serif; col=
or: rgb(31, 73, 125); font-size: 11pt;">John</span><span style=3D"COLOR: bl=
ack"></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"font-family: Calibri, sans-serif; =
color: black; font-size: 11pt;">From:</span></b><span style=3D"font-family:=
 Calibri, sans-serif; color: black; font-size: 11pt;"> BESS [<a href=3D"mai=
lto:bess-bounces@ietf.org" target=3D"_blank">mailto:bess-bounces@ietf.org</=
a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Hi Ali,</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Thanks fo=
r your information. I scanned through this draft, it really introduces inte=
r-chassis message for DF election handshaking, the requirements in this dra=
ft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"COL=
OR: black"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Also for =
accuracy, i think handshaking state machine on each multi-homed PE is also =
needed. From solution perspective, in my draft, no inter-chassis message is=
 introduced, only one new extended
 community is introduced, i think the process is comparatively simple than =
your following draft.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Current E=
VPN DF election has some drawbacks, so there are three new drafts about DF =
election emerged. I think BESS WG can consider these three drafts in global=
 view,&nbsp; a single,comprehensive new
 DF election draft is hoped.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">thanks.</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">Thanks,</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"font-family: Tahoma, sans-serif; color: black;">weiguo</s=
pan><span style=3D"COLOR: black"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"font=
-family: Tahoma, sans-serif; color: black; font-size: 10pt;">From:</span></=
b><span style=3D"font-family: Tahoma, sans-serif; color: black; font-size: =
10pt;"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">be=
ss-bounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</span></div>
</div>
</div>
</div>
</div>
</blockquote>
</span>
</body>
</html>

--_000_D139EE3A4EFCFpbrissetciscocom_--

--_005_D139EE3A4EFCFpbrissetciscocom_
Content-Type: image/jpeg; name="~WRD000.jpg"
Content-Description: ~WRD000.jpg
Content-Disposition: attachment; filename="~WRD000.jpg"; size=823;
	creation-date="Thu, 26 Mar 2015 22:13:21 GMT";
	modification-date="Thu, 26 Mar 2015 22:13:21 GMT"
Content-ID: <~WRD000.jpg>
Content-Transfer-Encoding: base64
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--_005_D139EE3A4EFCFpbrissetciscocom_
Content-Type: image/jpeg; name="image001.jpg"
Content-Description: image001.jpg
Content-Disposition: attachment; filename="image001.jpg"; size=437;
	creation-date="Thu, 26 Mar 2015 22:13:21 GMT";
	modification-date="Thu, 26 Mar 2015 22:13:21 GMT"
Content-ID: <image001.jpg@01D067DF.BB3C17F0>
Content-Transfer-Encoding: base64

/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAoHBwkHBgoJCAkLCwoMDxkQDw4ODx4WFxIZJCAmJSMg
IyIoLTkwKCo2KyIjMkQyNjs9QEBAJjBGS0U+Sjk/QD3/wAALCABJAG4BAREA/8QAHwAAAQUBAQEB
AQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1Fh
ByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZ
WmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXG
x8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/9oACAEBAAA/APZqKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK//9k=

--_005_D139EE3A4EFCFpbrissetciscocom_--


From nobody Thu Mar 26 15:17:18 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AC9E61A0174 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 15:17:16 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.61
X-Spam-Level: 
X-Spam-Status: No, score=-3.61 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4K3pNcxLtY4I for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 15:17:12 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BE1791A015F for <bess@ietf.org>; Thu, 26 Mar 2015 15:17:08 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml406-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BUE37834; Thu, 26 Mar 2015 22:17:07 +0000 (GMT)
Received: from NKGEML401-HUB.china.huawei.com (10.98.56.32) by lhreml406-hub.china.huawei.com (10.201.5.243) with Microsoft SMTP Server (TLS) id 14.3.158.1; Thu, 26 Mar 2015 22:17:05 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml401-hub.china.huawei.com ([10.98.56.32]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 06:16:58 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: John E Drake <jdrake@juniper.net>, "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy8=
Date: Thu, 26 Mar 2015 22:16:58 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com>
In-Reply-To: <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.156.0]
Content-Type: multipart/related; boundary="_005_DD5FC8DE455C3348B94340C0AB5517334F857541nkgeml501mbschi_"; type="multipart/alternative"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/zuRjuvPeX_WhT0sK1HafQ7fvnqw>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 22:17:17 -0000

--_005_DD5FC8DE455C3348B94340C0AB5517334F857541nkgeml501mbschi_
Content-Type: multipart/alternative;
	boundary="_000_DD5FC8DE455C3348B94340C0AB5517334F857541nkgeml501mbschi_"

--_000_DD5FC8DE455C3348B94340C0AB5517334F857541nkgeml501mbschi_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Pls see below.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 6:00
To: Haoweiguo; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

To recap,

We have established that your proposal is untenable because of its control =
plane load.

We have established that your proposal is based upon a flawed understanding=
 of the DF election in RFC 7432.
[weiguo]:  In ethernet world, traffic loop is serious than short timer traf=
fic disruption. If you want to implement  transiet traffic loop process, i =
will modify my draft to solve your issue.
If i am the developer, i will prefer short timer traffic disruption based o=
n current EVPN protocol.

What you are now arguing is that your draft prevents two or more PEs from b=
eing DF simultaneously.  This is clearly nonsense.
[weiguo]: I will modify the draft problem statements, and use the same hand=
shaking solution to solve it.

Furthermore, we have established that having two or more DFs for what even =
you admit is a brief transient leads to duplicate traffic, which is accepta=
ble, but not loops, your assertion to the contrary.
[weiguo]: It is transient loop and traffic duplication issue.
Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 5:37 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


John,

As your understanding of the EVPN draft, the DF election mechanism has more=
 serious side effect, it will have short time traffic loop,i.e., dual DF PE=
s will exist for a short time. I think dual DF PEs is absolutely not tolera=
ted, because native ethernet header has no TTL, up to several hundred ms tr=
affic loop normally not tolerated in commertial networks.



As your understanding, the PEs should do as following:

1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have rarely=
 deployment.

Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins ESI an=
d trigger DF re-election. When reception timer expires:

PE1 upgrades to DF PE.

After reception timer+ ES route transmission timer:

PE2 downloads to non-DF PE.

So in timer clock sync case, dual DF PEs will exist at least transmission t=
imer.



If NTP is used for timer sync, because it has bad accuracy, dual DF PEs wil=
l exist more longer timer.



So as your understanding for DF election, the drawback is more clear.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 4:41
To: Haoweiguo; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

We have already established that your proposal is untenable because of its =
control plane load.

What we are now discussing is that your proposal is based upon a misunderst=
anding of the algorithm in RFC 7432.  You are assuming that PE1 will advert=
ise an ES route and then wait for the configured interval before performing=
 the DF election while PE2 and PE3 will perform the DF election as soon as =
they receive the ES route from PE1.  This is not what RFC 7432 says.

Rather, what is says is that the advertisement of the ES route by PE1 and i=
ts receipt by PE2 and PE3 causes all three PEs to start the configured inte=
rval timer  - =933. When the timer expires, each PE builds an ordered list =
of the IP addresses of all the PE nodes connected to the Ethernet segment (=
including itself), in increasing numeric value.=94

Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 3:26 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Pls read my detail replies to Satya. If you still can't catch it, pls read =
my draft and EVPN base protocol,  thanks

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 1:28
To: Patrice Brissette (pbrisset)
Cc: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:
Weiguo,

I=92m not sure I=92m following here.
Don=92t you have the same issue with your handshaking mechanism?
If you don=92t know your peer, how can you handshake?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

You mention "But if your draft have not solved all issues=94,
Can you explain what Satya=92s draft is not solving?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can=92t  use the ES route to co-ordinate DF election because you would =
need to carry your new extended community for each EVI and they would not a=
ll fit.  You also can=92t use the Per EVI Ethernet AD route because that is=
 processed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya=92s draft is sufficiently stable that it renders your draft moot, even=
 if it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_DD5FC8DE455C3348B94340C0AB5517334F857541nkgeml501mbschi_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html dir=3D"ltr">
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
<style>@font-face {
	font-family: Cambria Math;
}
@font-face {
	font-family: Calibri;
}
@font-face {
	font-family: Tahoma;
}
@font-face {
	font-family: Consolas;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
LI.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
DIV.MsoNormal {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 12pt
}
A:link {
	COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
	COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
	COLOR: purple; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
	COLOR: purple; TEXT-DECORATION: underline
}
PRE {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: Consolas; FONT-SIZE: 10pt
}
SPAN.HTMLPreformattedChar {
	FONT-FAMILY: Consolas
}
P.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
LI.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
DIV.msochpdefault {
	MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman",serif; FONT-SIZE: 10pt
}
SPAN.emailstyle18 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
SPAN.emailstyle21 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
SPAN.htmlpreformattedchar0 {
	FONT-FAMILY: Consolas
}
SPAN.EmailStyle25 {
	FONT-FAMILY: "Calibri",sans-serif; COLOR: #1f497d
}
.msochpdefault {
	FONT-SIZE: 10pt
}
</style><style id=3D"owaParaStyle">P {
	MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
</style>
</head>
<body lang=3D"EN-US" link=3D"blue" vlink=3D"purple" fPStyle=3D"1" ocsi=3D"0=
">
<div style=3D"direction: ltr;font-family: Tahoma;color: #000000;font-size: =
10pt;">
<p><font size=3D"3" face=3D"Calibri">Pls see below.</font></p>
<p><font size=3D"3" face=3D"Calibri"></font>&nbsp;</p>
<p><font size=3D"3" face=3D"Calibri">Thanks,</font></p>
<p><font size=3D"3" face=3D"Calibri">weiguo</font></p>
<div style=3D"FONT-FAMILY: Times New Roman; COLOR: #000000; FONT-SIZE: 16px=
">
<hr tabindex=3D"-1">
<div style=3D"DIRECTION: ltr" id=3D"divRpF887072"><font color=3D"#000000" s=
ize=3D"2" face=3D"Tahoma"><b>From:</b> John E Drake [jdrake@juniper.net]<br=
>
<b>Sent:</b> Friday, March 27, 2015 6:00<br>
<b>To:</b> Haoweiguo; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<br>
</font><br>
</div>
<div></div>
<div>
<div class=3D"WordSection1">
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">To recap,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">We have established that your proposal is un=
tenable because of its control plane load.
</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">We have established that your proposal is ba=
sed upon a flawed understanding of the DF election in RFC 7432.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">[weiguo]:&nbsp; In ethernet world, traffic l=
oop is serious than short timer traffic disruption. If you want to implemen=
t &nbsp;transiet traffic loop process, i will
 modify my draft to solve your issue.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">If i am the developer, i will prefer short t=
imer traffic disruption based on current EVPN protocol.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">What you are now arguing is that your draft =
prevents two or more PEs from being DF simultaneously.&nbsp; This is clearl=
y nonsense.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">[weiguo]: I will modify the draft problem st=
atements, and use the same handshaking solution to solve it.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Furthermore, we have established that having=
 two or more DFs for what even you admit is a brief transient leads to dupl=
icate traffic, which is acceptable,
 but not loops, your assertion to the contrary.</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">[weiguo]: It is transient loop and traffic d=
uplication issue.</span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Yours Irrespectively,</span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">John</span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif;=
 FONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY: 'Calibri',san=
s-serif; FONT-SIZE: 11pt"> Haoweiguo [mailto:haoweiguo@huawei.com]
<br>
<b>Sent:</b> Thursday, March 26, 2015 5:37 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span></p>
</div>
</div>
<p class=3D"MsoNormal">&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">John,</s=
pan><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZ=
E: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">As your =
understanding of the EVPN draft, the DF election mechanism has more serious=
 side effect, it will have short time traffic loop,i.e., dual DF PEs will e=
xist for a short time. I think dual
 DF PEs is absolutely not tolerated, because native ethernet header has no =
TTL, up to several hundred ms traffic loop normally not tolerated in commer=
tial networks.</span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR=
: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE=
: 10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">As your =
understanding, the PEs should do as following:</span><span style=3D"FONT-FA=
MILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">1. Accur=
ate timer sync. NTP accuracy is bad, 1588v2 is good but have rarely deploym=
ent.</span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; F=
ONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Assuming=
 PE1,PE2 and PE3 have consistent timer clock, when PE3 joins ESI and trigge=
r DF re-election. When reception timer expires:</span><span style=3D"FONT-F=
AMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">PE1 upgr=
ades to DF PE.</span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR=
: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">After re=
ception timer&#43; ES route transmission timer:</span><span style=3D"FONT-F=
AMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">PE2 down=
loads to non-DF PE.</span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; =
COLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">So in ti=
mer clock sync case, dual DF PEs will exist at least transmission timer.</s=
pan><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZ=
E: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE=
: 10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">If NTP i=
s used for timer sync, because it has bad accuracy, dual DF PEs will exist =
more longer timer.</span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; C=
OLOR: black; FONT-SIZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE=
: 10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">So as yo=
ur understanding for DF election, the drawback is more clear.</span><span s=
tyle=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt"></=
span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE=
: 10pt"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Thanks,<=
/span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-S=
IZE: 10pt"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">weiguo</=
span><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SI=
ZE: 10pt"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF234411">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></=
b><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE:=
 10pt"> John E Drake [jdrake@juniper.net]<br>
<b>Sent:</b> Friday, March 27, 2015 4:41<br>
<b>To:</b> Haoweiguo; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Weiguo,</span><span style=3D"COLOR: black"><=
/span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">We have already established that your propos=
al is untenable because of its control plane load.
</span><span style=3D"COLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">What we are now discussing is that your prop=
osal is based upon a misunderstanding of the algorithm in RFC 7432.&nbsp; Y=
ou are assuming that PE1 will advertise an
 ES route and then wait for the configured interval before performing the D=
F election while PE2 and PE3 will perform the DF election as soon as they r=
eceive the ES route from PE1.&nbsp; This is not what RFC 7432 says.</span><=
span style=3D"COLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Rather, what is says is that the advertiseme=
nt of the ES route by PE1 and its receipt by PE2 and PE3 causes all three P=
Es to start the configured interval
 timer&nbsp; - =933. When the timer expires, each PE builds an ordered list=
 of the IP addresses of all the PE nodes connected to the Ethernet segment =
(including itself), in increasing numeric value.=94</span><span style=3D"CO=
LOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Yours Irrespectively,</span><span style=3D"C=
OLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">John</span><span style=3D"COLOR: black"></sp=
an></p>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif;=
 COLOR: black; FONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY:=
 'Calibri',sans-serif; COLOR: black; FONT-SIZE: 11pt"> Haoweiguo [<a href=
=3D"mailto:haoweiguo@huawei.com" target=3D"_blank">mailto:haoweiguo@huawei.=
com</a>]
<br>
<b>Sent:</b> Thursday, March 26, 2015 3:26 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Pls read=
 my detail replies to Satya. If you still can't catch it, pls read my draft=
 and EVPN base protocol, &nbsp;thanks</span><span style=3D"COLOR: black"></=
span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">weiguo</=
span><span style=3D"COLOR: black"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF478649">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></=
b><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE:=
 10pt"> John E Drake [jdrake@juniper.net]<br>
<b>Sent:</b> Friday, March 27, 2015 1:28<br>
<b>To:</b> Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">I think Patrice is corr=
ect. &nbsp;Your proposal doesn't solve the problem and it does so at huge c=
ost.<br>
<br>
Sent from my iPhone</span></p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"COLOR: =
black"><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt; wrot=
e:</span></p>
</div>
<blockquote style=3D"MARGIN-TOP: 5pt; MARGIN-BOTTOM: 5pt">
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Weiguo,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">I=92m not sure I=92m fo=
llowing here.</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Don=92t you have the sa=
me issue with your handshaking mechanism?&nbsp;</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">If you don=92t know you=
r peer, how can you handshake?</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Regards,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"COLOR: =
black">Patrice</span></p>
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"BORDER-BOTTOM: windowtex=
t 1pt solid; BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDIN=
G-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: windowtext 1pt solid; BORDER-R=
IGHT: windowtext 1pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1027" bord=
er=3D"0" alt=3D"Image removed by sender." src=3D"cid:image001.jpg@01D067EE.=
7BA7B450" width=3D"110" height=3D"73"></span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 0in;=
 PADDING-TOP: 0in" valign=3D"top" nowrap=3D"">
<p><strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #666666; =
FONT-SIZE: 8.5pt">Patrice Brissette</span></strong><span style=3D"FONT-FAMI=
LY: 'Arial',sans-serif; COLOR: #666666; FONT-SIZE: 8.5pt"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"COLO=
R: #666666; TEXT-DECORATION: none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif">&#43;1 613 2=
54 3336</span></strong></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 7.5pt; PADDING-LEFT: 15pt; WIDTH: 50%; PADDING=
-RIGHT: 0in; PADDING-TOP: 11.25pt" valign=3D"top" width=3D"50%">
<p><strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #666666; =
FONT-SIZE: 8.5pt">Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE<=
/span></strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #6666=
66; FONT-SIZE: 8.5pt"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"COLOR: #666666; TEXT-DECORATION: none">Cisco.com</span></a></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal">&nbsp;</p>
<table style=3D"WIDTH: 300pt" class=3D"MsoNormalTable" border=3D"0" cellspa=
cing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15pt=
; PADDING-TOP: 3.75pt">
<p class=3D"MsoNormal"><span style=3D"BORDER-BOTTOM: windowtext 1pt solid; =
BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; =
PADDING-RIGHT: 0in; FONT-FAMILY: 'Arial',sans-serif; COLOR: #009900; FONT-S=
IZE: 7.5pt; BORDER-TOP: windowtext 1pt solid; BORDER-RIGHT: windowtext 1pt =
solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1028" border=3D"0" alt=3D"Image=
 removed by sender." src=3D"cid:image002.jpg@01D067EE.7BA7B450" width=3D"10=
0" height=3D"100"></span><span style=3D"FONT-FAMILY: 'Arial',sans-serif; CO=
LOR: #009900; FONT-SIZE: 7.5pt">Think
 before you print.</span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 4.5pt; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15=
pt; PADDING-TOP: 5.25pt">
<p><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #999999; FONT-SIZ=
E: 7.5pt">This email may contain confidential and privileged material for t=
he sole use of the intended recipient. Any review, use, distribution or dis=
closure by others is strictly prohibited.
 If you are not the intended recipient (or authorized to receive for the re=
cipient), please contact the sender by reply email and delete all copies of=
 this message.</span></p>
<p><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #999999; FONT-SIZ=
E: 7.5pt">Please
<a title=3D"Legal Information" href=3D"http://www.cisco.com/web/about/doing=
_business/legal/cri/index.html" target=3D"_blank">
<span style=3D"COLOR: #0e58a0">click here</span></a> for Company Registrati=
on Information.</span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><br clear=3D"all">
</span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif;=
 COLOR: black; FONT-SIZE: 11pt">From:
</span></b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black; =
FONT-SIZE: 11pt">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.com" targ=
et=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Thursday, March 26, 2015 at 10:24 AM<br>
<b>To: </b>Patrice Brissette &lt;<a href=3D"mailto:pbrisset@cisco.com" targ=
et=3D"_blank">pbrisset@cisco.com</a>&gt;, John E Drake &lt;<a href=3D"mailt=
o:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali Saj=
assi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@cis=
co.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<b>Subject: </b>RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<blockquote style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: #b5c4df 4.5pt=
 solid; PADDING-BOTTOM: 0in; MARGIN: 5pt 0in 5pt 3.75pt; PADDING-LEFT: 4pt;=
 PADDING-RIGHT: 0in; BORDER-TOP: medium none; BORDER-RIGHT: medium none; PA=
DDING-TOP: 0in" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Hi Patri=
ce,</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Up to re=
ception timer traffic disruption in transient phase is one of the issues.</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Thanks,<=
/span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">weiguo</=
span><span style=3D"COLOR: black"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF422254">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></=
b><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE:=
 10pt"> Patrice Brissette (pbrisset) [<a href=3D"mailto:pbrisset@cisco.com"=
 target=3D"_blank">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt">Weiguo,</span><span style=3D"COLOR: black"><=
/span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black">You mention &quot;</span><span style=3D"COLOR: black">But if yo=
ur draft have not solved all issues=94,</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">Can you explain what Sa=
tya=92s draft is not solving?</span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt">Regards,</span><span style=3D"COLOR: black">=
</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><span style=3D"FONT-FA=
MILY: 'Calibri',sans-serif; COLOR: black; FONT-SIZE: 10.5pt">Patrice</span>=
<span style=3D"COLOR: black"></span></p>
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
<table style=3D"WIDTH: 407.25pt" class=3D"MsoNormalTable" border=3D"0" cell=
spacing=3D"0" cellpadding=3D"0" width=3D"543">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"BORDER-BOTTOM: windowtex=
t 1pt solid; BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDIN=
G-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: windowtext 1pt solid; BORDER-R=
IGHT: windowtext 1pt solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1030" bord=
er=3D"0" alt=3D"Image removed by sender." src=3D"cid:image001.jpg@01D067EE.=
7BA7B450" width=3D"110" height=3D"73"></span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 0in;=
 PADDING-TOP: 0in" valign=3D"top" nowrap=3D"">
<p><strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #666666; =
FONT-SIZE: 8.5pt">Patrice Brissette</span></strong><span style=3D"FONT-FAMI=
LY: 'Arial',sans-serif; COLOR: #666666; FONT-SIZE: 8.5pt"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"COLO=
R: #666666; TEXT-DECORATION: none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif">&#43;1 613 2=
54 3336</span></strong></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 7.5pt; PADDING-LEFT: 15pt; WIDTH: 50%; PADDING=
-RIGHT: 0in; PADDING-TOP: 11.25pt" valign=3D"top" width=3D"50%">
<p><strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #666666; =
FONT-SIZE: 8.5pt">Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE<=
/span></strong><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #6666=
66; FONT-SIZE: 8.5pt"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"COLOR: #666666; TEXT-DECORATION: none">Cisco.com</span></a></span></p>
</td>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in">
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; PA=
DDING-TOP: 0in" colspan=3D"3">
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal">&nbsp;</p>
<table style=3D"WIDTH: 300pt" class=3D"MsoNormalTable" border=3D"0" cellspa=
cing=3D"0" cellpadding=3D"0" width=3D"400">
<tbody>
<tr>
<td style=3D"PADDING-BOTTOM: 0in; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15pt=
; PADDING-TOP: 3.75pt">
<p class=3D"MsoNormal"><span style=3D"BORDER-BOTTOM: windowtext 1pt solid; =
BORDER-LEFT: windowtext 1pt solid; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; =
PADDING-RIGHT: 0in; FONT-FAMILY: 'Arial',sans-serif; COLOR: #009900; FONT-S=
IZE: 7.5pt; BORDER-TOP: windowtext 1pt solid; BORDER-RIGHT: windowtext 1pt =
solid; PADDING-TOP: 0in"><img id=3D"_x0000_i1031" border=3D"0" alt=3D"Image=
 removed by sender." src=3D"cid:image002.jpg@01D067EE.7BA7B450" width=3D"10=
0" height=3D"100"></span><span style=3D"FONT-FAMILY: 'Arial',sans-serif; CO=
LOR: #009900; FONT-SIZE: 7.5pt">Think
 before you print.</span></p>
</td>
</tr>
<tr>
<td style=3D"PADDING-BOTTOM: 4.5pt; PADDING-LEFT: 0.25in; PADDING-RIGHT: 15=
pt; PADDING-TOP: 5.25pt">
<p><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #999999; FONT-SIZ=
E: 7.5pt">This email may contain confidential and privileged material for t=
he sole use of the intended recipient. Any review, use, distribution or dis=
closure by others is strictly prohibited.
 If you are not the intended recipient (or authorized to receive for the re=
cipient), please contact the sender by reply email and delete all copies of=
 this message.</span></p>
<p><span style=3D"FONT-FAMILY: 'Arial',sans-serif; COLOR: #999999; FONT-SIZ=
E: 7.5pt">Please
<a title=3D"Legal Information" href=3D"http://www.cisco.com/web/about/doing=
_business/legal/cri/index.html" target=3D"_blank">
<span style=3D"COLOR: #0e58a0">click here</span></a> for Company Registrati=
on Information.</span></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: black; FONT-SIZE: 10.5pt"><br clear=3D"all">
</span><span style=3D"COLOR: black"></span></p>
</div>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif;=
 COLOR: black; FONT-SIZE: 11pt">From:
</span></b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black; =
FONT-SIZE: 11pt">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.com" targ=
et=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Wednesday, March 25, 2015 at 10:38 PM<br>
<b>To: </b>John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D=
"_blank">jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:saja=
ssi@cisco.com" target=3D"_blank">sajassi@cisco.com</a>&gt;, &quot;<a href=
=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<b>Subject: </b>Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<blockquote style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: #b5c4df 4.5pt=
 solid; PADDING-BOTTOM: 0in; MARGIN: 5pt 0in 5pt 3.75pt; PADDING-LEFT: 4pt;=
 PADDING-RIGHT: 0in; BORDER-TOP: medium none; BORDER-RIGHT: medium none; PA=
DDING-TOP: 0in" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Hi John,=
</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">&nbsp;Fi=
rstly i think EVPN community should reach consensus on the issues of curren=
t DF election mechanism. All these issues should be resolved in a single ne=
w DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">Thanks,<=
/span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; COLOR: black">weiguo</=
span><span style=3D"COLOR: black"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF271558">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></=
b><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE:=
 10pt"> John E Drake [<a href=3D"mailto:jdrake@juniper.net" target=3D"_blan=
k">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Weiguo,</span><span style=3D"COLOR: black"><=
/span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Your proposal introduces a control plane pro=
cessing load that is O(#EVIs * PEs) per DF election and given that there ca=
n be 4K EVIs per ES, this looks like
 a *<b>substantial</b>* load.&nbsp; Furthermore,</span><span style=3D"COLOR=
: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">you can=92t &nbsp;use the ES route to co-ord=
inate DF election because you would need to carry your new extended communi=
ty for each EVI and they would not all fit.&nbsp;
 You also can=92t use the Per EVI Ethernet AD route because that is process=
ed by all PEs in the EVI.</span><span style=3D"COLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">I think that from a practical perspective th=
e new DF election proposed in Satya=92s draft is sufficiently stable that i=
t renders your draft moot, even if it
 could be made to work.</span><span style=3D"COLOR: black"></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">Yours Irrespectively,</span><span style=3D"C=
OLOR: black"></span></p>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<p class=3D"MsoNormal"><span style=3D"FONT-FAMILY: 'Calibri',sans-serif; CO=
LOR: #1f497d; FONT-SIZE: 11pt">John</span><span style=3D"COLOR: black"></sp=
an></p>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: blue 1.5pt solid; PA=
DDING-BOTTOM: 0in; PADDING-LEFT: 4pt; PADDING-RIGHT: 0in; BORDER-TOP: mediu=
m none; BORDER-RIGHT: medium none; PADDING-TOP: 0in">
<div>
<div style=3D"BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING=
-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #e1e1e1 1p=
t solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class=3D"MsoNormal"><b><span style=3D"FONT-FAMILY: 'Calibri',sans-serif;=
 COLOR: black; FONT-SIZE: 11pt">From:</span></b><span style=3D"FONT-FAMILY:=
 'Calibri',sans-serif; COLOR: black; FONT-SIZE: 11pt"> BESS [<a href=3D"mai=
lto:bess-bounces@ietf.org" target=3D"_blank">mailto:bess-bounces@ietf.org</=
a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"COLOR: black"></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
<div>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Hi Ali,</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Thanks fo=
r your information. I scanned through this draft, it really introduces inte=
r-chassis message for DF election handshaking, the requirements in this dra=
ft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"COL=
OR: black"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Also for =
accuracy, i think handshaking state machine on each multi-homed PE is also =
needed. From solution perspective, in my draft, no inter-chassis message is=
 introduced, only one new extended
 community is introduced, i think the process is comparatively simple than =
your following draft.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Current E=
VPN DF election has some drawbacks, so there are three new drafts about DF =
election emerged. I think BESS WG can consider these three drafts in global=
 view,&nbsp; a single,comprehensive new
 DF election draft is hoped.</span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">thanks.</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"COLOR: black"></span>&nbsp;</p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">Thanks,</=
span><span style=3D"COLOR: black"></span></p>
<p><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black">weiguo</s=
pan><span style=3D"COLOR: black"></span></p>
<div>
<div style=3D"TEXT-ALIGN: center" class=3D"MsoNormal" align=3D"center"><spa=
n style=3D"COLOR: black">
<hr align=3D"center" size=3D"2" width=3D"100%">
</span></div>
<div id=3D"divRpF484031">
<p style=3D"MARGIN-BOTTOM: 12pt" class=3D"MsoNormal"><b><span style=3D"FONT=
-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE: 10pt">From:</span></=
b><span style=3D"FONT-FAMILY: 'Tahoma',sans-serif; COLOR: black; FONT-SIZE:=
 10pt"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">be=
ss-bounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"COLOR: black"></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">FYI- First published Ju=
ly 4, 2011</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"><a href=3D"https://data=
tracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_bla=
nk">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route=
</a>/</span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black"></span>&nbsp;</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"COLOR: black">-Ali</span></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_DD5FC8DE455C3348B94340C0AB5517334F857541nkgeml501mbschi_--

--_005_DD5FC8DE455C3348B94340C0AB5517334F857541nkgeml501mbschi_
Content-Type: image/jpeg; name="image001.jpg"
Content-Description: image001.jpg
Content-Disposition: inline; filename="image001.jpg"; size=437;
	creation-date="Thu, 26 Mar 2015 22:00:03 GMT";
	modification-date="Thu, 26 Mar 2015 22:00:03 GMT"
Content-ID: <image001.jpg@01D067EE.7BA7B450>
Content-Transfer-Encoding: base64

/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAoHBwkHBgoJCAkLCwoMDxkQDw4ODx4WFxIZJCAmJSMg
IyIoLTkwKCo2KyIjMkQyNjs9QEBAJjBGS0U+Sjk/QD3/wAALCABJAG4BAREA/8QAHwAAAQUBAQEB
AQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1Fh
ByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZ
WmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXG
x8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/9oACAEBAAA/APZqKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK//9k=

--_005_DD5FC8DE455C3348B94340C0AB5517334F857541nkgeml501mbschi_
Content-Type: image/jpeg; name="image002.jpg"
Content-Description: image002.jpg
Content-Disposition: inline; filename="image002.jpg"; size=823;
	creation-date="Thu, 26 Mar 2015 22:00:03 GMT";
	modification-date="Thu, 26 Mar 2015 22:00:03 GMT"
Content-ID: <image002.jpg@01D067EE.7BA7B450>
Content-Transfer-Encoding: base64
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--_005_DD5FC8DE455C3348B94340C0AB5517334F857541nkgeml501mbschi_--


From nobody Thu Mar 26 18:05:55 2015
Return-Path: <jdrake@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 67A441A887B for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 18:05:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.301
X-Spam-Level: 
X-Spam-Status: No, score=-1.301 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jYOrPvySqgoH for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 18:05:46 -0700 (PDT)
Received: from na01-bl2-obe.outbound.protection.outlook.com (mail-bl2on0135.outbound.protection.outlook.com [65.55.169.135]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 744471A0163 for <bess@ietf.org>; Thu, 26 Mar 2015 18:05:46 -0700 (PDT)
Received: from BLUPR05MB563.namprd05.prod.outlook.com (10.141.202.144) by BLUPR05MB038.namprd05.prod.outlook.com (10.255.210.147) with Microsoft SMTP Server (TLS) id 15.1.112.16; Fri, 27 Mar 2015 01:05:45 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) by BLUPR05MB563.namprd05.prod.outlook.com (10.141.202.144) with Microsoft SMTP Server (TLS) id 15.1.118.21; Fri, 27 Mar 2015 01:05:43 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) by BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) with mapi id 15.01.0118.022; Fri, 27 Mar 2015 01:05:43 +0000
From: John E Drake <jdrake@juniper.net>
To: Haoweiguo <haoweiguo@huawei.com>, "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8P/o146fJTTukSNC2GolBinQZ0u4lsAgAACyoCAAB/gjYAAILMAgAAMEtCAABidAIAAA48AgAAHqACAAC468A==
Date: Fri, 27 Mar 2015 01:05:42 +0000
Message-ID: <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator: 
x-originating-ip: [66.129.239.10]
authentication-results: huawei.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB563; UriScan:; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB038; 
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(69234005)(377454003)(164054003)(36304003)(24454002)(50854003)(15975445007)(102836002)(19625215002)(40100003)(99286002)(77096005)(19609705001)(19617315012)(19627595001)(46102003)(74316001)(76176999)(19580405001)(19580395003)(561944003)(122556002)(54356999)(2656002)(33656002)(87936001)(50986999)(106116001)(17760045003)(66066001)(18206015028)(86362001)(19300405004)(2900100001)(99936001)(76576001)(16236675004)(2950100001)(62966003)(77156002)(93886004)(92566002)(579004); DIR:OUT; SFP:1102; SCL:1; SRVR:BLUPR05MB563; H:BLUPR05MB562.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BLUPR05MB563B349CDC361D6CD6E95B1C7090@BLUPR05MB563.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BLUPR05MB563; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB563; 
x-forefront-prvs: 0528942FD8
Content-Type: multipart/related; boundary="_005_BLUPR05MB562A3B02E72B599D65C1F42C7090BLUPR05MB562namprd_"; type="multipart/alternative"
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-originalarrivaltime: 27 Mar 2015 01:05:42.8161 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BLUPR05MB563
X-OriginatorOrg: juniper.net
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/LZNf63MRti6v5PlRXPUWuhvfWkI>
Cc: "Ali Sajassi \(sajassi\)" <sajassi@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 01:05:54 -0000

--_005_BLUPR05MB562A3B02E72B599D65C1F42C7090BLUPR05MB562namprd_
Content-Type: multipart/alternative;
	boundary="_000_BLUPR05MB562A3B02E72B599D65C1F42C7090BLUPR05MB562namprd_"

--_000_BLUPR05MB562A3B02E72B599D65C1F42C7090BLUPR05MB562namprd_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Weiguo,

I guess I wasn't clear.  I think you draft, for the reasons I have detailed=
, is a non-solution to a non-problem with tremendous control plane cost.

Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 6:17 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Pls see below.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 6:00
To: Haoweiguo; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
To recap,

We have established that your proposal is untenable because of its control =
plane load.

We have established that your proposal is based upon a flawed understanding=
 of the DF election in RFC 7432.
[weiguo]:  In ethernet world, traffic loop is serious than short timer traf=
fic disruption. If you want to implement  transiet traffic loop process, i =
will modify my draft to solve your issue.
If i am the developer, i will prefer short timer traffic disruption based o=
n current EVPN protocol.

What you are now arguing is that your draft prevents two or more PEs from b=
eing DF simultaneously.  This is clearly nonsense.
[weiguo]: I will modify the draft problem statements, and use the same hand=
shaking solution to solve it.

Furthermore, we have established that having two or more DFs for what even =
you admit is a brief transient leads to duplicate traffic, which is accepta=
ble, but not loops, your assertion to the contrary.
[weiguo]: It is transient loop and traffic duplication issue.
Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 5:37 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


John,

As your understanding of the EVPN draft, the DF election mechanism has more=
 serious side effect, it will have short time traffic loop,i.e., dual DF PE=
s will exist for a short time. I think dual DF PEs is absolutely not tolera=
ted, because native ethernet header has no TTL, up to several hundred ms tr=
affic loop normally not tolerated in commertial networks.



As your understanding, the PEs should do as following:

1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have rarely=
 deployment.

Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins ESI an=
d trigger DF re-election. When reception timer expires:

PE1 upgrades to DF PE.

After reception timer+ ES route transmission timer:

PE2 downloads to non-DF PE.

So in timer clock sync case, dual DF PEs will exist at least transmission t=
imer.



If NTP is used for timer sync, because it has bad accuracy, dual DF PEs wil=
l exist more longer timer.



So as your understanding for DF election, the drawback is more clear.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 4:41
To: Haoweiguo; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

We have already established that your proposal is untenable because of its =
control plane load.

What we are now discussing is that your proposal is based upon a misunderst=
anding of the algorithm in RFC 7432.  You are assuming that PE1 will advert=
ise an ES route and then wait for the configured interval before performing=
 the DF election while PE2 and PE3 will perform the DF election as soon as =
they receive the ES route from PE1.  This is not what RFC 7432 says.

Rather, what is says is that the advertisement of the ES route by PE1 and i=
ts receipt by PE2 and PE3 causes all three PEs to start the configured inte=
rval timer  - "3. When the timer expires, each PE builds an ordered list of=
 the IP addresses of all the PE nodes connected to the Ethernet segment (in=
cluding itself), in increasing numeric value."

Yours Irrespectively,

John

From: Haoweiguo [mailto:haoweiguo@huawei.com]
Sent: Thursday, March 26, 2015 3:26 PM
To: John E Drake; Patrice Brissette (pbrisset)
Cc: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Pls read my detail replies to Satya. If you still can't catch it, pls read =
my draft and EVPN base protocol,  thanks

weiguo

________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 1:28
To: Patrice Brissette (pbrisset)
Cc: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
I think Patrice is correct.  Your proposal doesn't solve the problem and it=
 does so at huge cost.

Sent from my iPhone

On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) <pbrisset@cisco.=
com<mailto:pbrisset@cisco.com>> wrote:
Weiguo,

I'm not sure I'm following here.
Don't you have the same issue with your handshaking mechanism?
If you don't know your peer, how can you handshake?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Thursday, March 26, 2015 at 10:24 AM
To: Patrice Brissette <pbrisset@cisco.com<mailto:pbrisset@cisco.com>>, John=
 E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajassi <saja=
ssi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bess@ietf.o=
rg>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Patrice,

Up to reception timer traffic disruption in transient phase is one of the i=
ssues.



Thanks,

weiguo

________________________________
From: Patrice Brissette (pbrisset) [pbrisset@cisco.com<mailto:pbrisset@cisc=
o.com>]
Sent: Thursday, March 26, 2015 20:54
To: Haoweiguo; John E Drake; Ali Sajassi (sajassi); bess@ietf.org<mailto:be=
ss@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

You mention "But if your draft have not solved all issues",
Can you explain what Satya's draft is not solving?


Regards,

Patrice
   [Image removed by sender.]


Patrice Brissette
TECHNICAL LEADER.ENGINEERING

pbrisset@cisco.com<mailto:pbrisset@cisco.com>
Phone: +1 613 254 3336


Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE
Canada
Cisco.com<http://www.cisco.com/global/CA/>



[Image removed by sender.]Think before you print.


This email may contain confidential and privileged material for the sole us=
e of the intended recipient. Any review, use, distribution or disclosure by=
 others is strictly prohibited. If you are not the intended recipient (or a=
uthorized to receive for the recipient), please contact the sender by reply=
 email and delete all copies of this message.

Please click here<http://www.cisco.com/web/about/doing_business/legal/cri/i=
ndex.html> for Company Registration Information.




From: Haoweiguo <haoweiguo@huawei.com<mailto:haoweiguo@huawei.com>>
Date: Wednesday, March 25, 2015 at 10:38 PM
To: John E Drake <jdrake@juniper.net<mailto:jdrake@juniper.net>>, Ali Sajas=
si <sajassi@cisco.com<mailto:sajassi@cisco.com>>, "bess@ietf.org<mailto:bes=
s@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi John,

 Firstly i think EVPN community should reach consensus on the issues of cur=
rent DF election mechanism. All these issues should be resolved in a single=
 new DF election draft,rather than in multiple separate drafts. If your dra=
ft can solve all these issues and stable, i have no question for its progre=
ssing. But if your draft have not solved all issues, i think it had better =
combine with other drafts to provide a comprehensive solution. I think the =
issues listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-d=
f-handshaking-00 is valid, it should be resolved. So i think although your =
new Hash algorithm for DF election is good, it only includes partial enhanc=
ements, maybe it still needs some time for consensus.



Thanks,

weiguo

________________________________
From: John E Drake [jdrake@juniper.net<mailto:jdrake@juniper.net>]
Sent: Thursday, March 26, 2015 7:16
To: Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm
Weiguo,

Your proposal introduces a control plane processing load that is O(#EVIs * =
PEs) per DF election and given that there can be 4K EVIs per ES, this looks=
 like a *substantial* load.  Furthermore,
you can't  use the ES route to co-ordinate DF election because you would ne=
ed to carry your new extended community for each EVI and they would not all=
 fit.  You also can't use the Per EVI Ethernet AD route because that is pro=
cessed by all PEs in the EVI.

I think that from a practical perspective the new DF election proposed in S=
atya's draft is sufficiently stable that it renders your draft moot, even i=
f it could be made to work.
Yours Irrespectively,

John

From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
Sent: Wednesday, March 25, 2015 5:34 PM
To: Ali Sajassi (sajassi); bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm


Hi Ali,

Thanks for your information. I scanned through this draft, it really introd=
uces inter-chassis message for DF election handshaking, the requirements in=
 this draft is to eliminate transiet Loop and traffic duplication. Current =
EVPN DF election mechanism already eliminated loop and traffic duplication =
by configuring long reception timer on each multi-homed PE, but up to recep=
tion timer traffic disruption issue still exist. EVPN for DCI is an importa=
nt use case for EVPN, up to reception timer traffic disruption can't be tol=
erated for service providers, it should be improved.

Also for accuracy, i think handshaking state machine on each multi-homed PE=
 is also needed. From solution perspective, in my draft, no inter-chassis m=
essage is introduced, only one new extended community is introduced, i thin=
k the process is comparatively simple than your following draft.



Current EVPN DF election has some drawbacks, so there are three new drafts =
about DF election emerged. I think BESS WG can consider these three drafts =
in global view,  a single,comprehensive new DF election draft is hoped.



thanks.



Thanks,

weiguo

________________________________
From: BESS [bess-bounces@ietf.org<mailto:bess-bounces@ietf.org>] on behalf =
of Ali Sajassi (sajassi) [sajassi@cisco.com<mailto:sajassi@cisco.com>]
Sent: Thursday, March 26, 2015 0:21
To: bess@ietf.org<mailto:bess@ietf.org>
Subject: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorith=
m

FYI- First published July 4, 2011

https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/


-Ali

--_000_BLUPR05MB562A3B02E72B599D65C1F42C7090BLUPR05MB562namprd_
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=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)">
<!--[if !mso]><style>v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style><![endif]--><style><!--
/* Font Definitions */
@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:Consolas;
	panose-1:2 11 6 9 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;
	margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman",serif;}
pre
	{mso-style-priority:99;
	mso-style-link:"HTML Preformatted Char";
	margin:0in;
	margin-bottom:.0001pt;
	font-size:10.0pt;
	font-family:Consolas;}
span.HTMLPreformattedChar
	{mso-style-name:"HTML Preformatted Char";
	mso-style-priority:99;
	mso-style-link:"HTML Preformatted";
	font-family:Consolas;}
p.msochpdefault, li.msochpdefault, div.msochpdefault
	{mso-style-name:msochpdefault;
	margin:0in;
	margin-bottom:.0001pt;
	font-size:10.0pt;
	font-family:"Times New Roman",serif;}
span.htmlpreformattedchar0
	{mso-style-name:htmlpreformattedchar;
	font-family:Consolas;}
span.emailstyle18
	{mso-style-name:emailstyle18;
	font-family:"Calibri",sans-serif;
	color:#1F497D;}
span.emailstyle21
	{mso-style-name:emailstyle21;
	font-family:"Calibri",sans-serif;
	color:#1F497D;}
span.htmlpreformattedchar00
	{mso-style-name:htmlpreformattedchar0;
	font-family:Consolas;}
span.emailstyle25
	{mso-style-name:emailstyle25;
	font-family:"Calibri",sans-serif;
	color:#1F497D;}
span.EmailStyle27
	{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=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;,sans-serif;color:#1F497D">Weiguo,<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">I guess I wasn&#8217;t clear.&nbsp; I=
 think you draft, for the reasons I have detailed, is a non-solution to a n=
on-problem with tremendous control plane cost.<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D"><o:p>&nbsp;</o:p></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Yours Irrespectively,<o:p></o:p></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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;,sans-serif;color:#1F497D">John<o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;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 #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif">From:</span></b><span style=3D"font-size:11.0pt;=
font-family:&quot;Calibri&quot;,sans-serif"> Haoweiguo [mailto:haoweiguo@hu=
awei.com]
<br>
<b>Sent:</b> Thursday, March 26, 2015 6:17 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); bess@ietf.org<br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm<o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">P=
ls see below.</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahom=
a&quot;,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-seri=
f;color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">T=
hanks,</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;=
,sans-serif;color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,=
sans-serif;color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF887072">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> John E Drake [jdrake@juniper.net]<br>
<b>Sent:</b> Friday, March 27, 2015 6:00<br>
<b>To:</b> Haoweiguo; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org">bess@iet=
f.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">To recap,</span><span style=3D"color:=
black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">We have established that your proposa=
l is untenable because of its control plane load.
</span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">We have established that your proposa=
l is based upon a flawed understanding of the DF election in RFC 7432.</spa=
n><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">[weiguo]:&nbsp; In ethernet world, tr=
affic loop is serious than short timer traffic disruption. If you want to i=
mplement &nbsp;transiet traffic loop process, i will modify
 my draft to solve your issue.</span><span style=3D"color:black"><o:p></o:p=
></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">If i am the developer, i will prefer =
short timer traffic disruption based on current EVPN protocol.</span><span =
style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">What you are now arguing is that your=
 draft prevents two or more PEs from being DF simultaneously.&nbsp; This is=
 clearly nonsense.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">[weiguo]: I will modify the draft pro=
blem statements, and use the same handshaking solution to solve it.</span><=
span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Furthermore, we have established that=
 having two or more DFs for what even you admit is a brief transient leads =
to duplicate traffic, which is acceptable, but
 not loops, your assertion to the contrary.</span><span style=3D"color:blac=
k"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">[weiguo]: It is transient loop and tr=
affic duplication issue.</span><span style=3D"color:black"><o:p></o:p></spa=
n></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Yours Irrespectively,</span><span sty=
le=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">John</span><span style=3D"color:black=
"><o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></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 #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:</span></b><span style=3D"font-=
size:11.0pt;font-family:&quot;Calibri&quot;,sans-serif;color:black"> Haowei=
guo [<a href=3D"mailto:haoweiguo@huawei.com">mailto:haoweiguo@huawei.com</a=
>]
<br>
<b>Sent:</b> Thursday, March 26, 2015 5:37 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org">bess@iet=
f.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">J=
ohn,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">A=
s your understanding of the EVPN draft, the DF election mechanism has more =
serious side effect, it will have short time traffic loop,i.e., dual DF PEs=
 will exist for a short time. I think dual DF
 PEs is absolutely not tolerated, because native ethernet header has no TTL=
, up to several hundred ms traffic loop normally not tolerated in commertia=
l networks.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">A=
s your understanding, the PEs should do as following:</span><span style=3D"=
color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">1=
. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have rarely =
deployment.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">A=
ssuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins ESI and=
 trigger DF re-election. When reception timer expires:</span><span style=3D=
"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">P=
E1 upgrades to DF PE.</span><span style=3D"color:black"><o:p></o:p></span><=
/p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">A=
fter reception timer&#43; ES route transmission timer:</span><span style=3D=
"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">P=
E2 downloads to non-DF PE.</span><span style=3D"color:black"><o:p></o:p></s=
pan></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">S=
o in timer clock sync case, dual DF PEs will exist at least transmission ti=
mer.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">I=
f NTP is used for timer sync, because it has bad accuracy, dual DF PEs will=
 exist more longer timer.</span><span style=3D"color:black"><o:p></o:p></sp=
an></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">S=
o as your understanding for DF election, the drawback is more clear.</span>=
<span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">T=
hanks,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF234411">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> John E Drake [jdrake@juniper.net]<br>
<b>Sent:</b> Friday, March 27, 2015 4:41<br>
<b>To:</b> Haoweiguo; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Weiguo,</span><span style=3D"color:bl=
ack"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">We have already established that your=
 proposal is untenable because of its control plane load.
</span><span style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">What we are now discussing is that yo=
ur proposal is based upon a misunderstanding of the algorithm in RFC 7432.&=
nbsp; You are assuming that PE1 will advertise an ES
 route and then wait for the configured interval before performing the DF e=
lection while PE2 and PE3 will perform the DF election as soon as they rece=
ive the ES route from PE1.&nbsp; This is not what RFC 7432 says.</span><spa=
n style=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Rather, what is says is that the adve=
rtisement of the ES route by PE1 and its receipt by PE2 and PE3 causes all =
three PEs to start the configured interval timer&nbsp;
 - &#8220;3. When the timer expires, each PE builds an ordered list of the =
IP addresses of all the PE nodes connected to the Ethernet segment (includi=
ng itself), in increasing numeric value.&#8221;</span><span style=3D"color:=
black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Yours Irrespectively,</span><span sty=
le=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">John</span><span style=3D"color:black=
"><o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></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 #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:</span></b><span style=3D"font-=
size:11.0pt;font-family:&quot;Calibri&quot;,sans-serif;color:black"> Haowei=
guo [<a href=3D"mailto:haoweiguo@huawei.com" target=3D"_blank">mailto:haowe=
iguo@huawei.com</a>]
<br>
<b>Sent:</b> Thursday, March 26, 2015 3:26 PM<br>
<b>To:</b> John E Drake; Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">P=
ls read my detail replies to Satya. If you still can't catch it, pls read m=
y draft and EVPN base protocol, &nbsp;thanks</span><span style=3D"color:bla=
ck"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF478649">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> John E Drake [jdrake@juniper.net]<br>
<b>Sent:</b> Friday, March 27, 2015 1:28<br>
<b>To:</b> Patrice Brissette (pbrisset)<br>
<b>Cc:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">I think Patrice is corre=
ct. &nbsp;Your proposal doesn't solve the problem and it does so at huge co=
st.<br>
<br>
Sent from my iPhone<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><span style=3D"color:=
black"><br>
On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) &lt;<a href=3D"m=
ailto:pbrisset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>&gt; wrot=
e:<o:p></o:p></span></p>
</div>
<blockquote style=3D"margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Weiguo,<o:p></o:p></span=
></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">I&#8217;m not sure I&#82=
17;m following here.<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Don&#8217;t you have the=
 same issue with your handshaking mechanism?&nbsp;<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">If you don&#8217;t know =
your peer, how can you handshake?<o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Regards,<o:p></o:p></spa=
n></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><span style=3D"color:=
black">Patrice<o:p></o:p></span></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td style=3D"padding:0in 0in 0in 0in">
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"border:solid windowtext =
1.0pt;padding:0in"><img border=3D"0" width=3D"110" height=3D"73" id=3D"_x00=
00_i1028" src=3D"cid:image001.jpg@01D06808.9CF93830" alt=3D"Image removed b=
y sender."></span><o:p></o:p></p>
</td>
</tr>
<tr>
<td nowrap=3D"" valign=3D"top" style=3D"padding:0in 0in 0in .25in">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Patrice Brissette</span></strong><span style=3D"font=
-size:8.5pt;font-family:&quot;Arial&quot;,sans-serif;color:#666666"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"colo=
r:#666666;text-decoration:none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"font-family:&quot;Arial&quot;,sans-serif">&#4=
3;1 613 254 3336</span></strong></span><o:p></o:p></p>
</td>
<td width=3D"50%" valign=3D"top" style=3D"width:50.0%;padding:11.25pt 0in 7=
.5pt 15.0pt">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Cisco Systems Canada Co. / Les Systemes Cisco Canada=
 CIE</span></strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&=
quot;,sans-serif;color:#666666"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"color:#666666;text-decoration:none">Cisco.com</span></a></span><o:p></o=
:p></p>
</td>
<td style=3D"padding:0in 0in 0in 0in"></td>
</tr>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in"></td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal">&nbsp;<o:p></o:p></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"400" style=3D"width:300.0pt">
<tbody>
<tr>
<td style=3D"padding:3.75pt 15.0pt 0in .25in">
<p class=3D"MsoNormal"><span style=3D"font-size:7.5pt;font-family:&quot;Ari=
al&quot;,sans-serif;color:#009900;border:solid windowtext 1.0pt;padding:0in=
"><img border=3D"0" width=3D"100" height=3D"100" id=3D"_x0000_i1029" src=3D=
"cid:image002.jpg@01D06808.9CF93830" alt=3D"Image removed by sender."></spa=
n><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;c=
olor:#009900">Think
 before you print.</span><o:p></o:p></p>
</td>
</tr>
<tr>
<td style=3D"padding:5.25pt 15.0pt 4.5pt .25in">
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">This email may contain confidential and privileged material =
for the sole use of the intended recipient. Any review, use, distribution o=
r disclosure by others is strictly prohibited.
 If you are not the intended recipient (or authorized to receive for the re=
cipient), please contact the sender by reply email and delete all copies of=
 this message.</span><o:p></o:p></p>
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">Please
<a href=3D"http://www.cisco.com/web/about/doing_business/legal/cri/index.ht=
ml" target=3D"_blank" title=3D"Legal Information">
<span style=3D"color:#0E58A0">click here</span></a> for Company Registratio=
n Information.</span><o:p></o:p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"color:black"><br clear=3D"all">
<o:p></o:p></span></p>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></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:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:
</span></b><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,=
sans-serif;color:black">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.co=
m" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Thursday, March 26, 2015 at 10:24 AM<br>
<b>To: </b>Patrice Brissette &lt;<a href=3D"mailto:pbrisset@cisco.com" targ=
et=3D"_blank">pbrisset@cisco.com</a>&gt;, John E Drake &lt;<a href=3D"mailt=
o:jdrake@juniper.net" target=3D"_blank">jdrake@juniper.net</a>&gt;, Ali Saj=
assi &lt;<a href=3D"mailto:sajassi@cisco.com" target=3D"_blank">sajassi@cis=
co.com</a>&gt;,
 &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>=
&quot; &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a>&gt;<br>
<b>Subject: </b>RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<blockquote style=3D"border:none;border-left:solid #B5C4DF 4.5pt;padding:0i=
n 0in 0in 4.0pt;margin-left:3.75pt;margin-top:5.0pt;margin-right:0in;margin=
-bottom:5.0pt" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">H=
i Patrice,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">U=
p to reception timer traffic disruption in transient phase is one of the is=
sues.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">T=
hanks,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF422254">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> Patrice Brissette (pbrisset) [<a href=3D"mailto:pbri=
sset@cisco.com" target=3D"_blank">pbrisset@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 20:54<br>
<b>To:</b> Haoweiguo; John E Drake; Ali Sajassi (sajassi); <a href=3D"mailt=
o:bess@ietf.org" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black">Weiguo,</span><span style=3D"color:blac=
k"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-family:&quot;Calibri&quot;,sans-=
serif;color:black">You mention &quot;</span><span style=3D"color:black">But=
 if your draft have not solved all issues&#8221;,<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">Can you explain what Sat=
ya&#8217;s draft is not solving?<o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black">Regards,</span><span style=3D"color:bla=
ck"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><span style=3D"font-s=
ize:10.5pt;font-family:&quot;Calibri&quot;,sans-serif;color:black">Patrice<=
/span><span style=3D"color:black"><o:p></o:p></span></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td style=3D"padding:0in 0in 0in 0in">
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"543" style=3D"width:407.25pt">
<tbody>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in">
<p class=3D"MsoNormal">&nbsp; &nbsp;<span style=3D"border:solid windowtext =
1.0pt;padding:0in"><img border=3D"0" width=3D"110" height=3D"73" id=3D"_x00=
00_i1031" src=3D"cid:image001.jpg@01D06808.9CF93830" alt=3D"Image removed b=
y sender."></span><o:p></o:p></p>
</td>
</tr>
<tr>
<td nowrap=3D"" valign=3D"top" style=3D"padding:0in 0in 0in .25in">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Patrice Brissette</span></strong><span style=3D"font=
-size:8.5pt;font-family:&quot;Arial&quot;,sans-serif;color:#666666"><br>
TECHNICAL LEADER.ENGINEERING<br>
<br>
<a href=3D"mailto:pbrisset@cisco.com" target=3D"_blank"><span style=3D"colo=
r:#666666;text-decoration:none">pbrisset@cisco.com</span></a><br>
Phone: <strong><span style=3D"font-family:&quot;Arial&quot;,sans-serif">&#4=
3;1 613 254 3336</span></strong></span><o:p></o:p></p>
</td>
<td width=3D"50%" valign=3D"top" style=3D"width:50.0%;padding:11.25pt 0in 7=
.5pt 15.0pt">
<p><strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&quot;,san=
s-serif;color:#666666">Cisco Systems Canada Co. / Les Systemes Cisco Canada=
 CIE</span></strong><span style=3D"font-size:8.5pt;font-family:&quot;Arial&=
quot;,sans-serif;color:#666666"><br>
Canada<br>
<a href=3D"http://www.cisco.com/global/CA/" target=3D"_blank"><span style=
=3D"color:#666666;text-decoration:none">Cisco.com</span></a></span><o:p></o=
:p></p>
</td>
<td style=3D"padding:0in 0in 0in 0in"></td>
</tr>
<tr>
<td colspan=3D"3" style=3D"padding:0in 0in 0in 0in"></td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal">&nbsp;<o:p></o:p></p>
<table class=3D"MsoNormalTable" border=3D"0" cellspacing=3D"0" cellpadding=
=3D"0" width=3D"400" style=3D"width:300.0pt">
<tbody>
<tr>
<td style=3D"padding:3.75pt 15.0pt 0in .25in">
<p class=3D"MsoNormal"><span style=3D"font-size:7.5pt;font-family:&quot;Ari=
al&quot;,sans-serif;color:#009900;border:solid windowtext 1.0pt;padding:0in=
"><img border=3D"0" width=3D"100" height=3D"100" id=3D"_x0000_i1032" src=3D=
"cid:image002.jpg@01D06808.9CF93830" alt=3D"Image removed by sender."></spa=
n><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;c=
olor:#009900">Think
 before you print.</span><o:p></o:p></p>
</td>
</tr>
<tr>
<td style=3D"padding:5.25pt 15.0pt 4.5pt .25in">
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">This email may contain confidential and privileged material =
for the sole use of the intended recipient. Any review, use, distribution o=
r disclosure by others is strictly prohibited.
 If you are not the intended recipient (or authorized to receive for the re=
cipient), please contact the sender by reply email and delete all copies of=
 this message.</span><o:p></o:p></p>
<p><span style=3D"font-size:7.5pt;font-family:&quot;Arial&quot;,sans-serif;=
color:#999999">Please
<a href=3D"http://www.cisco.com/web/about/doing_business/legal/cri/index.ht=
ml" target=3D"_blank" title=3D"Legal Information">
<span style=3D"color:#0E58A0">click here</span></a> for Company Registratio=
n Information.</span><o:p></o:p></p>
</td>
</tr>
</tbody>
</table>
</td>
</tr>
</tbody>
</table>
<p class=3D"MsoNormal"><span style=3D"font-size:10.5pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:black"><br clear=3D"all">
</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
</div>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></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:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:
</span></b><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,=
sans-serif;color:black">Haoweiguo &lt;<a href=3D"mailto:haoweiguo@huawei.co=
m" target=3D"_blank">haoweiguo@huawei.com</a>&gt;<br>
<b>Date: </b>Wednesday, March 25, 2015 at 10:38 PM<br>
<b>To: </b>John E Drake &lt;<a href=3D"mailto:jdrake@juniper.net" target=3D=
"_blank">jdrake@juniper.net</a>&gt;, Ali Sajassi &lt;<a href=3D"mailto:saja=
ssi@cisco.com" target=3D"_blank">sajassi@cisco.com</a>&gt;, &quot;<a href=
=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&quot;
 &lt;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</a>&g=
t;<br>
<b>Subject: </b>Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<blockquote style=3D"border:none;border-left:solid #B5C4DF 4.5pt;padding:0i=
n 0in 0in 4.0pt;margin-left:3.75pt;margin-top:5.0pt;margin-right:0in;margin=
-bottom:5.0pt" id=3D"MAC_OUTLOOK_ATTRIBUTION_BLOCKQUOTE">
<div>
<div>
<div>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">H=
i John,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">&=
nbsp;Firstly i think EVPN community should reach consensus on the issues of=
 current DF election mechanism. All these issues should be resolved in a si=
ngle new DF election draft,rather than in multiple
 separate drafts. If your draft can solve all these issues and stable, i ha=
ve no question for its progressing. But if your draft have not solved all i=
ssues, i think it had better combine with other drafts to provide a compreh=
ensive solution. I think the issues
 listed in draft-rabadan-bess-evpn-ac-df-01 and draft-hao-bess-evpn-df-hand=
shaking-00 is valid, it should be resolved. So i think although your new Ha=
sh algorithm for DF election is good, it only includes partial enhancements=
, maybe it still needs some time
 for consensus.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">T=
hanks,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Calibri&quot;,sans-serif;color:black">w=
eiguo</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF271558">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> John E Drake [<a href=3D"mailto:jdrake@juniper.net" =
target=3D"_blank">jdrake@juniper.net</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 7:16<br>
<b>To:</b> Haoweiguo; Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.or=
g" target=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> RE: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Weiguo,</span><span style=3D"color:bl=
ack"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Your proposal introduces a control pl=
ane processing load that is O(#EVIs * PEs) per DF election and given that t=
here can be 4K EVIs per ES, this looks like a
 *<b>substantial</b>* load.&nbsp; Furthermore,</span><span style=3D"color:b=
lack"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">you can&#8217;t &nbsp;use the ES rout=
e to co-ordinate DF election because you would need to carry your new exten=
ded community for each EVI and they would not all fit.&nbsp; You
 also can&#8217;t use the Per EVI Ethernet AD route because that is process=
ed by all PEs in the EVI.</span><span style=3D"color:black"><o:p></o:p></sp=
an></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">I think that from a practical perspec=
tive the new DF election proposed in Satya&#8217;s draft is sufficiently st=
able that it renders your draft moot, even if it could
 be made to work.</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">Yours Irrespectively,</span><span sty=
le=3D"color:black"><o:p></o:p></span></p>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1F497D">John</span><span style=3D"color:black=
"><o:p></o:p></span></p>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></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 #E1E1E1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif;color:black">From:</span></b><span style=3D"font-=
size:11.0pt;font-family:&quot;Calibri&quot;,sans-serif;color:black"> BESS [=
<a href=3D"mailto:bess-bounces@ietf.org" target=3D"_blank">mailto:bess-boun=
ces@ietf.org</a>]
<b>On Behalf Of </b>Haoweiguo<br>
<b>Sent:</b> Wednesday, March 25, 2015 5:34 PM<br>
<b>To:</b> Ali Sajassi (sajassi); <a href=3D"mailto:bess@ietf.org" target=
=3D"_blank">
bess@ietf.org</a><br>
<b>Subject:</b> Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os algorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
</div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
<div>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Hi=
 Ali,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Th=
anks for your information. I scanned through this draft, it really introduc=
es inter-chassis message for DF election handshaking, the requirements in t=
his draft is to eliminate transiet Loop and
 traffic duplication. Current EVPN DF election mechanism already eliminated=
 loop and traffic duplication by configuring long reception timer on each m=
ulti-homed PE, but up to reception timer traffic disruption issue still exi=
st. EVPN for DCI is an important
 use case for EVPN, up to reception timer traffic disruption can't be toler=
ated for service providers, it should be improved.</span><span style=3D"col=
or:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Al=
so for accuracy, i think handshaking state machine on each multi-homed PE i=
s also needed. From solution perspective, in my draft, no inter-chassis mes=
sage is introduced, only one new extended community
 is introduced, i think the process is comparatively simple than your follo=
wing draft.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Cu=
rrent EVPN DF election has some drawbacks, so there are three new drafts ab=
out DF election emerged. I think BESS WG can consider these three drafts in=
 global view,&nbsp; a single,comprehensive new DF
 election draft is hoped.</span><span style=3D"color:black"><o:p></o:p></sp=
an></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">th=
anks.</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"color:black">&nbsp;<o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">Th=
anks,</span><span style=3D"color:black"><o:p></o:p></span></p>
<p><span style=3D"font-family:&quot;Tahoma&quot;,sans-serif;color:black">we=
iguo</span><span style=3D"color:black"><o:p></o:p></span></p>
<div>
<div class=3D"MsoNormal" align=3D"center" style=3D"text-align:center"><span=
 style=3D"color:black">
<hr size=3D"2" width=3D"100%" align=3D"center">
</span></div>
<div id=3D"divRpF484031">
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><b><span style=3D"fon=
t-size:10.0pt;font-family:&quot;Tahoma&quot;,sans-serif;color:black">From:<=
/span></b><span style=3D"font-size:10.0pt;font-family:&quot;Tahoma&quot;,sa=
ns-serif;color:black"> BESS [<a href=3D"mailto:bess-bounces@ietf.org" targe=
t=3D"_blank">bess-bounces@ietf.org</a>]
 on behalf of Ali Sajassi (sajassi) [<a href=3D"mailto:sajassi@cisco.com" t=
arget=3D"_blank">sajassi@cisco.com</a>]<br>
<b>Sent:</b> Thursday, March 26, 2015 0:21<br>
<b>To:</b> <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org=
</a><br>
<b>Subject:</b> [bess] Handshaking among PEs in an EVPN ES based on Paxos a=
lgorithm</span><span style=3D"color:black"><o:p></o:p></span></p>
</div>
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">FYI- First published Jul=
y 4, 2011<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black"><a href=3D"https://datat=
racker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route" target=3D"_blan=
k">https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route<=
/a>/<o:p></o:p></span></p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">&nbsp;<o:p></o:p></span>=
</p>
</div>
<div>
<p class=3D"MsoNormal"><span style=3D"color:black">-Ali<o:p></o:p></span></=
p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_BLUPR05MB562A3B02E72B599D65C1F42C7090BLUPR05MB562namprd_--

--_005_BLUPR05MB562A3B02E72B599D65C1F42C7090BLUPR05MB562namprd_
Content-Type: image/jpeg; name="image001.jpg"
Content-Description: image001.jpg
Content-Disposition: inline; filename="image001.jpg"; size=437;
	creation-date="Fri, 27 Mar 2015 01:05:41 GMT";
	modification-date="Fri, 27 Mar 2015 01:05:41 GMT"
Content-ID: <image001.jpg@01D06808.9CF93830>
Content-Transfer-Encoding: base64

/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAoHBwkHBgoJCAkLCwoMDxkQDw4ODx4WFxIZJCAmJSMg
IyIoLTkwKCo2KyIjMkQyNjs9QEBAJjBGS0U+Sjk/QD3/wAALCABJAG4BAREA/8QAHwAAAQUBAQEB
AQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1Fh
ByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZ
WmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXG
x8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/9oACAEBAAA/APZqKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK
KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK//9k=

--_005_BLUPR05MB562A3B02E72B599D65C1F42C7090BLUPR05MB562namprd_
Content-Type: image/jpeg; name="image002.jpg"
Content-Description: image002.jpg
Content-Disposition: inline; filename="image002.jpg"; size=823;
	creation-date="Fri, 27 Mar 2015 01:05:41 GMT";
	modification-date="Fri, 27 Mar 2015 01:05:41 GMT"
Content-ID: <image002.jpg@01D06808.9CF93830>
Content-Transfer-Encoding: base64
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--_005_BLUPR05MB562A3B02E72B599D65C1F42C7090BLUPR05MB562namprd_--


From nobody Thu Mar 26 19:50:47 2015
Return-Path: <thomas.morin@orange.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 89AFB1A905D for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 19:50:45 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FREEMAIL_FROM=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001, UNPARSEABLE_RELAY=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ufhF3FORaYfs for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 19:50:41 -0700 (PDT)
Received: from relais-inet.francetelecom.com (relais-ias245.francetelecom.com [80.12.204.245]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 097151A9059 for <bess@ietf.org>; Thu, 26 Mar 2015 19:50:40 -0700 (PDT)
Received: from omfeda07.si.francetelecom.fr (unknown [xx.xx.xx.200]) by omfeda14.si.francetelecom.fr (ESMTP service) with ESMTP id 7B0A82AC3C2 for <bess@ietf.org>; Fri, 27 Mar 2015 03:50:38 +0100 (CET)
Received: from Exchangemail-eme1.itn.ftgroup (unknown [10.114.1.183]) by omfeda07.si.francetelecom.fr (ESMTP service) with ESMTP id 5ACCD15805A for <bess@ietf.org>; Fri, 27 Mar 2015 03:50:38 +0100 (CET)
Received: from [10.193.116.37] (10.197.38.2) by PEXCVZYH02.corporate.adroot.infra.ftgroup (10.114.1.183) with Microsoft SMTP Server (TLS) id 14.3.224.2; Fri, 27 Mar 2015 03:50:37 +0100
Message-ID: <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com>
Date: Thu, 26 Mar 2015 21:50:35 -0500
From: <thomas.morin@orange.com>
Organization: Orange
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: <bess@ietf.org>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com>
In-Reply-To: <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit
X-Originating-IP: [10.197.38.2]
X-PMX-Version: 6.2.1.2478543, Antispam-Engine: 2.7.2.2107409, Antispam-Data: 2015.3.26.230622
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/zcBH7uCsrSxLV2sWGldOeEWeYdQ>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 02:50:45 -0000

Hi,

Let me attempt at clarifying the exchange between the few of you.

My understanding is that:
- RFC7432 already does a good job of avoiding forwarding loops thanks to 
the split-horizon procedure that does not depend on DF election and has 
no transient state
- RFC7432 may have transient periods where the DF election state is not 
yet synchronized between the two peers:
   * the transient period correspond to BGP route propagation times  
(not to the DF election delay, although the wording in RFC 7432 could be 
made clearer)
   * during this period, if PEs do not block BUM traffic, some traffic 
may be lost
   * during this period, if PEs do not block BUM traffic, some traffic 
may be duplicated, but this is not considered very harmful and hence 
preferred to packet loss

The opinion, as I understand, that has been expressed by some is that 
the reduction of the transient period where duplicates may arise, is not 
worth the cost of an handshaking approach.

HTH,

-Thomas


26/03/2015 20:05, John E Drake :
>
> Weiguo,
>
> I guess I wasn’t clear.  I think you draft, for the reasons I have 
> detailed, is a non-solution to a non-problem with tremendous control 
> plane cost.
>
> Yours Irrespectively,
>
> John
>
> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> *Sent:* Thursday, March 26, 2015 6:17 PM
> *To:* John E Drake; Patrice Brissette (pbrisset)
> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on 
> Paxos algorithm
>
> Pls see below.
>
> Thanks,
>
> weiguo
>
> ------------------------------------------------------------------------
>
> *From:*John E Drake [jdrake@juniper.net]
> *Sent:* Friday, March 27, 2015 6:00
> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on 
> Paxos algorithm
>
> To recap,
>
> We have established that your proposal is untenable because of its 
> control plane load.
>
> We have established that your proposal is based upon a flawed 
> understanding of the DF election in RFC 7432.
>
> [weiguo]: In ethernet world, traffic loop is serious than short timer 
> traffic disruption. If you want to implement  transiet traffic loop 
> process, i will modify my draft to solve your issue.
>
> If i am the developer, i will prefer short timer traffic disruption 
> based on current EVPN protocol.
>
> What you are now arguing is that your draft prevents two or more PEs 
> from being DF simultaneously. This is clearly nonsense.
>
> [weiguo]: I will modify the draft problem statements, and use the same 
> handshaking solution to solve it.
>
> Furthermore, we have established that having two or more DFs for what 
> even you admit is a brief transient leads to duplicate traffic, which 
> is acceptable, but not loops, your assertion to the contrary.
>
> [weiguo]: It is transient loop and traffic duplication issue.
>
> Yours Irrespectively,
>
> John
>
> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> *Sent:* Thursday, March 26, 2015 5:37 PM
> *To:* John E Drake; Patrice Brissette (pbrisset)
> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on 
> Paxos algorithm
>
> John,
>
> As your understanding of the EVPN draft, the DF election mechanism has 
> more serious side effect, it will have short time traffic loop,i.e., 
> dual DF PEs will exist for a short time. I think dual DF PEs is 
> absolutely not tolerated, because native ethernet header has no TTL, 
> up to several hundred ms traffic loop normally not tolerated in 
> commertial networks.
>
> As your understanding, the PEs should do as following:
>
> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have 
> rarely deployment.
>
> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins 
> ESI and trigger DF re-election. When reception timer expires:
>
> PE1 upgrades to DF PE.
>
> After reception timer+ ES route transmission timer:
>
> PE2 downloads to non-DF PE.
>
> So in timer clock sync case, dual DF PEs will exist at least 
> transmission timer.
>
> If NTP is used for timer sync, because it has bad accuracy, dual DF 
> PEs will exist more longer timer.
>
> So as your understanding for DF election, the drawback is more clear.
>
> Thanks,
>
> weiguo
>
> ------------------------------------------------------------------------
>
> *From:*John E Drake [jdrake@juniper.net]
> *Sent:* Friday, March 27, 2015 4:41
> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on 
> Paxos algorithm
>
> Weiguo,
>
> We have already established that your proposal is untenable because of 
> its control plane load.
>
> What we are now discussing is that your proposal is based upon a 
> misunderstanding of the algorithm in RFC 7432.  You are assuming that 
> PE1 will advertise an ES route and then wait for the configured 
> interval before performing the DF election while PE2 and PE3 will 
> perform the DF election as soon as they receive the ES route from PE1. 
> This is not what RFC 7432 says.
>
> Rather, what is says is that the advertisement of the ES route by PE1 
> and its receipt by PE2 and PE3 causes all three PEs to start the 
> configured interval timer  - “3. When the timer expires, each PE 
> builds an ordered list of the IP addresses of all the PE nodes 
> connected to the Ethernet segment (including itself), in increasing 
> numeric value.”
>
> Yours Irrespectively,
>
> John
>
> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> *Sent:* Thursday, March 26, 2015 3:26 PM
> *To:* John E Drake; Patrice Brissette (pbrisset)
> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on 
> Paxos algorithm
>
> Pls read my detail replies to Satya. If you still can't catch it, pls 
> read my draft and EVPN base protocol,  thanks
>
> weiguo
>
> ------------------------------------------------------------------------
>
> *From:*John E Drake [jdrake@juniper.net]
> *Sent:* Friday, March 27, 2015 1:28
> *To:* Patrice Brissette (pbrisset)
> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org 
> <mailto:bess@ietf.org>
> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on 
> Paxos algorithm
>
> I think Patrice is correct.  Your proposal doesn't solve the problem 
> and it does so at huge cost.
>
> Sent from my iPhone
>
>
> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset) 
> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>
>     Weiguo,
>
>     I’m not sure I’m following here.
>
>     Don’t you have the same issue with your handshaking mechanism?
>
>     If you don’t know your peer, how can you handshake?
>
>     Regards,
>
>     Patrice
>
>     Image removed by sender.
>
>     *Patrice Brissette*
>     TECHNICAL LEADER.ENGINEERING
>
>     pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>     Phone: *+1 613 254 3336*
>
>     	
>
>     *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>     Canada
>     Cisco.com <http://www.cisco.com/global/CA/>
>
>     	
>
>     Image removed by sender.Think before you print.
>
>     This email may contain confidential and privileged material for
>     the sole use of the intended recipient. Any review, use,
>     distribution or disclosure by others is strictly prohibited. If
>     you are not the intended recipient (or authorized to receive for
>     the recipient), please contact the sender by reply email and
>     delete all copies of this message.
>
>     Please click here
>     <http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>     for Company Registration Information.
>
>
>     *From: *Haoweiguo <haoweiguo@huawei.com <mailto:haoweiguo@huawei.com>>
>     *Date: *Thursday, March 26, 2015 at 10:24 AM
>     *To: *Patrice Brissette <pbrisset@cisco.com
>     <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>     <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>     <mailto:sajassi@cisco.com>>, "bess@ietf.org
>     <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>     *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES based on
>     Paxos algorithm
>
>         Hi Patrice,
>
>         Up to reception timer traffic disruption in transient phase is
>         one of the issues.
>
>         Thanks,
>
>         weiguo
>
>         ------------------------------------------------------------------------
>
>         *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>         <mailto:pbrisset@cisco.com>]
>         *Sent:* Thursday, March 26, 2015 20:54
>         *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>         bess@ietf.org <mailto:bess@ietf.org>
>         *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>         based on Paxos algorithm
>
>         Weiguo,
>
>         You mention "But if your draft have not solved all issues”,
>
>         Can you explain what Satya’s draft is not solving?
>
>         Regards,
>
>         Patrice
>
>         Image removed by sender.
>
>         *Patrice Brissette*
>         TECHNICAL LEADER.ENGINEERING
>
>         pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>         Phone: *+1 613 254 3336*
>
>         	
>
>         *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>         Canada
>         Cisco.com <http://www.cisco.com/global/CA/>
>
>         	
>
>         Image removed by sender.Think before you print.
>
>         This email may contain confidential and privileged material
>         for the sole use of the intended recipient. Any review, use,
>         distribution or disclosure by others is strictly prohibited.
>         If you are not the intended recipient (or authorized to
>         receive for the recipient), please contact the sender by reply
>         email and delete all copies of this message.
>
>         Please click here
>         <http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>         for Company Registration Information.
>
>
>         *From: *Haoweiguo <haoweiguo@huawei.com
>         <mailto:haoweiguo@huawei.com>>
>         *Date: *Wednesday, March 25, 2015 at 10:38 PM
>         *To: *John E Drake <jdrake@juniper.net
>         <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>         <mailto:sajassi@cisco.com>>, "bess@ietf.org
>         <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>         *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>         based on Paxos algorithm
>
>             Hi John,
>
>              Firstly i think EVPN community should reach consensus on
>             the issues of current DF election mechanism. All these
>             issues should be resolved in a single new DF election
>             draft,rather than in multiple separate drafts. If your
>             draft can solve all these issues and stable, i have no
>             question for its progressing. But if your draft have not
>             solved all issues, i think it had better combine with
>             other drafts to provide a comprehensive solution. I think
>             the issues listed in draft-rabadan-bess-evpn-ac-df-01 and
>             draft-hao-bess-evpn-df-handshaking-00 is valid, it should
>             be resolved. So i think although your new Hash algorithm
>             for DF election is good, it only includes partial
>             enhancements, maybe it still needs some time for consensus.
>
>             Thanks,
>
>             weiguo
>
>             ------------------------------------------------------------------------
>
>             *From:*John E Drake [jdrake@juniper.net
>             <mailto:jdrake@juniper.net>]
>             *Sent:* Thursday, March 26, 2015 7:16
>             *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>             <mailto:bess@ietf.org>
>             *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES
>             based on Paxos algorithm
>
>             Weiguo,
>
>             Your proposal introduces a control plane processing load
>             that is O(#EVIs * PEs) per DF election and given that
>             there can be 4K EVIs per ES, this looks like a
>             **substantial** load. Furthermore,
>
>             you can’t  use the ES route to co-ordinate DF election
>             because you would need to carry your new extended
>             community for each EVI and they would not all fit.  You
>             also can’t use the Per EVI Ethernet AD route because that
>             is processed by all PEs in the EVI.
>
>             I think that from a practical perspective the new DF
>             election proposed in Satya’s draft is sufficiently stable
>             that it renders your draft moot, even if it could be made
>             to work.
>
>             Yours Irrespectively,
>
>             John
>
>             *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf Of
>             *Haoweiguo
>             *Sent:* Wednesday, March 25, 2015 5:34 PM
>             *To:* Ali Sajassi (sajassi); bess@ietf.org
>             <mailto:bess@ietf.org>
>             *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>             based on Paxos algorithm
>
>             Hi Ali,
>
>             Thanks for your information. I scanned through this draft,
>             it really introduces inter-chassis message for DF election
>             handshaking, the requirements in this draft is to
>             eliminate transiet Loop and traffic duplication. Current
>             EVPN DF election mechanism already eliminated loop and
>             traffic duplication by configuring long reception timer on
>             each multi-homed PE, but up to reception timer traffic
>             disruption issue still exist. EVPN for DCI is an important
>             use case for EVPN, up to reception timer traffic
>             disruption can't be tolerated for service providers, it
>             should be improved.
>
>             Also for accuracy, i think handshaking state machine on
>             each multi-homed PE is also needed. From solution
>             perspective, in my draft, no inter-chassis message is
>             introduced, only one new extended community is introduced,
>             i think the process is comparatively simple than your
>             following draft.
>
>             Current EVPN DF election has some drawbacks, so there are
>             three new drafts about DF election emerged. I think BESS
>             WG can consider these three drafts in global view, a
>             single,comprehensive new DF election draft is hoped.
>
>             thanks.
>
>             Thanks,
>
>             weiguo
>
>             ------------------------------------------------------------------------
>
>             *From:*BESS [bess-bounces@ietf.org
>             <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajassi
>             (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>]
>             *Sent:* Thursday, March 26, 2015 0:21
>             *To:* bess@ietf.org <mailto:bess@ietf.org>
>             *Subject:* [bess] Handshaking among PEs in an EVPN ES
>             based on Paxos algorithm
>
>             FYI- First published July 4, 2011
>
>             https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/
>
>             -Ali
>
>
>
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess













_________________________________________________________________________________________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration,
Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci.

This message and its attachments may contain confidential or privileged information that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and delete this message and its attachments.
As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified.
Thank you.


From nobody Thu Mar 26 20:48:02 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 651BE1A3BA0 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 20:48:00 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.011
X-Spam-Level: 
X-Spam-Status: No, score=-3.011 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_42=0.6, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LL12fMMDbhXL for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 20:47:57 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 47D641A1EED for <bess@ietf.org>; Thu, 26 Mar 2015 20:47:56 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml405-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT33608; Fri, 27 Mar 2015 03:47:54 +0000 (GMT)
Received: from NKGEML410-HUB.china.huawei.com (10.98.56.41) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 03:47:53 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml410-hub.china.huawei.com ([10.98.56.41]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 11:47:45 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dU=
Date: Fri, 27 Mar 2015 03:47:44 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com>, <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com>
In-Reply-To: <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.153.94]
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/hRm6BnGclrtw-F9aGmO2my2j6U4>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 03:48:00 -0000

Hi Thomas,
Pls see inline with [weiguo].

Thanks,
weiguo

________________________________________
From: BESS [bess-bounces@ietf.org] on behalf of thomas.morin@orange.com [th=
omas.morin@orange.com]
Sent: Friday, March 27, 2015 10:50
To: bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Hi,

Let me attempt at clarifying the exchange between the few of you.

My understanding is that:
- RFC7432 already does a good job of avoiding forwarding loops thanks to
the split-horizon procedure that does not depend on DF election and has
no transient state
- RFC7432 may have transient periods where the DF election state is not
yet synchronized between the two peers:
[weiguo]: Yes, i think RFC 7432 has transient periods of traffic loop(In my=
 draft, i described it into short time traffic disruption, no state consist=
ency between PEs is blamed for the transient loop or traffic disruption).  =
John and patrice think current DF election has no transiet loop problem.

   * the transient period correspond to BGP route propagation times
(not to the DF election delay, although the wording in RFC 7432 could be
made clearer)
[weiguo]: Based on the assumption of strict timer clock synchronization on =
different PEs, the transiet period is BGP route propagation timer. In real =
world, no strict timer clock sync. If NTP is used for timer sync, timer dif=
ference will be large,say A seconds. So in NTP case, the transiet period is=
 A+ BGP route propagation timer or A- BGP route propagation timer. Up to th=
e transiet timer difference traffic loop is normally not allowed in etheren=
t network, because unlike IP frame, ethernet header has no TTL, Up to 1 sec=
ond or multiple seconds transiet loop will cause whole network crashes.

   * during this period, if PEs do not block BUM traffic, some traffic
may be lost
   * during this period, if PEs do not block BUM traffic, some traffic
may be duplicated, but this is not considered very harmful and hence
preferred to packet loss
[weiguo]: During this period,  dual DF PE will exist during DF re-election =
period, traffic loop and duplication will happen. When a PE fails and recov=
er again, it will trigger DF re-election.

The opinion, as I understand, that has been expressed by some is that
the reduction of the transient period where duplicates may arise, is not
worth the cost of an handshaking approach.
[weiguo]: I think at any time in any case, traffic loop is not allowed alth=
ough the transiet timer maybe very short. In ethernet network, no traffic l=
oop is not allowed.

HTH,

-Thomas


26/03/2015 20:05, John E Drake :
>
> Weiguo,
>
> I guess I wasn=92t clear.  I think you draft, for the reasons I have
> detailed, is a non-solution to a non-problem with tremendous control
> plane cost.
>
> Yours Irrespectively,
>
> John
>
> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> *Sent:* Thursday, March 26, 2015 6:17 PM
> *To:* John E Drake; Patrice Brissette (pbrisset)
> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> Paxos algorithm
>
> Pls see below.
>
> Thanks,
>
> weiguo
>
> ------------------------------------------------------------------------
>
> *From:*John E Drake [jdrake@juniper.net]
> *Sent:* Friday, March 27, 2015 6:00
> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> Paxos algorithm
>
> To recap,
>
> We have established that your proposal is untenable because of its
> control plane load.
>
> We have established that your proposal is based upon a flawed
> understanding of the DF election in RFC 7432.
>
> [weiguo]: In ethernet world, traffic loop is serious than short timer
> traffic disruption. If you want to implement  transiet traffic loop
> process, i will modify my draft to solve your issue.
>
> If i am the developer, i will prefer short timer traffic disruption
> based on current EVPN protocol.
>
> What you are now arguing is that your draft prevents two or more PEs
> from being DF simultaneously. This is clearly nonsense.
>
> [weiguo]: I will modify the draft problem statements, and use the same
> handshaking solution to solve it.
>
> Furthermore, we have established that having two or more DFs for what
> even you admit is a brief transient leads to duplicate traffic, which
> is acceptable, but not loops, your assertion to the contrary.
>
> [weiguo]: It is transient loop and traffic duplication issue.
>
> Yours Irrespectively,
>
> John
>
> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> *Sent:* Thursday, March 26, 2015 5:37 PM
> *To:* John E Drake; Patrice Brissette (pbrisset)
> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> Paxos algorithm
>
> John,
>
> As your understanding of the EVPN draft, the DF election mechanism has
> more serious side effect, it will have short time traffic loop,i.e.,
> dual DF PEs will exist for a short time. I think dual DF PEs is
> absolutely not tolerated, because native ethernet header has no TTL,
> up to several hundred ms traffic loop normally not tolerated in
> commertial networks.
>
> As your understanding, the PEs should do as following:
>
> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have
> rarely deployment.
>
> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins
> ESI and trigger DF re-election. When reception timer expires:
>
> PE1 upgrades to DF PE.
>
> After reception timer+ ES route transmission timer:
>
> PE2 downloads to non-DF PE.
>
> So in timer clock sync case, dual DF PEs will exist at least
> transmission timer.
>
> If NTP is used for timer sync, because it has bad accuracy, dual DF
> PEs will exist more longer timer.
>
> So as your understanding for DF election, the drawback is more clear.
>
> Thanks,
>
> weiguo
>
> ------------------------------------------------------------------------
>
> *From:*John E Drake [jdrake@juniper.net]
> *Sent:* Friday, March 27, 2015 4:41
> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> Paxos algorithm
>
> Weiguo,
>
> We have already established that your proposal is untenable because of
> its control plane load.
>
> What we are now discussing is that your proposal is based upon a
> misunderstanding of the algorithm in RFC 7432.  You are assuming that
> PE1 will advertise an ES route and then wait for the configured
> interval before performing the DF election while PE2 and PE3 will
> perform the DF election as soon as they receive the ES route from PE1.
> This is not what RFC 7432 says.
>
> Rather, what is says is that the advertisement of the ES route by PE1
> and its receipt by PE2 and PE3 causes all three PEs to start the
> configured interval timer  - =933. When the timer expires, each PE
> builds an ordered list of the IP addresses of all the PE nodes
> connected to the Ethernet segment (including itself), in increasing
> numeric value.=94
>
> Yours Irrespectively,
>
> John
>
> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> *Sent:* Thursday, March 26, 2015 3:26 PM
> *To:* John E Drake; Patrice Brissette (pbrisset)
> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> Paxos algorithm
>
> Pls read my detail replies to Satya. If you still can't catch it, pls
> read my draft and EVPN base protocol,  thanks
>
> weiguo
>
> ------------------------------------------------------------------------
>
> *From:*John E Drake [jdrake@juniper.net]
> *Sent:* Friday, March 27, 2015 1:28
> *To:* Patrice Brissette (pbrisset)
> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
> <mailto:bess@ietf.org>
> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
> Paxos algorithm
>
> I think Patrice is correct.  Your proposal doesn't solve the problem
> and it does so at huge cost.
>
> Sent from my iPhone
>
>
> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>
>     Weiguo,
>
>     I=92m not sure I=92m following here.
>
>     Don=92t you have the same issue with your handshaking mechanism?
>
>     If you don=92t know your peer, how can you handshake?
>
>     Regards,
>
>     Patrice
>
>     Image removed by sender.
>
>     *Patrice Brissette*
>     TECHNICAL LEADER.ENGINEERING
>
>     pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>     Phone: *+1 613 254 3336*
>
>
>
>     *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>     Canada
>     Cisco.com <http://www.cisco.com/global/CA/>
>
>
>
>     Image removed by sender.Think before you print.
>
>     This email may contain confidential and privileged material for
>     the sole use of the intended recipient. Any review, use,
>     distribution or disclosure by others is strictly prohibited. If
>     you are not the intended recipient (or authorized to receive for
>     the recipient), please contact the sender by reply email and
>     delete all copies of this message.
>
>     Please click here
>     <http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>     for Company Registration Information.
>
>
>     *From: *Haoweiguo <haoweiguo@huawei.com <mailto:haoweiguo@huawei.com>=
>
>     *Date: *Thursday, March 26, 2015 at 10:24 AM
>     *To: *Patrice Brissette <pbrisset@cisco.com
>     <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>     <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>     <mailto:sajassi@cisco.com>>, "bess@ietf.org
>     <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>     *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES based on
>     Paxos algorithm
>
>         Hi Patrice,
>
>         Up to reception timer traffic disruption in transient phase is
>         one of the issues.
>
>         Thanks,
>
>         weiguo
>
>         -----------------------------------------------------------------=
-------
>
>         *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>         <mailto:pbrisset@cisco.com>]
>         *Sent:* Thursday, March 26, 2015 20:54
>         *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>         bess@ietf.org <mailto:bess@ietf.org>
>         *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>         based on Paxos algorithm
>
>         Weiguo,
>
>         You mention "But if your draft have not solved all issues=94,
>
>         Can you explain what Satya=92s draft is not solving?
>
>         Regards,
>
>         Patrice
>
>         Image removed by sender.
>
>         *Patrice Brissette*
>         TECHNICAL LEADER.ENGINEERING
>
>         pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>         Phone: *+1 613 254 3336*
>
>
>
>         *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>         Canada
>         Cisco.com <http://www.cisco.com/global/CA/>
>
>
>
>         Image removed by sender.Think before you print.
>
>         This email may contain confidential and privileged material
>         for the sole use of the intended recipient. Any review, use,
>         distribution or disclosure by others is strictly prohibited.
>         If you are not the intended recipient (or authorized to
>         receive for the recipient), please contact the sender by reply
>         email and delete all copies of this message.
>
>         Please click here
>         <http://www.cisco.com/web/about/doing_business/legal/cri/index.ht=
ml>
>         for Company Registration Information.
>
>
>         *From: *Haoweiguo <haoweiguo@huawei.com
>         <mailto:haoweiguo@huawei.com>>
>         *Date: *Wednesday, March 25, 2015 at 10:38 PM
>         *To: *John E Drake <jdrake@juniper.net
>         <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>         <mailto:sajassi@cisco.com>>, "bess@ietf.org
>         <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>         *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>         based on Paxos algorithm
>
>             Hi John,
>
>              Firstly i think EVPN community should reach consensus on
>             the issues of current DF election mechanism. All these
>             issues should be resolved in a single new DF election
>             draft,rather than in multiple separate drafts. If your
>             draft can solve all these issues and stable, i have no
>             question for its progressing. But if your draft have not
>             solved all issues, i think it had better combine with
>             other drafts to provide a comprehensive solution. I think
>             the issues listed in draft-rabadan-bess-evpn-ac-df-01 and
>             draft-hao-bess-evpn-df-handshaking-00 is valid, it should
>             be resolved. So i think although your new Hash algorithm
>             for DF election is good, it only includes partial
>             enhancements, maybe it still needs some time for consensus.
>
>             Thanks,
>
>             weiguo
>
>             -------------------------------------------------------------=
-----------
>
>             *From:*John E Drake [jdrake@juniper.net
>             <mailto:jdrake@juniper.net>]
>             *Sent:* Thursday, March 26, 2015 7:16
>             *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>             <mailto:bess@ietf.org>
>             *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES
>             based on Paxos algorithm
>
>             Weiguo,
>
>             Your proposal introduces a control plane processing load
>             that is O(#EVIs * PEs) per DF election and given that
>             there can be 4K EVIs per ES, this looks like a
>             **substantial** load. Furthermore,
>
>             you can=92t  use the ES route to co-ordinate DF election
>             because you would need to carry your new extended
>             community for each EVI and they would not all fit.  You
>             also can=92t use the Per EVI Ethernet AD route because that
>             is processed by all PEs in the EVI.
>
>             I think that from a practical perspective the new DF
>             election proposed in Satya=92s draft is sufficiently stable
>             that it renders your draft moot, even if it could be made
>             to work.
>
>             Yours Irrespectively,
>
>             John
>
>             *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf Of
>             *Haoweiguo
>             *Sent:* Wednesday, March 25, 2015 5:34 PM
>             *To:* Ali Sajassi (sajassi); bess@ietf.org
>             <mailto:bess@ietf.org>
>             *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>             based on Paxos algorithm
>
>             Hi Ali,
>
>             Thanks for your information. I scanned through this draft,
>             it really introduces inter-chassis message for DF election
>             handshaking, the requirements in this draft is to
>             eliminate transiet Loop and traffic duplication. Current
>             EVPN DF election mechanism already eliminated loop and
>             traffic duplication by configuring long reception timer on
>             each multi-homed PE, but up to reception timer traffic
>             disruption issue still exist. EVPN for DCI is an important
>             use case for EVPN, up to reception timer traffic
>             disruption can't be tolerated for service providers, it
>             should be improved.
>
>             Also for accuracy, i think handshaking state machine on
>             each multi-homed PE is also needed. From solution
>             perspective, in my draft, no inter-chassis message is
>             introduced, only one new extended community is introduced,
>             i think the process is comparatively simple than your
>             following draft.
>
>             Current EVPN DF election has some drawbacks, so there are
>             three new drafts about DF election emerged. I think BESS
>             WG can consider these three drafts in global view, a
>             single,comprehensive new DF election draft is hoped.
>
>             thanks.
>
>             Thanks,
>
>             weiguo
>
>             -------------------------------------------------------------=
-----------
>
>             *From:*BESS [bess-bounces@ietf.org
>             <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajassi
>             (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>]
>             *Sent:* Thursday, March 26, 2015 0:21
>             *To:* bess@ietf.org <mailto:bess@ietf.org>
>             *Subject:* [bess] Handshaking among PEs in an EVPN ES
>             based on Paxos algorithm
>
>             FYI- First published July 4, 2011
>
>             https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-seg=
ment-route/
>
>             -Ali
>
>
>
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess













___________________________________________________________________________=
______________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confiden=
tielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu=
 ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages el=
ectroniques etant susceptibles d'alteration,
Orange decline toute responsabilite si ce message a ete altere, deforme ou =
falsifie. Merci.

This message and its attachments may contain confidential or privileged inf=
ormation that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and dele=
te this message and its attachments.
As emails may be altered, Orange is not liable for messages that have been =
modified, changed or falsified.
Thank you.

_______________________________________________
BESS mailing list
BESS@ietf.org
https://www.ietf.org/mailman/listinfo/bess=


From nobody Thu Mar 26 20:59:30 2015
Return-Path: <thomas.morin@orange.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 72EA71A3BA0 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 20:59:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FREEMAIL_FROM=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001, UNPARSEABLE_RELAY=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UstnaIsf13LZ for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 20:59:25 -0700 (PDT)
Received: from relais-inet.francetelecom.com (relais-ias91.francetelecom.com [193.251.215.91]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 546541A1EED for <bess@ietf.org>; Thu, 26 Mar 2015 20:59:24 -0700 (PDT)
Received: from omfedm07.si.francetelecom.fr (unknown [xx.xx.xx.3]) by omfedm10.si.francetelecom.fr (ESMTP service) with ESMTP id D2979264320; Fri, 27 Mar 2015 04:59:22 +0100 (CET)
Received: from Exchangemail-eme1.itn.ftgroup (unknown [10.114.1.183]) by omfedm07.si.francetelecom.fr (ESMTP service) with ESMTP id B762A4C056; Fri, 27 Mar 2015 04:59:22 +0100 (CET)
Received: from [10.193.116.37] (10.197.38.6) by PEXCVZYH02.corporate.adroot.infra.ftgroup (10.114.1.183) with Microsoft SMTP Server (TLS) id 14.3.224.2; Fri, 27 Mar 2015 04:59:21 +0100
Message-ID: <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com>
Date: Thu, 26 Mar 2015 22:59:18 -0500
From: <thomas.morin@orange.com>
Organization: Orange
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: Haoweiguo <haoweiguo@huawei.com>, "bess@ietf.org" <bess@ietf.org>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com>, <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit
X-Originating-IP: [10.197.38.6]
X-PMX-Version: 6.2.1.2478543, Antispam-Engine: 2.7.2.2107409, Antispam-Data: 2015.3.26.230622
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/5BsLy2yVS7DfU6d5tNA0gQphVC4>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 03:59:28 -0000

Hi Weiguo,

2015-03-26, Haoweiguo:
 > Thomas:
> > - RFC7432 may have transient periods where the DF election state is
> > not yet synchronized between the two peers:
 >
> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic loop

Note well that I didn't write that there can be transient _loops_ .
I tried to capture the exchange you had all, and what I gather is that 
the split-horizon procedure _prevents_loops_, independently of any 
transient period where DF state is not synchronized and which may lead 
to transient _duplicates_.

-Thomas

>
> 26/03/2015 20:05, John E Drake :
>>
>> Weiguo,
>>
>> I guess I wasn’t clear.  I think you draft, for the reasons I have
>> detailed, is a non-solution to a non-problem with tremendous control
>> plane cost.
>>
>> Yours Irrespectively,
>>
>> John
>>
>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>> *Sent:* Thursday, March 26, 2015 6:17 PM
>> *To:* John E Drake; Patrice Brissette (pbrisset)
>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> Pls see below.
>>
>> Thanks,
>>
>> weiguo
>>
>> ------------------------------------------------------------------------
>>
>> *From:*John E Drake [jdrake@juniper.net]
>> *Sent:* Friday, March 27, 2015 6:00
>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> To recap,
>>
>> We have established that your proposal is untenable because of its
>> control plane load.
>>
>> We have established that your proposal is based upon a flawed
>> understanding of the DF election in RFC 7432.
>>
>> [weiguo]: In ethernet world, traffic loop is serious than short timer
>> traffic disruption. If you want to implement  transiet traffic loop
>> process, i will modify my draft to solve your issue.
>>
>> If i am the developer, i will prefer short timer traffic disruption
>> based on current EVPN protocol.
>>
>> What you are now arguing is that your draft prevents two or more PEs
>> from being DF simultaneously. This is clearly nonsense.
>>
>> [weiguo]: I will modify the draft problem statements, and use the same
>> handshaking solution to solve it.
>>
>> Furthermore, we have established that having two or more DFs for what
>> even you admit is a brief transient leads to duplicate traffic, which
>> is acceptable, but not loops, your assertion to the contrary.
>>
>> [weiguo]: It is transient loop and traffic duplication issue.
>>
>> Yours Irrespectively,
>>
>> John
>>
>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>> *Sent:* Thursday, March 26, 2015 5:37 PM
>> *To:* John E Drake; Patrice Brissette (pbrisset)
>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> John,
>>
>> As your understanding of the EVPN draft, the DF election mechanism has
>> more serious side effect, it will have short time traffic loop,i.e.,
>> dual DF PEs will exist for a short time. I think dual DF PEs is
>> absolutely not tolerated, because native ethernet header has no TTL,
>> up to several hundred ms traffic loop normally not tolerated in
>> commertial networks.
>>
>> As your understanding, the PEs should do as following:
>>
>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have
>> rarely deployment.
>>
>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins
>> ESI and trigger DF re-election. When reception timer expires:
>>
>> PE1 upgrades to DF PE.
>>
>> After reception timer+ ES route transmission timer:
>>
>> PE2 downloads to non-DF PE.
>>
>> So in timer clock sync case, dual DF PEs will exist at least
>> transmission timer.
>>
>> If NTP is used for timer sync, because it has bad accuracy, dual DF
>> PEs will exist more longer timer.
>>
>> So as your understanding for DF election, the drawback is more clear.
>>
>> Thanks,
>>
>> weiguo
>>
>> ------------------------------------------------------------------------
>>
>> *From:*John E Drake [jdrake@juniper.net]
>> *Sent:* Friday, March 27, 2015 4:41
>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> Weiguo,
>>
>> We have already established that your proposal is untenable because of
>> its control plane load.
>>
>> What we are now discussing is that your proposal is based upon a
>> misunderstanding of the algorithm in RFC 7432.  You are assuming that
>> PE1 will advertise an ES route and then wait for the configured
>> interval before performing the DF election while PE2 and PE3 will
>> perform the DF election as soon as they receive the ES route from PE1.
>> This is not what RFC 7432 says.
>>
>> Rather, what is says is that the advertisement of the ES route by PE1
>> and its receipt by PE2 and PE3 causes all three PEs to start the
>> configured interval timer  - “3. When the timer expires, each PE
>> builds an ordered list of the IP addresses of all the PE nodes
>> connected to the Ethernet segment (including itself), in increasing
>> numeric value.”
>>
>> Yours Irrespectively,
>>
>> John
>>
>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>> *Sent:* Thursday, March 26, 2015 3:26 PM
>> *To:* John E Drake; Patrice Brissette (pbrisset)
>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> Pls read my detail replies to Satya. If you still can't catch it, pls
>> read my draft and EVPN base protocol,  thanks
>>
>> weiguo
>>
>> ------------------------------------------------------------------------
>>
>> *From:*John E Drake [jdrake@juniper.net]
>> *Sent:* Friday, March 27, 2015 1:28
>> *To:* Patrice Brissette (pbrisset)
>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>> <mailto:bess@ietf.org>
>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> I think Patrice is correct.  Your proposal doesn't solve the problem
>> and it does so at huge cost.
>>
>> Sent from my iPhone
>>
>>
>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>
>>      Weiguo,
>>
>>      I’m not sure I’m following here.
>>
>>      Don’t you have the same issue with your handshaking mechanism?
>>
>>      If you don’t know your peer, how can you handshake?
>>
>>      Regards,
>>
>>      Patrice
>>
>>      Image removed by sender.
>>
>>      *Patrice Brissette*
>>      TECHNICAL LEADER.ENGINEERING
>>
>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>      Phone: *+1 613 254 3336*
>>
>>
>>
>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>      Canada
>>      Cisco.com <http://www.cisco.com/global/CA/>
>>
>>
>>
>>      Image removed by sender.Think before you print.
>>
>>      This email may contain confidential and privileged material for
>>      the sole use of the intended recipient. Any review, use,
>>      distribution or disclosure by others is strictly prohibited. If
>>      you are not the intended recipient (or authorized to receive for
>>      the recipient), please contact the sender by reply email and
>>      delete all copies of this message.
>>
>>      Please click here
>>      <http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>      for Company Registration Information.
>>
>>
>>      *From: *Haoweiguo <haoweiguo@huawei.com <mailto:haoweiguo@huawei.com>>
>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
>>      *To: *Patrice Brissette <pbrisset@cisco.com
>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES based on
>>      Paxos algorithm
>>
>>          Hi Patrice,
>>
>>          Up to reception timer traffic disruption in transient phase is
>>          one of the issues.
>>
>>          Thanks,
>>
>>          weiguo
>>
>>          ------------------------------------------------------------------------
>>
>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>          <mailto:pbrisset@cisco.com>]
>>          *Sent:* Thursday, March 26, 2015 20:54
>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>          bess@ietf.org <mailto:bess@ietf.org>
>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>          based on Paxos algorithm
>>
>>          Weiguo,
>>
>>          You mention "But if your draft have not solved all issues”,
>>
>>          Can you explain what Satya’s draft is not solving?
>>
>>          Regards,
>>
>>          Patrice
>>
>>          Image removed by sender.
>>
>>          *Patrice Brissette*
>>          TECHNICAL LEADER.ENGINEERING
>>
>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>          Phone: *+1 613 254 3336*
>>
>>
>>
>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>          Canada
>>          Cisco.com <http://www.cisco.com/global/CA/>
>>
>>
>>
>>          Image removed by sender.Think before you print.
>>
>>          This email may contain confidential and privileged material
>>          for the sole use of the intended recipient. Any review, use,
>>          distribution or disclosure by others is strictly prohibited.
>>          If you are not the intended recipient (or authorized to
>>          receive for the recipient), please contact the sender by reply
>>          email and delete all copies of this message.
>>
>>          Please click here
>>          <http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>          for Company Registration Information.
>>
>>
>>          *From: *Haoweiguo <haoweiguo@huawei.com
>>          <mailto:haoweiguo@huawei.com>>
>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>          *To: *John E Drake <jdrake@juniper.net
>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>          <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>          based on Paxos algorithm
>>
>>              Hi John,
>>
>>               Firstly i think EVPN community should reach consensus on
>>              the issues of current DF election mechanism. All these
>>              issues should be resolved in a single new DF election
>>              draft,rather than in multiple separate drafts. If your
>>              draft can solve all these issues and stable, i have no
>>              question for its progressing. But if your draft have not
>>              solved all issues, i think it had better combine with
>>              other drafts to provide a comprehensive solution. I think
>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01 and
>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it should
>>              be resolved. So i think although your new Hash algorithm
>>              for DF election is good, it only includes partial
>>              enhancements, maybe it still needs some time for consensus.
>>
>>              Thanks,
>>
>>              weiguo
>>
>>              ------------------------------------------------------------------------
>>
>>              *From:*John E Drake [jdrake@juniper.net
>>              <mailto:jdrake@juniper.net>]
>>              *Sent:* Thursday, March 26, 2015 7:16
>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>              <mailto:bess@ietf.org>
>>              *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES
>>              based on Paxos algorithm
>>
>>              Weiguo,
>>
>>              Your proposal introduces a control plane processing load
>>              that is O(#EVIs * PEs) per DF election and given that
>>              there can be 4K EVIs per ES, this looks like a
>>              **substantial** load. Furthermore,
>>
>>              you can’t  use the ES route to co-ordinate DF election
>>              because you would need to carry your new extended
>>              community for each EVI and they would not all fit.  You
>>              also can’t use the Per EVI Ethernet AD route because that
>>              is processed by all PEs in the EVI.
>>
>>              I think that from a practical perspective the new DF
>>              election proposed in Satya’s draft is sufficiently stable
>>              that it renders your draft moot, even if it could be made
>>              to work.
>>
>>              Yours Irrespectively,
>>
>>              John
>>
>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf Of
>>              *Haoweiguo
>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
>>              <mailto:bess@ietf.org>
>>              *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>              based on Paxos algorithm
>>
>>              Hi Ali,
>>
>>              Thanks for your information. I scanned through this draft,
>>              it really introduces inter-chassis message for DF election
>>              handshaking, the requirements in this draft is to
>>              eliminate transiet Loop and traffic duplication. Current
>>              EVPN DF election mechanism already eliminated loop and
>>              traffic duplication by configuring long reception timer on
>>              each multi-homed PE, but up to reception timer traffic
>>              disruption issue still exist. EVPN for DCI is an important
>>              use case for EVPN, up to reception timer traffic
>>              disruption can't be tolerated for service providers, it
>>              should be improved.
>>
>>              Also for accuracy, i think handshaking state machine on
>>              each multi-homed PE is also needed. From solution
>>              perspective, in my draft, no inter-chassis message is
>>              introduced, only one new extended community is introduced,
>>              i think the process is comparatively simple than your
>>              following draft.
>>
>>              Current EVPN DF election has some drawbacks, so there are
>>              three new drafts about DF election emerged. I think BESS
>>              WG can consider these three drafts in global view, a
>>              single,comprehensive new DF election draft is hoped.
>>
>>              thanks.
>>
>>              Thanks,
>>
>>              weiguo
>>
>>              ------------------------------------------------------------------------
>>
>>              *From:*BESS [bess-bounces@ietf.org
>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajassi
>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>]
>>              *Sent:* Thursday, March 26, 2015 0:21
>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>              based on Paxos algorithm
>>
>>              FYI- First published July 4, 2011
>>
>>              https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/
>>
>>              -Ali
>>
>>
>>
>> _______________________________________________
>> BESS mailing list
>> BESS@ietf.org
>> https://www.ietf.org/mailman/listinfo/bess
>
>
>
>
>
>
>
>
>
>
>
>
>
> _________________________________________________________________________________________________________________________
>
> Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc
> pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler
> a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration,
> Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci.
>
> This message and its attachments may contain confidential or privileged information that may be protected by law;
> they should not be distributed, used or copied without authorisation.
> If you have received this email in error, please notify the sender and delete this message and its attachments.
> As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified.
> Thank you.
>
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess
>


_________________________________________________________________________________________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration,
Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci.

This message and its attachments may contain confidential or privileged information that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and delete this message and its attachments.
As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified.
Thank you.


From nobody Thu Mar 26 21:08:50 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D9E6C1A1A8F for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 21:08:48 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.611
X-Spam-Level: 
X-Spam-Status: No, score=-3.611 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id C8mUHssuuFAN for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 21:08:45 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B5F051A0058 for <bess@ietf.org>; Thu, 26 Mar 2015 21:08:44 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml405-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT34820; Fri, 27 Mar 2015 04:08:43 +0000 (GMT)
Received: from nkgeml407-hub.china.huawei.com (10.98.56.38) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 04:08:42 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml407-hub.china.huawei.com ([10.98.56.38]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 12:08:38 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr
Date: Fri, 27 Mar 2015 04:08:38 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com>, <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com>, <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com>
In-Reply-To: <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.153.94]
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/-8mwM2TbD1s1GEpBBW0I6YqD9i0>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 04:08:49 -0000

Hi Thomas,
I know you are neutral, sorry for my reply vulnerable to be miunderstood.

Thanks,
weiguo

________________________________________
From: thomas.morin@orange.com [thomas.morin@orange.com]
Sent: Friday, March 27, 2015 11:59
To: Haoweiguo; bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Hi Weiguo,

2015-03-26, Haoweiguo:
 > Thomas:
> > - RFC7432 may have transient periods where the DF election state is
> > not yet synchronized between the two peers:
 >
> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic loop

Note well that I didn't write that there can be transient _loops_ .
I tried to capture the exchange you had all, and what I gather is that
the split-horizon procedure _prevents_loops_, independently of any
transient period where DF state is not synchronized and which may lead
to transient _duplicates_.

-Thomas

>
> 26/03/2015 20:05, John E Drake :
>>
>> Weiguo,
>>
>> I guess I wasn=92t clear.  I think you draft, for the reasons I have
>> detailed, is a non-solution to a non-problem with tremendous control
>> plane cost.
>>
>> Yours Irrespectively,
>>
>> John
>>
>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>> *Sent:* Thursday, March 26, 2015 6:17 PM
>> *To:* John E Drake; Patrice Brissette (pbrisset)
>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> Pls see below.
>>
>> Thanks,
>>
>> weiguo
>>
>> ------------------------------------------------------------------------
>>
>> *From:*John E Drake [jdrake@juniper.net]
>> *Sent:* Friday, March 27, 2015 6:00
>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> To recap,
>>
>> We have established that your proposal is untenable because of its
>> control plane load.
>>
>> We have established that your proposal is based upon a flawed
>> understanding of the DF election in RFC 7432.
>>
>> [weiguo]: In ethernet world, traffic loop is serious than short timer
>> traffic disruption. If you want to implement  transiet traffic loop
>> process, i will modify my draft to solve your issue.
>>
>> If i am the developer, i will prefer short timer traffic disruption
>> based on current EVPN protocol.
>>
>> What you are now arguing is that your draft prevents two or more PEs
>> from being DF simultaneously. This is clearly nonsense.
>>
>> [weiguo]: I will modify the draft problem statements, and use the same
>> handshaking solution to solve it.
>>
>> Furthermore, we have established that having two or more DFs for what
>> even you admit is a brief transient leads to duplicate traffic, which
>> is acceptable, but not loops, your assertion to the contrary.
>>
>> [weiguo]: It is transient loop and traffic duplication issue.
>>
>> Yours Irrespectively,
>>
>> John
>>
>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>> *Sent:* Thursday, March 26, 2015 5:37 PM
>> *To:* John E Drake; Patrice Brissette (pbrisset)
>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> John,
>>
>> As your understanding of the EVPN draft, the DF election mechanism has
>> more serious side effect, it will have short time traffic loop,i.e.,
>> dual DF PEs will exist for a short time. I think dual DF PEs is
>> absolutely not tolerated, because native ethernet header has no TTL,
>> up to several hundred ms traffic loop normally not tolerated in
>> commertial networks.
>>
>> As your understanding, the PEs should do as following:
>>
>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have
>> rarely deployment.
>>
>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins
>> ESI and trigger DF re-election. When reception timer expires:
>>
>> PE1 upgrades to DF PE.
>>
>> After reception timer+ ES route transmission timer:
>>
>> PE2 downloads to non-DF PE.
>>
>> So in timer clock sync case, dual DF PEs will exist at least
>> transmission timer.
>>
>> If NTP is used for timer sync, because it has bad accuracy, dual DF
>> PEs will exist more longer timer.
>>
>> So as your understanding for DF election, the drawback is more clear.
>>
>> Thanks,
>>
>> weiguo
>>
>> ------------------------------------------------------------------------
>>
>> *From:*John E Drake [jdrake@juniper.net]
>> *Sent:* Friday, March 27, 2015 4:41
>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> Weiguo,
>>
>> We have already established that your proposal is untenable because of
>> its control plane load.
>>
>> What we are now discussing is that your proposal is based upon a
>> misunderstanding of the algorithm in RFC 7432.  You are assuming that
>> PE1 will advertise an ES route and then wait for the configured
>> interval before performing the DF election while PE2 and PE3 will
>> perform the DF election as soon as they receive the ES route from PE1.
>> This is not what RFC 7432 says.
>>
>> Rather, what is says is that the advertisement of the ES route by PE1
>> and its receipt by PE2 and PE3 causes all three PEs to start the
>> configured interval timer  - =933. When the timer expires, each PE
>> builds an ordered list of the IP addresses of all the PE nodes
>> connected to the Ethernet segment (including itself), in increasing
>> numeric value.=94
>>
>> Yours Irrespectively,
>>
>> John
>>
>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>> *Sent:* Thursday, March 26, 2015 3:26 PM
>> *To:* John E Drake; Patrice Brissette (pbrisset)
>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> Pls read my detail replies to Satya. If you still can't catch it, pls
>> read my draft and EVPN base protocol,  thanks
>>
>> weiguo
>>
>> ------------------------------------------------------------------------
>>
>> *From:*John E Drake [jdrake@juniper.net]
>> *Sent:* Friday, March 27, 2015 1:28
>> *To:* Patrice Brissette (pbrisset)
>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>> <mailto:bess@ietf.org>
>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>> Paxos algorithm
>>
>> I think Patrice is correct.  Your proposal doesn't solve the problem
>> and it does so at huge cost.
>>
>> Sent from my iPhone
>>
>>
>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>
>>      Weiguo,
>>
>>      I=92m not sure I=92m following here.
>>
>>      Don=92t you have the same issue with your handshaking mechanism?
>>
>>      If you don=92t know your peer, how can you handshake?
>>
>>      Regards,
>>
>>      Patrice
>>
>>      Image removed by sender.
>>
>>      *Patrice Brissette*
>>      TECHNICAL LEADER.ENGINEERING
>>
>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>      Phone: *+1 613 254 3336*
>>
>>
>>
>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>      Canada
>>      Cisco.com <http://www.cisco.com/global/CA/>
>>
>>
>>
>>      Image removed by sender.Think before you print.
>>
>>      This email may contain confidential and privileged material for
>>      the sole use of the intended recipient. Any review, use,
>>      distribution or disclosure by others is strictly prohibited. If
>>      you are not the intended recipient (or authorized to receive for
>>      the recipient), please contact the sender by reply email and
>>      delete all copies of this message.
>>
>>      Please click here
>>      <http://www.cisco.com/web/about/doing_business/legal/cri/index.html=
>
>>      for Company Registration Information.
>>
>>
>>      *From: *Haoweiguo <haoweiguo@huawei.com <mailto:haoweiguo@huawei.co=
m>>
>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
>>      *To: *Patrice Brissette <pbrisset@cisco.com
>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES based on
>>      Paxos algorithm
>>
>>          Hi Patrice,
>>
>>          Up to reception timer traffic disruption in transient phase is
>>          one of the issues.
>>
>>          Thanks,
>>
>>          weiguo
>>
>>          ---------------------------------------------------------------=
---------
>>
>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>          <mailto:pbrisset@cisco.com>]
>>          *Sent:* Thursday, March 26, 2015 20:54
>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>          bess@ietf.org <mailto:bess@ietf.org>
>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>          based on Paxos algorithm
>>
>>          Weiguo,
>>
>>          You mention "But if your draft have not solved all issues=94,
>>
>>          Can you explain what Satya=92s draft is not solving?
>>
>>          Regards,
>>
>>          Patrice
>>
>>          Image removed by sender.
>>
>>          *Patrice Brissette*
>>          TECHNICAL LEADER.ENGINEERING
>>
>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>          Phone: *+1 613 254 3336*
>>
>>
>>
>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>          Canada
>>          Cisco.com <http://www.cisco.com/global/CA/>
>>
>>
>>
>>          Image removed by sender.Think before you print.
>>
>>          This email may contain confidential and privileged material
>>          for the sole use of the intended recipient. Any review, use,
>>          distribution or disclosure by others is strictly prohibited.
>>          If you are not the intended recipient (or authorized to
>>          receive for the recipient), please contact the sender by reply
>>          email and delete all copies of this message.
>>
>>          Please click here
>>          <http://www.cisco.com/web/about/doing_business/legal/cri/index.=
html>
>>          for Company Registration Information.
>>
>>
>>          *From: *Haoweiguo <haoweiguo@huawei.com
>>          <mailto:haoweiguo@huawei.com>>
>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>          *To: *John E Drake <jdrake@juniper.net
>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>          <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>          based on Paxos algorithm
>>
>>              Hi John,
>>
>>               Firstly i think EVPN community should reach consensus on
>>              the issues of current DF election mechanism. All these
>>              issues should be resolved in a single new DF election
>>              draft,rather than in multiple separate drafts. If your
>>              draft can solve all these issues and stable, i have no
>>              question for its progressing. But if your draft have not
>>              solved all issues, i think it had better combine with
>>              other drafts to provide a comprehensive solution. I think
>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01 and
>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it should
>>              be resolved. So i think although your new Hash algorithm
>>              for DF election is good, it only includes partial
>>              enhancements, maybe it still needs some time for consensus.
>>
>>              Thanks,
>>
>>              weiguo
>>
>>              -----------------------------------------------------------=
-------------
>>
>>              *From:*John E Drake [jdrake@juniper.net
>>              <mailto:jdrake@juniper.net>]
>>              *Sent:* Thursday, March 26, 2015 7:16
>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>              <mailto:bess@ietf.org>
>>              *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES
>>              based on Paxos algorithm
>>
>>              Weiguo,
>>
>>              Your proposal introduces a control plane processing load
>>              that is O(#EVIs * PEs) per DF election and given that
>>              there can be 4K EVIs per ES, this looks like a
>>              **substantial** load. Furthermore,
>>
>>              you can=92t  use the ES route to co-ordinate DF election
>>              because you would need to carry your new extended
>>              community for each EVI and they would not all fit.  You
>>              also can=92t use the Per EVI Ethernet AD route because that
>>              is processed by all PEs in the EVI.
>>
>>              I think that from a practical perspective the new DF
>>              election proposed in Satya=92s draft is sufficiently stable
>>              that it renders your draft moot, even if it could be made
>>              to work.
>>
>>              Yours Irrespectively,
>>
>>              John
>>
>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf Of
>>              *Haoweiguo
>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
>>              <mailto:bess@ietf.org>
>>              *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>              based on Paxos algorithm
>>
>>              Hi Ali,
>>
>>              Thanks for your information. I scanned through this draft,
>>              it really introduces inter-chassis message for DF election
>>              handshaking, the requirements in this draft is to
>>              eliminate transiet Loop and traffic duplication. Current
>>              EVPN DF election mechanism already eliminated loop and
>>              traffic duplication by configuring long reception timer on
>>              each multi-homed PE, but up to reception timer traffic
>>              disruption issue still exist. EVPN for DCI is an important
>>              use case for EVPN, up to reception timer traffic
>>              disruption can't be tolerated for service providers, it
>>              should be improved.
>>
>>              Also for accuracy, i think handshaking state machine on
>>              each multi-homed PE is also needed. From solution
>>              perspective, in my draft, no inter-chassis message is
>>              introduced, only one new extended community is introduced,
>>              i think the process is comparatively simple than your
>>              following draft.
>>
>>              Current EVPN DF election has some drawbacks, so there are
>>              three new drafts about DF election emerged. I think BESS
>>              WG can consider these three drafts in global view, a
>>              single,comprehensive new DF election draft is hoped.
>>
>>              thanks.
>>
>>              Thanks,
>>
>>              weiguo
>>
>>              -----------------------------------------------------------=
-------------
>>
>>              *From:*BESS [bess-bounces@ietf.org
>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajassi
>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>]
>>              *Sent:* Thursday, March 26, 2015 0:21
>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>              based on Paxos algorithm
>>
>>              FYI- First published July 4, 2011
>>
>>              https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-s=
egment-route/
>>
>>              -Ali
>>
>>
>>
>> _______________________________________________
>> BESS mailing list
>> BESS@ietf.org
>> https://www.ietf.org/mailman/listinfo/bess
>
>
>
>
>
>
>
>
>
>
>
>
>
> _________________________________________________________________________=
________________________________________________
>
> Ce message et ses pieces jointes peuvent contenir des informations confid=
entielles ou privilegiees et ne doivent donc
> pas etre diffuses, exploites ou copies sans autorisation. Si vous avez re=
cu ce message par erreur, veuillez le signaler
> a l'expediteur et le detruire ainsi que les pieces jointes. Les messages =
electroniques etant susceptibles d'alteration,
> Orange decline toute responsabilite si ce message a ete altere, deforme o=
u falsifie. Merci.
>
> This message and its attachments may contain confidential or privileged i=
nformation that may be protected by law;
> they should not be distributed, used or copied without authorisation.
> If you have received this email in error, please notify the sender and de=
lete this message and its attachments.
> As emails may be altered, Orange is not liable for messages that have bee=
n modified, changed or falsified.
> Thank you.
>
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess
>


___________________________________________________________________________=
______________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confiden=
tielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu=
 ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages el=
ectroniques etant susceptibles d'alteration,
Orange decline toute responsabilite si ce message a ete altere, deforme ou =
falsifie. Merci.

This message and its attachments may contain confidential or privileged inf=
ormation that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and dele=
te this message and its attachments.
As emails may be altered, Orange is not liable for messages that have been =
modified, changed or falsified.
Thank you.=


From nobody Thu Mar 26 21:10:49 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B074B1A0058 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 21:10:48 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id rAovKTYwMr08 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 21:10:44 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5B65D1A1B46 for <bess@ietf.org>; Thu, 26 Mar 2015 21:10:44 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id F0A4D250C8DAE; Fri, 27 Mar 2015 04:10:41 +0000 (GMT)
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2R4Aguw010920 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 05:10:42 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR711WXCHHUB01.zeu.alcatel-lucent.com ([135.239.2.111]) with mapi id 14.03.0195.001; Fri, 27 Mar 2015 05:10:42 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: "thomas.morin@orange.com" <thomas.morin@orange.com>, Haoweiguo <haoweiguo@huawei.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQaCo/Eaq8IGmBqka8wmUCt0qdaZ0vkICAgAAP+ACAAAM7AP//jdIA
Date: Fri, 27 Mar 2015 04:10:41 +0000
Message-ID: <D13A24D9.6C852%jorge.rabadan@alcatel-lucent.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com>
In-Reply-To: <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.39]
Content-Type: text/plain; charset="utf-8"
Content-ID: <B95E47D86CAD8448A0104C08076A3690@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/3Zl-9Of2yUuwFP5CP8m0ft0eEtw>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 04:10:48 -0000
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From nobody Thu Mar 26 21:12:04 2015
Return-Path: <satyamoh@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9C6621A0058 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 21:12:03 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -13.911
X-Spam-Level: 
X-Spam-Status: No, score=-13.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ckt-CyXyKcDa for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 21:12:00 -0700 (PDT)
Received: from rcdn-iport-6.cisco.com (rcdn-iport-6.cisco.com [173.37.86.77]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CB0441A1B46 for <bess@ietf.org>; Thu, 26 Mar 2015 21:11:59 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=20343; q=dns/txt; s=iport; t=1427429519; x=1428639119; h=from:to:subject:date:message-id:references:in-reply-to: content-id:content-transfer-encoding:mime-version; bh=S74ZToFKVrAlaU1kwAcprys0rrlpTM/w3uTctw45RZY=; b=BEJj34ABLaNBTPxQF2sjnVIhVQ4SGRvmWEbU9ioQo4JosNqM3kXE9dQg IoEGuDCgQ/Zi/2BULMWfa7o7LADAXwN+36Q2LJMyAednrnN6gOT3ZIbhe 3ATZrm4d0Gn39t+q+M/Y9ReRkcbtqrXwDgxUPFF+Xnc63kPKgEhZzyfb1 k=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0AfBQAL2BRV/5FdJa1ZA4MGUloExSgMhW4BAgEBAoE/TAEBAQEBAX2EFAEBAQQBAQFjCBkCAgEIEQMBAQEBJwcbDAsUCQgCBAESFAeIFA2zJZhtAQEBAQEBAQEBAQEBAQEBAQEBAQEBFwSLJIQNCAERAQ4PDRYLBwUGC4QcBYYYhEuFbYNvhgCBGzqCdoJaiUGDRyKCAR2BUG8BgQMHFyJ/AQEB
X-IronPort-AV: E=Sophos;i="5.11,476,1422921600"; d="scan'208";a="407073764"
Received: from rcdn-core-9.cisco.com ([173.37.93.145]) by rcdn-iport-6.cisco.com with ESMTP; 27 Mar 2015 04:11:58 +0000
Received: from xhc-rcd-x11.cisco.com (xhc-rcd-x11.cisco.com [173.37.183.85]) by rcdn-core-9.cisco.com (8.14.5/8.14.5) with ESMTP id t2R4Bw96007744 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 04:11:58 GMT
Received: from xmb-aln-x07.cisco.com ([169.254.2.199]) by xhc-rcd-x11.cisco.com ([173.37.183.85]) with mapi id 14.03.0195.001; Thu, 26 Mar 2015 23:11:57 -0500
From: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>
To: Haoweiguo <haoweiguo@huawei.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ1HYE/ZJUJbtWUGUXbn7g+eTDJ0ucdUAgACbTgCAACnuAIAAAsqAgAAf4ACAACCzAIAAFSOAgAAPjACAAAZ+AIAABLkAgAAvJACAAB1OgIAAD/gAgAADOwCAAAKcAP//i5CA
Date: Fri, 27 Mar 2015 04:11:57 +0000
Message-ID: <D13A263C.2CB9E%satyamoh@cisco.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.65.50.166]
Content-Type: text/plain; charset="iso-8859-1"
Content-ID: <4D3B29084C805046955D7C3EAF7DC4A9@emea.cisco.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/c2n2_8kNSfh7UlGTFvhc_i70leI>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 04:12:03 -0000

Thomas,

Thanks for your summary. It captures the essence of the discussion.

One feature of the HRW scheme is that it computes the Active and the
Backup DF upfront.
Whenever the Active PE, say PE1, dies; the Backup PE, PE2, can become the
Active immediately upon receiving the withdraw of the ES route from PE1.
It need not even have to wait for the 3 seconds time.

On the other hand, the 3 second timer will serve as a bulk and batch
mechanism for the case when new PE(s) is(are) coming up, so we don=B9t have
to do a DF election ever so often.

Haoweiguo,  I think the "loop" scenario is speculative.
I think everybody understands that, yes, there may be some transient
duplication of bum traffic; however, there is no "loop" problem to be
solved.

Thanks,
--Satya



On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:

>Hi Thomas,
>I know you are neutral, sorry for my reply vulnerable to be miunderstood.
>
>Thanks,
>weiguo
>
>________________________________________
>From: thomas.morin@orange.com [thomas.morin@orange.com]
>Sent: Friday, March 27, 2015 11:59
>To: Haoweiguo; bess@ietf.org
>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>algorithm
>
>Hi Weiguo,
>
>2015-03-26, Haoweiguo:
> > Thomas:
>> > - RFC7432 may have transient periods where the DF election state is
>> > not yet synchronized between the two peers:
> >
>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic loop
>
>Note well that I didn't write that there can be transient _loops_ .
>I tried to capture the exchange you had all, and what I gather is that
>the split-horizon procedure _prevents_loops_, independently of any
>transient period where DF state is not synchronized and which may lead
>to transient _duplicates_.
>
>-Thomas
>
>>
>> 26/03/2015 20:05, John E Drake :
>>>
>>> Weiguo,
>>>
>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I have
>>> detailed, is a non-solution to a non-problem with tremendous control
>>> plane cost.
>>>
>>> Yours Irrespectively,
>>>
>>> John
>>>
>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>> *Sent:* Thursday, March 26, 2015 6:17 PM
>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> Pls see below.
>>>
>>> Thanks,
>>>
>>> weiguo
>>>
>>>=20
>>>------------------------------------------------------------------------
>>>
>>> *From:*John E Drake [jdrake@juniper.net]
>>> *Sent:* Friday, March 27, 2015 6:00
>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> To recap,
>>>
>>> We have established that your proposal is untenable because of its
>>> control plane load.
>>>
>>> We have established that your proposal is based upon a flawed
>>> understanding of the DF election in RFC 7432.
>>>
>>> [weiguo]: In ethernet world, traffic loop is serious than short timer
>>> traffic disruption. If you want to implement  transiet traffic loop
>>> process, i will modify my draft to solve your issue.
>>>
>>> If i am the developer, i will prefer short timer traffic disruption
>>> based on current EVPN protocol.
>>>
>>> What you are now arguing is that your draft prevents two or more PEs
>>> from being DF simultaneously. This is clearly nonsense.
>>>
>>> [weiguo]: I will modify the draft problem statements, and use the same
>>> handshaking solution to solve it.
>>>
>>> Furthermore, we have established that having two or more DFs for what
>>> even you admit is a brief transient leads to duplicate traffic, which
>>> is acceptable, but not loops, your assertion to the contrary.
>>>
>>> [weiguo]: It is transient loop and traffic duplication issue.
>>>
>>> Yours Irrespectively,
>>>
>>> John
>>>
>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>> *Sent:* Thursday, March 26, 2015 5:37 PM
>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> John,
>>>
>>> As your understanding of the EVPN draft, the DF election mechanism has
>>> more serious side effect, it will have short time traffic loop,i.e.,
>>> dual DF PEs will exist for a short time. I think dual DF PEs is
>>> absolutely not tolerated, because native ethernet header has no TTL,
>>> up to several hundred ms traffic loop normally not tolerated in
>>> commertial networks.
>>>
>>> As your understanding, the PEs should do as following:
>>>
>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have
>>> rarely deployment.
>>>
>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins
>>> ESI and trigger DF re-election. When reception timer expires:
>>>
>>> PE1 upgrades to DF PE.
>>>
>>> After reception timer+ ES route transmission timer:
>>>
>>> PE2 downloads to non-DF PE.
>>>
>>> So in timer clock sync case, dual DF PEs will exist at least
>>> transmission timer.
>>>
>>> If NTP is used for timer sync, because it has bad accuracy, dual DF
>>> PEs will exist more longer timer.
>>>
>>> So as your understanding for DF election, the drawback is more clear.
>>>
>>> Thanks,
>>>
>>> weiguo
>>>
>>>=20
>>>------------------------------------------------------------------------
>>>
>>> *From:*John E Drake [jdrake@juniper.net]
>>> *Sent:* Friday, March 27, 2015 4:41
>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> Weiguo,
>>>
>>> We have already established that your proposal is untenable because of
>>> its control plane load.
>>>
>>> What we are now discussing is that your proposal is based upon a
>>> misunderstanding of the algorithm in RFC 7432.  You are assuming that
>>> PE1 will advertise an ES route and then wait for the configured
>>> interval before performing the DF election while PE2 and PE3 will
>>> perform the DF election as soon as they receive the ES route from PE1.
>>> This is not what RFC 7432 says.
>>>
>>> Rather, what is says is that the advertisement of the ES route by PE1
>>> and its receipt by PE2 and PE3 causes all three PEs to start the
>>> configured interval timer  - =B33. When the timer expires, each PE
>>> builds an ordered list of the IP addresses of all the PE nodes
>>> connected to the Ethernet segment (including itself), in increasing
>>> numeric value.=B2
>>>
>>> Yours Irrespectively,
>>>
>>> John
>>>
>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>> *Sent:* Thursday, March 26, 2015 3:26 PM
>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> Pls read my detail replies to Satya. If you still can't catch it, pls
>>> read my draft and EVPN base protocol,  thanks
>>>
>>> weiguo
>>>
>>>=20
>>>------------------------------------------------------------------------
>>>
>>> *From:*John E Drake [jdrake@juniper.net]
>>> *Sent:* Friday, March 27, 2015 1:28
>>> *To:* Patrice Brissette (pbrisset)
>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>> <mailto:bess@ietf.org>
>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> I think Patrice is correct.  Your proposal doesn't solve the problem
>>> and it does so at huge cost.
>>>
>>> Sent from my iPhone
>>>
>>>
>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>>
>>>      Weiguo,
>>>
>>>      I=B9m not sure I=B9m following here.
>>>
>>>      Don=B9t you have the same issue with your handshaking mechanism?
>>>
>>>      If you don=B9t know your peer, how can you handshake?
>>>
>>>      Regards,
>>>
>>>      Patrice
>>>
>>>      Image removed by sender.
>>>
>>>      *Patrice Brissette*
>>>      TECHNICAL LEADER.ENGINEERING
>>>
>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>      Phone: *+1 613 254 3336*
>>>
>>>
>>>
>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>      Canada
>>>      Cisco.com <http://www.cisco.com/global/CA/>
>>>
>>>
>>>
>>>      Image removed by sender.Think before you print.
>>>
>>>      This email may contain confidential and privileged material for
>>>      the sole use of the intended recipient. Any review, use,
>>>      distribution or disclosure by others is strictly prohibited. If
>>>      you are not the intended recipient (or authorized to receive for
>>>      the recipient), please contact the sender by reply email and
>>>      delete all copies of this message.
>>>
>>>      Please click here
>>>     =20
>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>      for Company Registration Information.
>>>
>>>
>>>      *From: *Haoweiguo <haoweiguo@huawei.com
>>><mailto:haoweiguo@huawei.com>>
>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
>>>      *To: *Patrice Brissette <pbrisset@cisco.com
>>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>      Paxos algorithm
>>>
>>>          Hi Patrice,
>>>
>>>          Up to reception timer traffic disruption in transient phase is
>>>          one of the issues.
>>>
>>>          Thanks,
>>>
>>>          weiguo
>>>
>>>         =20
>>>------------------------------------------------------------------------
>>>
>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>>          <mailto:pbrisset@cisco.com>]
>>>          *Sent:* Thursday, March 26, 2015 20:54
>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>>          bess@ietf.org <mailto:bess@ietf.org>
>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>          based on Paxos algorithm
>>>
>>>          Weiguo,
>>>
>>>          You mention "But if your draft have not solved all issues=B2,
>>>
>>>          Can you explain what Satya=B9s draft is not solving?
>>>
>>>          Regards,
>>>
>>>          Patrice
>>>
>>>          Image removed by sender.
>>>
>>>          *Patrice Brissette*
>>>          TECHNICAL LEADER.ENGINEERING
>>>
>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>          Phone: *+1 613 254 3336*
>>>
>>>
>>>
>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>          Canada
>>>          Cisco.com <http://www.cisco.com/global/CA/>
>>>
>>>
>>>
>>>          Image removed by sender.Think before you print.
>>>
>>>          This email may contain confidential and privileged material
>>>          for the sole use of the intended recipient. Any review, use,
>>>          distribution or disclosure by others is strictly prohibited.
>>>          If you are not the intended recipient (or authorized to
>>>          receive for the recipient), please contact the sender by reply
>>>          email and delete all copies of this message.
>>>
>>>          Please click here
>>>         =20
>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>          for Company Registration Information.
>>>
>>>
>>>          *From: *Haoweiguo <haoweiguo@huawei.com
>>>          <mailto:haoweiguo@huawei.com>>
>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>>          *To: *John E Drake <jdrake@juniper.net
>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>          <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>>          based on Paxos algorithm
>>>
>>>              Hi John,
>>>
>>>               Firstly i think EVPN community should reach consensus on
>>>              the issues of current DF election mechanism. All these
>>>              issues should be resolved in a single new DF election
>>>              draft,rather than in multiple separate drafts. If your
>>>              draft can solve all these issues and stable, i have no
>>>              question for its progressing. But if your draft have not
>>>              solved all issues, i think it had better combine with
>>>              other drafts to provide a comprehensive solution. I think
>>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01 and
>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it should
>>>              be resolved. So i think although your new Hash algorithm
>>>              for DF election is good, it only includes partial
>>>              enhancements, maybe it still needs some time for
>>>consensus.
>>>
>>>              Thanks,
>>>
>>>              weiguo
>>>
>>>             =20
>>>------------------------------------------------------------------------
>>>
>>>              *From:*John E Drake [jdrake@juniper.net
>>>              <mailto:jdrake@juniper.net>]
>>>              *Sent:* Thursday, March 26, 2015 7:16
>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>              <mailto:bess@ietf.org>
>>>              *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES
>>>              based on Paxos algorithm
>>>
>>>              Weiguo,
>>>
>>>              Your proposal introduces a control plane processing load
>>>              that is O(#EVIs * PEs) per DF election and given that
>>>              there can be 4K EVIs per ES, this looks like a
>>>              **substantial** load. Furthermore,
>>>
>>>              you can=B9t  use the ES route to co-ordinate DF election
>>>              because you would need to carry your new extended
>>>              community for each EVI and they would not all fit.  You
>>>              also can=B9t use the Per EVI Ethernet AD route because tha=
t
>>>              is processed by all PEs in the EVI.
>>>
>>>              I think that from a practical perspective the new DF
>>>              election proposed in Satya=B9s draft is sufficiently stabl=
e
>>>              that it renders your draft moot, even if it could be made
>>>              to work.
>>>
>>>              Yours Irrespectively,
>>>
>>>              John
>>>
>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf Of
>>>              *Haoweiguo
>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
>>>              <mailto:bess@ietf.org>
>>>              *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>              based on Paxos algorithm
>>>
>>>              Hi Ali,
>>>
>>>              Thanks for your information. I scanned through this draft,
>>>              it really introduces inter-chassis message for DF election
>>>              handshaking, the requirements in this draft is to
>>>              eliminate transiet Loop and traffic duplication. Current
>>>              EVPN DF election mechanism already eliminated loop and
>>>              traffic duplication by configuring long reception timer on
>>>              each multi-homed PE, but up to reception timer traffic
>>>              disruption issue still exist. EVPN for DCI is an important
>>>              use case for EVPN, up to reception timer traffic
>>>              disruption can't be tolerated for service providers, it
>>>              should be improved.
>>>
>>>              Also for accuracy, i think handshaking state machine on
>>>              each multi-homed PE is also needed. From solution
>>>              perspective, in my draft, no inter-chassis message is
>>>              introduced, only one new extended community is introduced,
>>>              i think the process is comparatively simple than your
>>>              following draft.
>>>
>>>              Current EVPN DF election has some drawbacks, so there are
>>>              three new drafts about DF election emerged. I think BESS
>>>              WG can consider these three drafts in global view, a
>>>              single,comprehensive new DF election draft is hoped.
>>>
>>>              thanks.
>>>
>>>              Thanks,
>>>
>>>              weiguo
>>>
>>>             =20
>>>------------------------------------------------------------------------
>>>
>>>              *From:*BESS [bess-bounces@ietf.org
>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajassi
>>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>]
>>>              *Sent:* Thursday, March 26, 2015 0:21
>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
>>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>>              based on Paxos algorithm
>>>
>>>              FYI- First published July 4, 2011
>>>
>>>             =20
>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/
>>>
>>>              -Ali
>>>
>>>
>>>
>>> _______________________________________________
>>> BESS mailing list
>>> BESS@ietf.org
>>> https://www.ietf.org/mailman/listinfo/bess
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>=20
>>_________________________________________________________________________
>>________________________________________________
>>
>> Ce message et ses pieces jointes peuvent contenir des informations
>>confidentielles ou privilegiees et ne doivent donc
>> pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>>recu ce message par erreur, veuillez le signaler
>> a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>messages electroniques etant susceptibles d'alteration,
>> Orange decline toute responsabilite si ce message a ete altere, deforme
>>ou falsifie. Merci.
>>
>> This message and its attachments may contain confidential or privileged
>>information that may be protected by law;
>> they should not be distributed, used or copied without authorisation.
>> If you have received this email in error, please notify the sender and
>>delete this message and its attachments.
>> As emails may be altered, Orange is not liable for messages that have
>>been modified, changed or falsified.
>> Thank you.
>>
>> _______________________________________________
>> BESS mailing list
>> BESS@ietf.org
>> https://www.ietf.org/mailman/listinfo/bess
>>
>
>
>__________________________________________________________________________
>_______________________________________________
>
>Ce message et ses pieces jointes peuvent contenir des informations
>confidentielles ou privilegiees et ne doivent donc
>pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>recu ce message par erreur, veuillez le signaler
>a l'expediteur et le detruire ainsi que les pieces jointes. Les messages
>electroniques etant susceptibles d'alteration,
>Orange decline toute responsabilite si ce message a ete altere, deforme
>ou falsifie. Merci.
>
>This message and its attachments may contain confidential or privileged
>information that may be protected by law;
>they should not be distributed, used or copied without authorisation.
>If you have received this email in error, please notify the sender and
>delete this message and its attachments.
>As emails may be altered, Orange is not liable for messages that have
>been modified, changed or falsified.
>Thank you.
>_______________________________________________
>BESS mailing list
>BESS@ietf.org
>https://www.ietf.org/mailman/listinfo/bess


From nobody Thu Mar 26 21:58:29 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BB07F1A8A28 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 21:58:27 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.611
X-Spam-Level: 
X-Spam-Status: No, score=-3.611 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id njJfgBw6YThC for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 21:58:24 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 21CD31A8A0C for <bess@ietf.org>; Thu, 26 Mar 2015 21:58:22 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT37515; Fri, 27 Mar 2015 04:58:21 +0000 (GMT)
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 14.3.158.1; Fri, 27 Mar 2015 04:55:51 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml401-hub.china.huawei.com ([10.98.56.32]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 12:55:44 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2g==
Date: Fri, 27 Mar 2015 04:55:43 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com>, <D13A263C.2CB9E%satyamoh@cisco.com>
In-Reply-To: <D13A263C.2CB9E%satyamoh@cisco.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.153.170]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/9tZHM3wGy82ltTPtZLTkrzK8huo>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 04:58:27 -0000

Hi Satya,
Pls see inline.

Thanks,
weiguo

________________________________________
From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
Sent: Friday, March 27, 2015 12:11
To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Thomas,

Thanks for your summary. It captures the essence of the discussion.
[weiguo]: Yes, i agree.

One feature of the HRW scheme is that it computes the Active and the
Backup DF upfront.
Whenever the Active PE, say PE1, dies;=20
the Backup PE, PE2, can become the
Active immediately upon receiving the withdraw of the ES route from PE1.
It need not even have to wait for the 3 seconds time.
[weiguo]: I think maybe your negligence, dies PE1 can't send withdraw messa=
ge:) But when PE1 leaves ESI, it can withdraw the ES route immediately.
When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2 will bo=
th act as PE for some transiet time.

On the other hand, the 3 second timer will serve as a bulk and batch
mechanism for the case when new PE(s) is(are) coming up, so we don=B9t have
to do a DF election ever so often.

Haoweiguo,  I think the "loop" scenario is speculative.
I think everybody understands that, yes, there may be some transient
duplication of bum traffic; however, there is no "loop" problem to be
solved.
[weiguo]: Traffic duplication and loop issue both exist in dual DF PE case.=
 In multi-homed device case, say CE1 is multi-homed to PE1 and PE2 and both=
 PE are DF PE. When CE1 sends traffic to PE1, the traffic will loop back to=
 CE1 through PE2. If CE1 can filter and drop the traffic , it is fortunate,=
 maybe no loop.=20
For multi-homed network scenario, say local network1(CE1 and CE2) is multi-=
homed to PE1 and PE2. Similarly, the traffic from CE1 will loop back throug=
h PE1-->PE2-->CE2, it is very serious. It is absolutely not allowed.

Thanks,
--Satya



On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:

>Hi Thomas,
>I know you are neutral, sorry for my reply vulnerable to be miunderstood.
>
>Thanks,
>weiguo
>
>________________________________________
>From: thomas.morin@orange.com [thomas.morin@orange.com]
>Sent: Friday, March 27, 2015 11:59
>To: Haoweiguo; bess@ietf.org
>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>algorithm
>
>Hi Weiguo,
>
>2015-03-26, Haoweiguo:
> > Thomas:
>> > - RFC7432 may have transient periods where the DF election state is
>> > not yet synchronized between the two peers:
> >
>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic loop
>
>Note well that I didn't write that there can be transient _loops_ .
>I tried to capture the exchange you had all, and what I gather is that
>the split-horizon procedure _prevents_loops_, independently of any
>transient period where DF state is not synchronized and which may lead
>to transient _duplicates_.
>
>-Thomas
>
>>
>> 26/03/2015 20:05, John E Drake :
>>>
>>> Weiguo,
>>>
>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I have
>>> detailed, is a non-solution to a non-problem with tremendous control
>>> plane cost.
>>>
>>> Yours Irrespectively,
>>>
>>> John
>>>
>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>> *Sent:* Thursday, March 26, 2015 6:17 PM
>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> Pls see below.
>>>
>>> Thanks,
>>>
>>> weiguo
>>>
>>>
>>>------------------------------------------------------------------------
>>>
>>> *From:*John E Drake [jdrake@juniper.net]
>>> *Sent:* Friday, March 27, 2015 6:00
>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> To recap,
>>>
>>> We have established that your proposal is untenable because of its
>>> control plane load.
>>>
>>> We have established that your proposal is based upon a flawed
>>> understanding of the DF election in RFC 7432.
>>>
>>> [weiguo]: In ethernet world, traffic loop is serious than short timer
>>> traffic disruption. If you want to implement  transiet traffic loop
>>> process, i will modify my draft to solve your issue.
>>>
>>> If i am the developer, i will prefer short timer traffic disruption
>>> based on current EVPN protocol.
>>>
>>> What you are now arguing is that your draft prevents two or more PEs
>>> from being DF simultaneously. This is clearly nonsense.
>>>
>>> [weiguo]: I will modify the draft problem statements, and use the same
>>> handshaking solution to solve it.
>>>
>>> Furthermore, we have established that having two or more DFs for what
>>> even you admit is a brief transient leads to duplicate traffic, which
>>> is acceptable, but not loops, your assertion to the contrary.
>>>
>>> [weiguo]: It is transient loop and traffic duplication issue.
>>>
>>> Yours Irrespectively,
>>>
>>> John
>>>
>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>> *Sent:* Thursday, March 26, 2015 5:37 PM
>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> John,
>>>
>>> As your understanding of the EVPN draft, the DF election mechanism has
>>> more serious side effect, it will have short time traffic loop,i.e.,
>>> dual DF PEs will exist for a short time. I think dual DF PEs is
>>> absolutely not tolerated, because native ethernet header has no TTL,
>>> up to several hundred ms traffic loop normally not tolerated in
>>> commertial networks.
>>>
>>> As your understanding, the PEs should do as following:
>>>
>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have
>>> rarely deployment.
>>>
>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins
>>> ESI and trigger DF re-election. When reception timer expires:
>>>
>>> PE1 upgrades to DF PE.
>>>
>>> After reception timer+ ES route transmission timer:
>>>
>>> PE2 downloads to non-DF PE.
>>>
>>> So in timer clock sync case, dual DF PEs will exist at least
>>> transmission timer.
>>>
>>> If NTP is used for timer sync, because it has bad accuracy, dual DF
>>> PEs will exist more longer timer.
>>>
>>> So as your understanding for DF election, the drawback is more clear.
>>>
>>> Thanks,
>>>
>>> weiguo
>>>
>>>
>>>------------------------------------------------------------------------
>>>
>>> *From:*John E Drake [jdrake@juniper.net]
>>> *Sent:* Friday, March 27, 2015 4:41
>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> Weiguo,
>>>
>>> We have already established that your proposal is untenable because of
>>> its control plane load.
>>>
>>> What we are now discussing is that your proposal is based upon a
>>> misunderstanding of the algorithm in RFC 7432.  You are assuming that
>>> PE1 will advertise an ES route and then wait for the configured
>>> interval before performing the DF election while PE2 and PE3 will
>>> perform the DF election as soon as they receive the ES route from PE1.
>>> This is not what RFC 7432 says.
>>>
>>> Rather, what is says is that the advertisement of the ES route by PE1
>>> and its receipt by PE2 and PE3 causes all three PEs to start the
>>> configured interval timer  - =B33. When the timer expires, each PE
>>> builds an ordered list of the IP addresses of all the PE nodes
>>> connected to the Ethernet segment (including itself), in increasing
>>> numeric value.=B2
>>>
>>> Yours Irrespectively,
>>>
>>> John
>>>
>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>> *Sent:* Thursday, March 26, 2015 3:26 PM
>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> Pls read my detail replies to Satya. If you still can't catch it, pls
>>> read my draft and EVPN base protocol,  thanks
>>>
>>> weiguo
>>>
>>>
>>>------------------------------------------------------------------------
>>>
>>> *From:*John E Drake [jdrake@juniper.net]
>>> *Sent:* Friday, March 27, 2015 1:28
>>> *To:* Patrice Brissette (pbrisset)
>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>> <mailto:bess@ietf.org>
>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>>> Paxos algorithm
>>>
>>> I think Patrice is correct.  Your proposal doesn't solve the problem
>>> and it does so at huge cost.
>>>
>>> Sent from my iPhone
>>>
>>>
>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>>
>>>      Weiguo,
>>>
>>>      I=B9m not sure I=B9m following here.
>>>
>>>      Don=B9t you have the same issue with your handshaking mechanism?
>>>
>>>      If you don=B9t know your peer, how can you handshake?
>>>
>>>      Regards,
>>>
>>>      Patrice
>>>
>>>      Image removed by sender.
>>>
>>>      *Patrice Brissette*
>>>      TECHNICAL LEADER.ENGINEERING
>>>
>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>      Phone: *+1 613 254 3336*
>>>
>>>
>>>
>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>      Canada
>>>      Cisco.com <http://www.cisco.com/global/CA/>
>>>
>>>
>>>
>>>      Image removed by sender.Think before you print.
>>>
>>>      This email may contain confidential and privileged material for
>>>      the sole use of the intended recipient. Any review, use,
>>>      distribution or disclosure by others is strictly prohibited. If
>>>      you are not the intended recipient (or authorized to receive for
>>>      the recipient), please contact the sender by reply email and
>>>      delete all copies of this message.
>>>
>>>      Please click here
>>>
>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>      for Company Registration Information.
>>>
>>>
>>>      *From: *Haoweiguo <haoweiguo@huawei.com
>>><mailto:haoweiguo@huawei.com>>
>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
>>>      *To: *Patrice Brissette <pbrisset@cisco.com
>>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>      Paxos algorithm
>>>
>>>          Hi Patrice,
>>>
>>>          Up to reception timer traffic disruption in transient phase is
>>>          one of the issues.
>>>
>>>          Thanks,
>>>
>>>          weiguo
>>>
>>>
>>>------------------------------------------------------------------------
>>>
>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>>          <mailto:pbrisset@cisco.com>]
>>>          *Sent:* Thursday, March 26, 2015 20:54
>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>>          bess@ietf.org <mailto:bess@ietf.org>
>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>          based on Paxos algorithm
>>>
>>>          Weiguo,
>>>
>>>          You mention "But if your draft have not solved all issues=B2,
>>>
>>>          Can you explain what Satya=B9s draft is not solving?
>>>
>>>          Regards,
>>>
>>>          Patrice
>>>
>>>          Image removed by sender.
>>>
>>>          *Patrice Brissette*
>>>          TECHNICAL LEADER.ENGINEERING
>>>
>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>          Phone: *+1 613 254 3336*
>>>
>>>
>>>
>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>          Canada
>>>          Cisco.com <http://www.cisco.com/global/CA/>
>>>
>>>
>>>
>>>          Image removed by sender.Think before you print.
>>>
>>>          This email may contain confidential and privileged material
>>>          for the sole use of the intended recipient. Any review, use,
>>>          distribution or disclosure by others is strictly prohibited.
>>>          If you are not the intended recipient (or authorized to
>>>          receive for the recipient), please contact the sender by reply
>>>          email and delete all copies of this message.
>>>
>>>          Please click here
>>>
>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>          for Company Registration Information.
>>>
>>>
>>>          *From: *Haoweiguo <haoweiguo@huawei.com
>>>          <mailto:haoweiguo@huawei.com>>
>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>>          *To: *John E Drake <jdrake@juniper.net
>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>          <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>>          based on Paxos algorithm
>>>
>>>              Hi John,
>>>
>>>               Firstly i think EVPN community should reach consensus on
>>>              the issues of current DF election mechanism. All these
>>>              issues should be resolved in a single new DF election
>>>              draft,rather than in multiple separate drafts. If your
>>>              draft can solve all these issues and stable, i have no
>>>              question for its progressing. But if your draft have not
>>>              solved all issues, i think it had better combine with
>>>              other drafts to provide a comprehensive solution. I think
>>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01 and
>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it should
>>>              be resolved. So i think although your new Hash algorithm
>>>              for DF election is good, it only includes partial
>>>              enhancements, maybe it still needs some time for
>>>consensus.
>>>
>>>              Thanks,
>>>
>>>              weiguo
>>>
>>>
>>>------------------------------------------------------------------------
>>>
>>>              *From:*John E Drake [jdrake@juniper.net
>>>              <mailto:jdrake@juniper.net>]
>>>              *Sent:* Thursday, March 26, 2015 7:16
>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>              <mailto:bess@ietf.org>
>>>              *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES
>>>              based on Paxos algorithm
>>>
>>>              Weiguo,
>>>
>>>              Your proposal introduces a control plane processing load
>>>              that is O(#EVIs * PEs) per DF election and given that
>>>              there can be 4K EVIs per ES, this looks like a
>>>              **substantial** load. Furthermore,
>>>
>>>              you can=B9t  use the ES route to co-ordinate DF election
>>>              because you would need to carry your new extended
>>>              community for each EVI and they would not all fit.  You
>>>              also can=B9t use the Per EVI Ethernet AD route because tha=
t
>>>              is processed by all PEs in the EVI.
>>>
>>>              I think that from a practical perspective the new DF
>>>              election proposed in Satya=B9s draft is sufficiently stabl=
e
>>>              that it renders your draft moot, even if it could be made
>>>              to work.
>>>
>>>              Yours Irrespectively,
>>>
>>>              John
>>>
>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf Of
>>>              *Haoweiguo
>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
>>>              <mailto:bess@ietf.org>
>>>              *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>              based on Paxos algorithm
>>>
>>>              Hi Ali,
>>>
>>>              Thanks for your information. I scanned through this draft,
>>>              it really introduces inter-chassis message for DF election
>>>              handshaking, the requirements in this draft is to
>>>              eliminate transiet Loop and traffic duplication. Current
>>>              EVPN DF election mechanism already eliminated loop and
>>>              traffic duplication by configuring long reception timer on
>>>              each multi-homed PE, but up to reception timer traffic
>>>              disruption issue still exist. EVPN for DCI is an important
>>>              use case for EVPN, up to reception timer traffic
>>>              disruption can't be tolerated for service providers, it
>>>              should be improved.
>>>
>>>              Also for accuracy, i think handshaking state machine on
>>>              each multi-homed PE is also needed. From solution
>>>              perspective, in my draft, no inter-chassis message is
>>>              introduced, only one new extended community is introduced,
>>>              i think the process is comparatively simple than your
>>>              following draft.
>>>
>>>              Current EVPN DF election has some drawbacks, so there are
>>>              three new drafts about DF election emerged. I think BESS
>>>              WG can consider these three drafts in global view, a
>>>              single,comprehensive new DF election draft is hoped.
>>>
>>>              thanks.
>>>
>>>              Thanks,
>>>
>>>              weiguo
>>>
>>>
>>>------------------------------------------------------------------------
>>>
>>>              *From:*BESS [bess-bounces@ietf.org
>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajassi
>>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>]
>>>              *Sent:* Thursday, March 26, 2015 0:21
>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
>>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>>              based on Paxos algorithm
>>>
>>>              FYI- First published July 4, 2011
>>>
>>>
>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route/
>>>
>>>              -Ali
>>>
>>>
>>>
>>> _______________________________________________
>>> BESS mailing list
>>> BESS@ietf.org
>>> https://www.ietf.org/mailman/listinfo/bess
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>_________________________________________________________________________
>>________________________________________________
>>
>> Ce message et ses pieces jointes peuvent contenir des informations
>>confidentielles ou privilegiees et ne doivent donc
>> pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>>recu ce message par erreur, veuillez le signaler
>> a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>messages electroniques etant susceptibles d'alteration,
>> Orange decline toute responsabilite si ce message a ete altere, deforme
>>ou falsifie. Merci.
>>
>> This message and its attachments may contain confidential or privileged
>>information that may be protected by law;
>> they should not be distributed, used or copied without authorisation.
>> If you have received this email in error, please notify the sender and
>>delete this message and its attachments.
>> As emails may be altered, Orange is not liable for messages that have
>>been modified, changed or falsified.
>> Thank you.
>>
>> _______________________________________________
>> BESS mailing list
>> BESS@ietf.org
>> https://www.ietf.org/mailman/listinfo/bess
>>
>
>
>__________________________________________________________________________
>_______________________________________________
>
>Ce message et ses pieces jointes peuvent contenir des informations
>confidentielles ou privilegiees et ne doivent donc
>pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>recu ce message par erreur, veuillez le signaler
>a l'expediteur et le detruire ainsi que les pieces jointes. Les messages
>electroniques etant susceptibles d'alteration,
>Orange decline toute responsabilite si ce message a ete altere, deforme
>ou falsifie. Merci.
>
>This message and its attachments may contain confidential or privileged
>information that may be protected by law;
>they should not be distributed, used or copied without authorisation.
>If you have received this email in error, please notify the sender and
>delete this message and its attachments.
>As emails may be altered, Orange is not liable for messages that have
>been modified, changed or falsified.
>Thank you.
>_______________________________________________
>BESS mailing list
>BESS@ietf.org
>https://www.ietf.org/mailman/listinfo/bess=


From nobody Thu Mar 26 22:12:58 2015
Return-Path: <satyamoh@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C1E211A89B1 for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 22:12:56 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -13.91
X-Spam-Level: 
X-Spam-Status: No, score=-13.91 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id B3yn_RMK9CqD for <bess@ietfa.amsl.com>; Thu, 26 Mar 2015 22:12:50 -0700 (PDT)
Received: from rcdn-iport-2.cisco.com (rcdn-iport-2.cisco.com [173.37.86.73]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2B22C1A7023 for <bess@ietf.org>; Thu, 26 Mar 2015 22:12:50 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=87675; q=dns/txt; s=iport; t=1427433170; x=1428642770; h=from:to:subject:date:message-id:references:in-reply-to: mime-version; bh=ncHw4ogZV7BF+5SJFPeoVyInq6nAzUxFv6L9SrvzP6s=; b=mIXBZXCvZ9yyCnrWm++cPS1uQ0dLAq6WLf6tSjSJDUxaghkFstRr53LN eXX64/m6n9axPVlpuGT9uMYrFyWs3Z+r+A4QglhT9Uix+Df6dEK/0Udpv OS/lBWzEWUAIsK/9dG/EHaKsdZbyKV8V2Fsc91iPjDrkmYevUslnuCEzz 4=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0AhBQA+5hRV/51dJa1ZA4JDQ1JaBIMPwgAZAQuFcwIcgSBMAQEBAQEBfYQUAQEBAwEBAQEXAQw/CBAJAgIBCBEDAQEBIQEGBQICGQwLFAkIAgQBEhQHiAwIDZR2nG4GgUyYaQEBAQEBAQEBAQEBAQEBAQEBAQEBARcEiySEDQgBEQEOFgYLCgELAQsGC4JRgUsFhhiIKoIOg2+GAIEbOoJ2glqJQYNHIoIBHYFQbwGBAwcXIn8BAQE
X-IronPort-AV: E=Sophos;i="5.11,477,1422921600";  d="scan'208,217";a="407424423"
Received: from rcdn-core-6.cisco.com ([173.37.93.157]) by rcdn-iport-2.cisco.com with ESMTP; 27 Mar 2015 05:12:48 +0000
Received: from xhc-rcd-x07.cisco.com (xhc-rcd-x07.cisco.com [173.37.183.81]) by rcdn-core-6.cisco.com (8.14.5/8.14.5) with ESMTP id t2R5Cm5m026815 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 05:12:48 GMT
Received: from xmb-aln-x07.cisco.com ([169.254.2.199]) by xhc-rcd-x07.cisco.com ([173.37.183.81]) with mapi id 14.03.0195.001; Fri, 27 Mar 2015 00:12:48 -0500
From: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>
To: Haoweiguo <haoweiguo@huawei.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ1HYE/ZJUJbtWUGUXbn7g+eTDJ0ucdUAgACbTgCAACnuAIAAAsqAgAAf4ACAACCzAIAAFSOAgAAPjACAAAZ+AIAABLkAgAAvJACAAB1OgIAAD/gAgAADOwCAAAKcAP//i5CAgACBl4D//49dAA==
Date: Fri, 27 Mar 2015 05:12:47 +0000
Message-ID: <D13A32A3.2CC08%satyamoh@cisco.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.65.50.166]
Content-Type: multipart/alternative; boundary="_000_D13A32A32CC08satyamohciscocom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/FtEe-ldYs0NyAYSf9Ea0DCnUnBU>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 05:12:57 -0000

--_000_D13A32A32CC08satyamohciscocom_
Content-Type: text/plain; charset="euc-kr"
Content-Transfer-Encoding: base64
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--_000_D13A32A32CC08satyamohciscocom_
Content-Type: text/html; charset="euc-kr"
Content-ID: <D5DABA22E6A13546BDE0AF5492A805F0@emea.cisco.com>
Content-Transfer-Encoding: base64
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==

--_000_D13A32A32CC08satyamohciscocom_--


From nobody Fri Mar 27 02:52:42 2015
Return-Path: <zhuangshunwan@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5B10F1ACCFF for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 02:52:40 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.21
X-Spam-Level: 
X-Spam-Status: No, score=-4.21 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id rChIDL7OXW62 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 02:52:37 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A76241AC413 for <bess@ietf.org>; Fri, 27 Mar 2015 02:52:36 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml401-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BUE85537; Fri, 27 Mar 2015 09:52:34 +0000 (GMT)
Received: from nkgeml407-hub.china.huawei.com (10.98.56.38) by lhreml401-hub.china.huawei.com (10.201.5.240) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 09:52:34 +0000
Received: from NKGEML512-MBS.china.huawei.com ([169.254.8.209]) by nkgeml407-hub.china.huawei.com ([10.98.56.38]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 17:52:30 +0800
From: Zhuangshunwan <zhuangshunwan@huawei.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [BESS] Comment to draft-ietf-bess-evpn-usage-00.txt
Thread-Index: AdBoc7wvAICze0pmQBu6f1RPvywcrA==
Date: Fri, 27 Mar 2015 09:52:29 +0000
Message-ID: <19AB2A007F56DB4E8257F949A2FB98588D289A52@NKGEML512-MBS.china.huawei.com>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.86.195]
Content-Type: multipart/alternative; boundary="_000_19AB2A007F56DB4E8257F949A2FB98588D289A52NKGEML512MBSchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/K8rqT23bJG3-sQJ5iZ2oH-UK8aE>
Subject: [bess] [BESS] Comment to draft-ietf-bess-evpn-usage-00.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 09:52:40 -0000

--_000_19AB2A007F56DB4E8257F949A2FB98588D289A52NKGEML512MBSchi_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Dear Authors,

I have a comment, see inline with [Shunwan].

5.2.1. Service startup procedures

   As soon as the EVIs are created in PE1, PE2 and PE3, the following
   control plane actions are carried out:

   o Flooding tree setup per EVI (4k routes): Each PE will send one
     Inclusive Multicast Ethernet Tag route per EVI (up to 4k routes per
     PE) so that the flooding tree per EVI can be setup. Note that
     ingress replication, P2MP LSPs or MP2MP LSPs can optionally be
     signaled in the PMSI Tunnel attribute and the corresponding tree be
     created.

   o Ethernet A-D routes per ESI (a set of routes for ESI12): A set of
     A-D routes with a list of 4k RTs (one per EVI) for ESI12 will be
     issued from PE1 and PE2 (it has to be a set of routes so that the
     total number of RTs can be conveyed). This set will also include

[Shunwan]:
Sorry, maybe I miss something about this feature, I just don't know how to =
convey a bgp route with 4k RTs.
The BGP specification mandates a maximum BGP message size of 4096 octets, s=
o a bgp route can only carry approximately 500 RTs once.
If we send one Ethernet A-D route per ESI with 4k RTs using a set of routes=
 (Each route carrying a sub-set of 4K RTs) , because such set of routes hav=
e the same NLRI, in line with the behavior of BGP [RFC4271], the last route=
 will override the previous routes, so the receiver PE can only maintain th=
e last one other than a set of routes for one Ethernet A-D route with 4k RT=
s.

If I miss something, please let me know, thanks.


Rabadan-Palislamovic et al.Expires May 17, 2015                [Page 11]


Internet-Draft                 EVPN Usage              November 13, 2014


     ESI Label extended communities with the active-standby flag set to
     zero (all-active multi-homing type) and an ESI Label different from
     zero (used for split-horizon functions). These routes will be
     imported by the three PEs, since the RTs match the EVI RTs locally
     configured. The A-D routes per ESI will be used for fast
     convergence and split-horizon functions, as discussed in [EVPN].



Regards,

Shunwan




--_000_19AB2A007F56DB4E8257F949A2FB98588D289A52NKGEML512MBSchi_
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=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:0cm;
	margin-bottom:.0001pt;
	text-align:justify;
	text-justify:inter-ideograph;
	font-size:10.5pt;
	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.MsoPlainText, li.MsoPlainText, div.MsoPlainText
	{mso-style-priority:99;
	mso-style-link:"\7EAF\6587\672C Char";
	margin:0cm;
	margin-bottom:.0001pt;
	font-size:10.5pt;
	font-family:"Calibri","sans-serif";}
span.EmailStyle17
	{mso-style-type:personal-compose;
	font-family:"Calibri","sans-serif";
	color:windowtext;}
span.Char
	{mso-style-name:"\7EAF\6587\672C Char";
	mso-style-priority:99;
	mso-style-link:\7EAF\6587\672C;
	font-family:"Calibri","sans-serif";}
.MsoChpDefault
	{mso-style-type:export-only;}
/* Page Definitions */
@page WordSection1
	{size:612.0pt 792.0pt;
	margin:72.0pt 90.0pt 72.0pt 90.0pt;}
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"ZH-CN" link=3D"blue" vlink=3D"purple" style=3D"text-justify-t=
rim:punctuation">
<div class=3D"WordSection1">
<p class=3D"MsoPlainText"><span lang=3D"EN-US" style=3D"color:#4F81BD">Dear=
 Authors,<o:p></o:p></span></p>
<p class=3D"MsoPlainText"><span lang=3D"EN-US" style=3D"color:#4F81BD">I ha=
ve a comment, see inline with [Shunwan].<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">5.2.1. Service startup procedur=
es<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp; As soon as the EVI=
s are created in PE1, PE2 and PE3, the following<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp; control plane acti=
ons are carried out:<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp; o Flooding tree se=
tup per EVI (4k routes): Each PE will send one<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; Inclus=
ive Multicast Ethernet Tag route per EVI (up to 4k routes per<o:p></o:p></s=
pan></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; PE) so=
 that the flooding tree per EVI can be setup. Note that<o:p></o:p></span></=
p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; ingres=
s replication, P2MP LSPs or MP2MP LSPs can optionally be<o:p></o:p></span><=
/p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; signal=
ed in the PMSI Tunnel attribute and the corresponding tree be<o:p></o:p></s=
pan></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; create=
d.<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp; o Ethernet A-D rou=
tes per ESI (a set of routes for ESI12): A set of<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; A-D ro=
utes with a list of 4k RTs (one per EVI) for ESI12 will be<o:p></o:p></span=
></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; issued=
 from PE1 and PE2 (it has to be a set of routes so that the<o:p></o:p></spa=
n></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; total =
number of RTs can be conveyed). This set will also include<o:p></o:p></span=
></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"color:#1F497D">[Shunwa=
n]: <o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"color:#1F497D">Sorry, =
maybe I miss something about this feature, I just don&#8217;t know how to c=
onvey a bgp route with 4k RTs.<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"color:#1F497D">The BGP=
 specification mandates a maximum BGP message size of 4096 octets, so a bgp=
 route can only carry approximately 500 RTs once.<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"color:#1F497D">If we s=
end one Ethernet A-D route per ESI with 4k RTs using a set of routes (Each =
route carrying a sub-set of 4K RTs) , because such set of routes have the s=
ame NLRI, in line with the behavior of
 BGP [RFC4271], the last route will override the previous routes, so the re=
ceiver PE can only maintain the last one other than a set of routes for one=
 Ethernet A-D route with 4k RTs.<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"color:#1F497D"><o:p>&n=
bsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"color:#1F497D">If I mi=
ss something, please let me know, thanks.<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">Rabadan-Palislamovic et al.Expi=
res May 17, 2015&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; [Page 11]<o:p></o:p></span></p>
<span lang=3D"EN-US" style=3D"font-size:10.5pt;font-family:&quot;Calibri&qu=
ot;,&quot;sans-serif&quot;"><br clear=3D"all" style=3D"page-break-before:al=
ways">
</span>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">Internet-Draft&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp; EVPN Usage&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=
&nbsp;&nbsp;&nbsp; November 13, 2014<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; ESI La=
bel extended communities with the active-standby flag set to<o:p></o:p></sp=
an></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; zero (=
all-active multi-homing type) and an ESI Label different from<o:p></o:p></s=
pan></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; zero (=
used for split-horizon functions). These routes will be<o:p></o:p></span></=
p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; import=
ed by the three PEs, since the RTs match the EVI RTs locally<o:p></o:p></sp=
an></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; config=
ured. The A-D routes per ESI will be used for fast<o:p></o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US">&nbsp;&nbsp;&nbsp;&nbsp; conver=
gence and split-horizon functions, as discussed in [EVPN].<o:p></o:p></span=
></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoPlainText"><span lang=3D"EN-US">Regards,<o:p></o:p></span></=
p>
<p class=3D"MsoPlainText"><span lang=3D"EN-US">Shunwan<o:p></o:p></span></p=
>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
<p class=3D"MsoNormal"><span lang=3D"EN-US"><o:p>&nbsp;</o:p></span></p>
</div>
</body>
</html>

--_000_19AB2A007F56DB4E8257F949A2FB98588D289A52NKGEML512MBSchi_--


From nobody Fri Mar 27 05:15:07 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4DE581A1B53 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 05:15:05 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.909
X-Spam-Level: 
X-Spam-Status: No, score=-6.909 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gD_8wYmLZXw9 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 05:15:03 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A907E1A8775 for <bess@ietf.org>; Fri, 27 Mar 2015 05:15:01 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id 6E8463DF1FC2E; Fri, 27 Mar 2015 12:14:54 +0000 (GMT)
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2RCEuRj026317 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 13:14:56 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR711WXCHHUB01.zeu.alcatel-lucent.com ([135.239.2.111]) with mapi id 14.03.0195.001; Fri, 27 Mar 2015 13:14:56 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Zhuangshunwan <zhuangshunwan@huawei.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] [BESS] Comment to draft-ietf-bess-evpn-usage-00.txt
Thread-Index: AQHQaIeiNE+ntgeI/0aPTMZtEe/f9g==
Date: Fri, 27 Mar 2015 12:14:55 +0000
Message-ID: <D13A9625.6C8AC%jorge.rabadan@alcatel-lucent.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.39]
Content-Type: multipart/alternative; boundary="_000_D13A96256C8ACjorgerabadanalcatellucentcom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/Ql2JMhAG3KcTvWrocV-HcZXp330>
Subject: Re: [bess] [BESS] Comment to draft-ietf-bess-evpn-usage-00.txt
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 12:15:05 -0000

--_000_D13A96256C8ACjorgerabadanalcatellucentcom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_D13A96256C8ACjorgerabadanalcatellucentcom_
Content-Type: text/html; charset="utf-8"
Content-ID: <A2EC4F782E6FD840AECEB19BF0BD7C01@exchange.lucent.com>
Content-Transfer-Encoding: base64
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--_000_D13A96256C8ACjorgerabadanalcatellucentcom_--


From nobody Fri Mar 27 05:26:59 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 553F01ACD93 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 05:26:57 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mAEEPCKV_N5D for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 05:26:52 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 736111ACD8B for <bess@ietf.org>; Fri, 27 Mar 2015 05:26:51 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id 59B7A53746B7C; Fri, 27 Mar 2015 12:26:48 +0000 (GMT)
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2RCQoA1015321 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 13:26:50 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR712WXCHHUB03.zeu.alcatel-lucent.com ([135.239.2.74]) with mapi id 14.03.0195.001; Fri, 27 Mar 2015 13:26:50 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Haoweiguo <haoweiguo@huawei.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8QfEaq8IGmBqka8wmUCt0qdaZ0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HOA
Date: Fri, 27 Mar 2015 12:26:50 +0000
Message-ID: <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.39]
Content-Type: text/plain; charset="utf-8"
Content-ID: <42D6B9730E56EE4092B55CC1A06D4982@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/kuAjxIgmD5k4WV2lbWgz6jIKBWY>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 12:26:57 -0000
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From nobody Fri Mar 27 06:02:16 2015
Return-Path: <pbrisset@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 589D01ACDCC for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 06:02:15 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -13.911
X-Spam-Level: 
X-Spam-Status: No, score=-13.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id vezPlgboQXCd for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 06:02:10 -0700 (PDT)
Received: from mtv-iport-2.cisco.com (mtv-iport-2.cisco.com [173.36.130.13]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5F3AF1ACDD2 for <bess@ietf.org>; Fri, 27 Mar 2015 06:02:10 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=30122; q=dns/txt; s=iport; t=1427461330; x=1428670930; h=from:to:subject:date:message-id:references:in-reply-to: content-id:content-transfer-encoding:mime-version; bh=nPtHO6OJ6cvELYImqXqpfSzS6RU1eXg5Nqb3pjsfsi8=; b=T5POo8BAu55XQTlSjEXZbzMxKgzZjob0A9nWipXg0LSeFU/xpQ/hr+m0 bhOXFciUsWxMogJ3ckCUcca0vHqvd3HnCo0DPXmR1Ikl8JnBLdOTnPS7t bdiGpgRQdkbZDDjCJTkA2eIxP7/h4Pwkr2e/Fj5+pHMcBWUuK0TuYJ3MM A=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0BKCQDpUxVV/51dJa1ZA4MGUloEgw+/b4IxCoV1AhyBJEwBAQEBAQF9hBQBAQEEAQEBMTIIGQICAQgRAwEBAQEEIwUCAhkMCxQJCAIEARIUB4gUDZRtnG4GgUyYYgEBAQEBAQEBAQEBAQEBAQEBAQEBARcEgReKDYQNCAERAQ4cCwsLDAYLglGBSwWGGoo9g2+GAYEbOoJ3glqJQ4NIIoIBHYFQbwGBAwcXIn8BAQE
X-IronPort-AV: E=Sophos;i="5.11,478,1422921600"; d="scan'208";a="114590243"
Received: from rcdn-core-6.cisco.com ([173.37.93.157]) by mtv-iport-2.cisco.com with ESMTP; 27 Mar 2015 13:02:08 +0000
Received: from xhc-rcd-x09.cisco.com (xhc-rcd-x09.cisco.com [173.37.183.83]) by rcdn-core-6.cisco.com (8.14.5/8.14.5) with ESMTP id t2RD28ub005139 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 13:02:08 GMT
Received: from xmb-aln-x09.cisco.com ([169.254.4.43]) by xhc-rcd-x09.cisco.com ([173.37.183.83]) with mapi id 14.03.0195.001; Fri, 27 Mar 2015 08:02:08 -0500
From: "Patrice Brissette (pbrisset)" <pbrisset@cisco.com>
To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, Haoweiguo <haoweiguo@huawei.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQaI46Ws9VLcZ0TUWO8xI+wRZimQ==
Date: Fri, 27 Mar 2015 13:02:06 +0000
Message-ID: <D13ABEC3.4F07E%pbrisset@cisco.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com>
In-Reply-To: <D13A263C.2CB9E%satyamoh@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.89.14.251]
Content-Type: text/plain; charset="euc-kr"
Content-ID: <DC0F5F379EDAA6449499EE3F5BE14196@emea.cisco.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/CVP8CydKMShQU-FPTvrBv8fgGQg>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 13:02:15 -0000
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From nobody Fri Mar 27 07:29:57 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1C1841AC7E8 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:29:56 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.611
X-Spam-Level: 
X-Spam-Status: No, score=-3.611 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ASMmpa9PH_Lr for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:29:48 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0B53C1A9031 for <bess@ietf.org>; Fri, 27 Mar 2015 07:29:47 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml402-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT89814; Fri, 27 Mar 2015 14:29:46 +0000 (GMT)
Received: from NKGEML401-HUB.china.huawei.com (10.98.56.32) by lhreml402-hub.china.huawei.com (10.201.5.241) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 14:29:45 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml401-hub.china.huawei.com ([10.98.56.32]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 22:29:40 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HYAABS9Z9A=
Date: Fri, 27 Mar 2015 14:29:39 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com>, <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com>
In-Reply-To: <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.144.149]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/MezBR0oqDTh70V6WyvTPv6NF8Uk>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 14:29:56 -0000

Hi Jorge,
Sorry, maybe i have not explain clearly, pls allow me to clarify it more cl=
early.
In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2 is s=
ingle homed to remote PE3. CE1 and CE2 forms one layer 2 network, so it mul=
ti-homed network scenario.
           CE3
             |
            PE3


PE1                  PE2
  |                       |
CE1-----------CE2

In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends BUM=
 traffic to CE1, the traffic will reach PE1 and PE2, CE1 will receive two c=
opy of the traffic from CE3, also loop is formed, the form path:

CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3

I think the traffic loop is absolutely not allowed. In MHN case, the loop w=
ill formed due to dual-DF PE in transiet period.

Thanks,
weiguo



________________________________________
From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
Sent: Friday, March 27, 2015 20:26
To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com; bess@ietf=
.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Hi Weiguo,

About this:

[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
loop back to CE1 through PE2. If CE1 can filter and drop the traffic , it
is fortunate, maybe no loop.
For multi-homed network scenario, say local network1(CE1 and CE2) is
multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop back
through PE1-->PE2-->CE2, it is very serious. It is absolutely not allowed.


[JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2 will
not send the traffic back to CE1 due to the ESI-label and split-horizon
procedures.
In MH network case PE2 will NOT send the traffic to CE2 for the same
reason.

Let me know if you disagree please.

Thanks.
Jorge

-----Original Message-----
From: Haoweiguo <haoweiguo@huawei.com>
Date: Thursday, March 26, 2015 at 9:55 PM
To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
"thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org"
<bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
algorithm

>Hi Satya,
>Pls see inline.
>
>Thanks,
>weiguo
>
>________________________________________
>From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
>Sent: Friday, March 27, 2015 12:11
>To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>algorithm
>
>Thomas,
>
>Thanks for your summary. It captures the essence of the discussion.
>[weiguo]: Yes, i agree.
>
>One feature of the HRW scheme is that it computes the Active and the
>Backup DF upfront.
>Whenever the Active PE, say PE1, dies;
>the Backup PE, PE2, can become the
>Active immediately upon receiving the withdraw of the ES route from PE1.
>It need not even have to wait for the 3 seconds time.
>[weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
>message:) But when PE1 leaves ESI, it can withdraw the ES route
>immediately.
>When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2 will
>both act as PE for some transiet time.
>
>On the other hand, the 3 second timer will serve as a bulk and batch
>mechanism for the case when new PE(s) is(are) coming up, so we don=B9t hav=
e
>to do a DF election ever so often.
>
>Haoweiguo,  I think the "loop" scenario is speculative.
>I think everybody understands that, yes, there may be some transient
>duplication of bum traffic; however, there is no "loop" problem to be
>solved.
>[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
>and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>loop back to CE1 through PE2. If CE1 can filter and drop the traffic , it
>is fortunate, maybe no loop.
>For multi-homed network scenario, say local network1(CE1 and CE2) is
>multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
>allowed.
>
>Thanks,
>--Satya
>
>
>
>On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
>
>>Hi Thomas,
>>I know you are neutral, sorry for my reply vulnerable to be miunderstood.
>>
>>Thanks,
>>weiguo
>>
>>________________________________________
>>From: thomas.morin@orange.com [thomas.morin@orange.com]
>>Sent: Friday, March 27, 2015 11:59
>>To: Haoweiguo; bess@ietf.org
>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>algorithm
>>
>>Hi Weiguo,
>>
>>2015-03-26, Haoweiguo:
>> > Thomas:
>>> > - RFC7432 may have transient periods where the DF election state is
>>> > not yet synchronized between the two peers:
>> >
>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic loop
>>
>>Note well that I didn't write that there can be transient _loops_ .
>>I tried to capture the exchange you had all, and what I gather is that
>>the split-horizon procedure _prevents_loops_, independently of any
>>transient period where DF state is not synchronized and which may lead
>>to transient _duplicates_.
>>
>>-Thomas
>>
>>>
>>> 26/03/2015 20:05, John E Drake :
>>>>
>>>> Weiguo,
>>>>
>>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I have
>>>> detailed, is a non-solution to a non-problem with tremendous control
>>>> plane cost.
>>>>
>>>> Yours Irrespectively,
>>>>
>>>> John
>>>>
>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> Pls see below.
>>>>
>>>> Thanks,
>>>>
>>>> weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>> *From:*John E Drake [jdrake@juniper.net]
>>>> *Sent:* Friday, March 27, 2015 6:00
>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> To recap,
>>>>
>>>> We have established that your proposal is untenable because of its
>>>> control plane load.
>>>>
>>>> We have established that your proposal is based upon a flawed
>>>> understanding of the DF election in RFC 7432.
>>>>
>>>> [weiguo]: In ethernet world, traffic loop is serious than short timer
>>>> traffic disruption. If you want to implement  transiet traffic loop
>>>> process, i will modify my draft to solve your issue.
>>>>
>>>> If i am the developer, i will prefer short timer traffic disruption
>>>> based on current EVPN protocol.
>>>>
>>>> What you are now arguing is that your draft prevents two or more PEs
>>>> from being DF simultaneously. This is clearly nonsense.
>>>>
>>>> [weiguo]: I will modify the draft problem statements, and use the same
>>>> handshaking solution to solve it.
>>>>
>>>> Furthermore, we have established that having two or more DFs for what
>>>> even you admit is a brief transient leads to duplicate traffic, which
>>>> is acceptable, but not loops, your assertion to the contrary.
>>>>
>>>> [weiguo]: It is transient loop and traffic duplication issue.
>>>>
>>>> Yours Irrespectively,
>>>>
>>>> John
>>>>
>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> John,
>>>>
>>>> As your understanding of the EVPN draft, the DF election mechanism has
>>>> more serious side effect, it will have short time traffic loop,i.e.,
>>>> dual DF PEs will exist for a short time. I think dual DF PEs is
>>>> absolutely not tolerated, because native ethernet header has no TTL,
>>>> up to several hundred ms traffic loop normally not tolerated in
>>>> commertial networks.
>>>>
>>>> As your understanding, the PEs should do as following:
>>>>
>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have
>>>> rarely deployment.
>>>>
>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins
>>>> ESI and trigger DF re-election. When reception timer expires:
>>>>
>>>> PE1 upgrades to DF PE.
>>>>
>>>> After reception timer+ ES route transmission timer:
>>>>
>>>> PE2 downloads to non-DF PE.
>>>>
>>>> So in timer clock sync case, dual DF PEs will exist at least
>>>> transmission timer.
>>>>
>>>> If NTP is used for timer sync, because it has bad accuracy, dual DF
>>>> PEs will exist more longer timer.
>>>>
>>>> So as your understanding for DF election, the drawback is more clear.
>>>>
>>>> Thanks,
>>>>
>>>> weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>> *From:*John E Drake [jdrake@juniper.net]
>>>> *Sent:* Friday, March 27, 2015 4:41
>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> Weiguo,
>>>>
>>>> We have already established that your proposal is untenable because of
>>>> its control plane load.
>>>>
>>>> What we are now discussing is that your proposal is based upon a
>>>> misunderstanding of the algorithm in RFC 7432.  You are assuming that
>>>> PE1 will advertise an ES route and then wait for the configured
>>>> interval before performing the DF election while PE2 and PE3 will
>>>> perform the DF election as soon as they receive the ES route from PE1.
>>>> This is not what RFC 7432 says.
>>>>
>>>> Rather, what is says is that the advertisement of the ES route by PE1
>>>> and its receipt by PE2 and PE3 causes all three PEs to start the
>>>> configured interval timer  - =B33. When the timer expires, each PE
>>>> builds an ordered list of the IP addresses of all the PE nodes
>>>> connected to the Ethernet segment (including itself), in increasing
>>>> numeric value.=B2
>>>>
>>>> Yours Irrespectively,
>>>>
>>>> John
>>>>
>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> Pls read my detail replies to Satya. If you still can't catch it, pls
>>>> read my draft and EVPN base protocol,  thanks
>>>>
>>>> weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>> *From:*John E Drake [jdrake@juniper.net]
>>>> *Sent:* Friday, March 27, 2015 1:28
>>>> *To:* Patrice Brissette (pbrisset)
>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>> <mailto:bess@ietf.org>
>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> I think Patrice is correct.  Your proposal doesn't solve the problem
>>>> and it does so at huge cost.
>>>>
>>>> Sent from my iPhone
>>>>
>>>>
>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>>>
>>>>      Weiguo,
>>>>
>>>>      I=B9m not sure I=B9m following here.
>>>>
>>>>      Don=B9t you have the same issue with your handshaking mechanism?
>>>>
>>>>      If you don=B9t know your peer, how can you handshake?
>>>>
>>>>      Regards,
>>>>
>>>>      Patrice
>>>>
>>>>      Image removed by sender.
>>>>
>>>>      *Patrice Brissette*
>>>>      TECHNICAL LEADER.ENGINEERING
>>>>
>>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>      Phone: *+1 613 254 3336*
>>>>
>>>>
>>>>
>>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>      Canada
>>>>      Cisco.com <http://www.cisco.com/global/CA/>
>>>>
>>>>
>>>>
>>>>      Image removed by sender.Think before you print.
>>>>
>>>>      This email may contain confidential and privileged material for
>>>>      the sole use of the intended recipient. Any review, use,
>>>>      distribution or disclosure by others is strictly prohibited. If
>>>>      you are not the intended recipient (or authorized to receive for
>>>>      the recipient), please contact the sender by reply email and
>>>>      delete all copies of this message.
>>>>
>>>>      Please click here
>>>>
>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>>      for Company Registration Information.
>>>>
>>>>
>>>>      *From: *Haoweiguo <haoweiguo@huawei.com
>>>><mailto:haoweiguo@huawei.com>>
>>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
>>>>      *To: *Patrice Brissette <pbrisset@cisco.com
>>>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES based
>>>>on
>>>>      Paxos algorithm
>>>>
>>>>          Hi Patrice,
>>>>
>>>>          Up to reception timer traffic disruption in transient phase
>>>>is
>>>>          one of the issues.
>>>>
>>>>          Thanks,
>>>>
>>>>          weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>>>          <mailto:pbrisset@cisco.com>]
>>>>          *Sent:* Thursday, March 26, 2015 20:54
>>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>>>          bess@ietf.org <mailto:bess@ietf.org>
>>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>>          based on Paxos algorithm
>>>>
>>>>          Weiguo,
>>>>
>>>>          You mention "But if your draft have not solved all issues=B2,
>>>>
>>>>          Can you explain what Satya=B9s draft is not solving?
>>>>
>>>>          Regards,
>>>>
>>>>          Patrice
>>>>
>>>>          Image removed by sender.
>>>>
>>>>          *Patrice Brissette*
>>>>          TECHNICAL LEADER.ENGINEERING
>>>>
>>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>          Phone: *+1 613 254 3336*
>>>>
>>>>
>>>>
>>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>          Canada
>>>>          Cisco.com <http://www.cisco.com/global/CA/>
>>>>
>>>>
>>>>
>>>>          Image removed by sender.Think before you print.
>>>>
>>>>          This email may contain confidential and privileged material
>>>>          for the sole use of the intended recipient. Any review, use,
>>>>          distribution or disclosure by others is strictly prohibited.
>>>>          If you are not the intended recipient (or authorized to
>>>>          receive for the recipient), please contact the sender by
>>>>reply
>>>>          email and delete all copies of this message.
>>>>
>>>>          Please click here
>>>>
>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>>          for Company Registration Information.
>>>>
>>>>
>>>>          *From: *Haoweiguo <haoweiguo@huawei.com
>>>>          <mailto:haoweiguo@huawei.com>>
>>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>>>          *To: *John E Drake <jdrake@juniper.net
>>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>          <mailto:bess@ietf.org>" <bess@ietf.org
>>>><mailto:bess@ietf.org>>
>>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>>>          based on Paxos algorithm
>>>>
>>>>              Hi John,
>>>>
>>>>               Firstly i think EVPN community should reach consensus on
>>>>              the issues of current DF election mechanism. All these
>>>>              issues should be resolved in a single new DF election
>>>>              draft,rather than in multiple separate drafts. If your
>>>>              draft can solve all these issues and stable, i have no
>>>>              question for its progressing. But if your draft have not
>>>>              solved all issues, i think it had better combine with
>>>>              other drafts to provide a comprehensive solution. I think
>>>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01 and
>>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it should
>>>>              be resolved. So i think although your new Hash algorithm
>>>>              for DF election is good, it only includes partial
>>>>              enhancements, maybe it still needs some time for
>>>>consensus.
>>>>
>>>>              Thanks,
>>>>
>>>>              weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>>              *From:*John E Drake [jdrake@juniper.net
>>>>              <mailto:jdrake@juniper.net>]
>>>>              *Sent:* Thursday, March 26, 2015 7:16
>>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>              <mailto:bess@ietf.org>
>>>>              *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES
>>>>              based on Paxos algorithm
>>>>
>>>>              Weiguo,
>>>>
>>>>              Your proposal introduces a control plane processing load
>>>>              that is O(#EVIs * PEs) per DF election and given that
>>>>              there can be 4K EVIs per ES, this looks like a
>>>>              **substantial** load. Furthermore,
>>>>
>>>>              you can=B9t  use the ES route to co-ordinate DF election
>>>>              because you would need to carry your new extended
>>>>              community for each EVI and they would not all fit.  You
>>>>              also can=B9t use the Per EVI Ethernet AD route because th=
at
>>>>              is processed by all PEs in the EVI.
>>>>
>>>>              I think that from a practical perspective the new DF
>>>>              election proposed in Satya=B9s draft is sufficiently stab=
le
>>>>              that it renders your draft moot, even if it could be made
>>>>              to work.
>>>>
>>>>              Yours Irrespectively,
>>>>
>>>>              John
>>>>
>>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf Of
>>>>              *Haoweiguo
>>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
>>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
>>>>              <mailto:bess@ietf.org>
>>>>              *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>>              based on Paxos algorithm
>>>>
>>>>              Hi Ali,
>>>>
>>>>              Thanks for your information. I scanned through this
>>>>draft,
>>>>              it really introduces inter-chassis message for DF
>>>>election
>>>>              handshaking, the requirements in this draft is to
>>>>              eliminate transiet Loop and traffic duplication. Current
>>>>              EVPN DF election mechanism already eliminated loop and
>>>>              traffic duplication by configuring long reception timer
>>>>on
>>>>              each multi-homed PE, but up to reception timer traffic
>>>>              disruption issue still exist. EVPN for DCI is an
>>>>important
>>>>              use case for EVPN, up to reception timer traffic
>>>>              disruption can't be tolerated for service providers, it
>>>>              should be improved.
>>>>
>>>>              Also for accuracy, i think handshaking state machine on
>>>>              each multi-homed PE is also needed. From solution
>>>>              perspective, in my draft, no inter-chassis message is
>>>>              introduced, only one new extended community is
>>>>introduced,
>>>>              i think the process is comparatively simple than your
>>>>              following draft.
>>>>
>>>>              Current EVPN DF election has some drawbacks, so there are
>>>>              three new drafts about DF election emerged. I think BESS
>>>>              WG can consider these three drafts in global view, a
>>>>              single,comprehensive new DF election draft is hoped.
>>>>
>>>>              thanks.
>>>>
>>>>              Thanks,
>>>>
>>>>              weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>>              *From:*BESS [bess-bounces@ietf.org
>>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajassi
>>>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>]
>>>>              *Sent:* Thursday, March 26, 2015 0:21
>>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
>>>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>>>              based on Paxos algorithm
>>>>
>>>>              FYI- First published July 4, 2011
>>>>
>>>>
>>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route
>>>>/
>>>>
>>>>              -Ali
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> BESS mailing list
>>>> BESS@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/bess
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>________________________________________________________________________
>>>_
>>>________________________________________________
>>>
>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>confidentielles ou privilegiees et ne doivent donc
>>> pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>>>recu ce message par erreur, veuillez le signaler
>>> a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>>messages electroniques etant susceptibles d'alteration,
>>> Orange decline toute responsabilite si ce message a ete altere, deforme
>>>ou falsifie. Merci.
>>>
>>> This message and its attachments may contain confidential or privileged
>>>information that may be protected by law;
>>> they should not be distributed, used or copied without authorisation.
>>> If you have received this email in error, please notify the sender and
>>>delete this message and its attachments.
>>> As emails may be altered, Orange is not liable for messages that have
>>>been modified, changed or falsified.
>>> Thank you.
>>>
>>> _______________________________________________
>>> BESS mailing list
>>> BESS@ietf.org
>>> https://www.ietf.org/mailman/listinfo/bess
>>>
>>
>>
>>_________________________________________________________________________
>>_
>>_______________________________________________
>>
>>Ce message et ses pieces jointes peuvent contenir des informations
>>confidentielles ou privilegiees et ne doivent donc
>>pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>>recu ce message par erreur, veuillez le signaler
>>a l'expediteur et le detruire ainsi que les pieces jointes. Les messages
>>electroniques etant susceptibles d'alteration,
>>Orange decline toute responsabilite si ce message a ete altere, deforme
>>ou falsifie. Merci.
>>
>>This message and its attachments may contain confidential or privileged
>>information that may be protected by law;
>>they should not be distributed, used or copied without authorisation.
>>If you have received this email in error, please notify the sender and
>>delete this message and its attachments.
>>As emails may be altered, Orange is not liable for messages that have
>>been modified, changed or falsified.
>>Thank you.
>>_______________________________________________
>>BESS mailing list
>>BESS@ietf.org
>>https://www.ietf.org/mailman/listinfo/bess
>_______________________________________________
>BESS mailing list
>BESS@ietf.org
>https://www.ietf.org/mailman/listinfo/bess=


From nobody Fri Mar 27 07:40:33 2015
Return-Path: <jdrake@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3F77D1ACE84 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:40:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.302
X-Spam-Level: 
X-Spam-Status: No, score=-1.302 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id aAVhIUSWANh4 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:40:26 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1on0750.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::750]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 25F871ACE8F for <bess@ietf.org>; Fri, 27 Mar 2015 07:40:26 -0700 (PDT)
Received: from BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) by BLUPR05MB564.namprd05.prod.outlook.com (10.141.202.150) with Microsoft SMTP Server (TLS) id 15.1.118.21; Fri, 27 Mar 2015 14:40:09 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) by BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) with mapi id 15.01.0118.022; Fri, 27 Mar 2015 14:40:09 +0000
From: John E Drake <jdrake@juniper.net>
To: "EXT - thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8P/o146fJTTukSNC2GolBinQZ0u4lsAgAACyoCAAB/gjYAAILMAgAAMEtCAABidAIAAA48AgAAHqACAAC468IAAHjiAgADGMpA=
Date: Fri, 27 Mar 2015 14:40:09 +0000
Message-ID: <BLUPR05MB5625939AF8D75B6B1724447C7090@BLUPR05MB562.namprd05.prod.outlook.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8571E7@nkgeml501-mbs.china.huawei.com> <BLUPR05MB5629642F4A86ED28469DF19C70B0@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com>, <D13990E2.4EE3E%pbrisset@cisco.com>, <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com>
In-Reply-To: <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [66.129.239.10]
authentication-results: orange.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BLUPR05MB564;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(479174004)(24454002)(164054003)(69234005)(51704005)(36304003)(13464003)(377454003)(62966003)(77156002)(122556002)(106116001)(93886004)(50986999)(561944003)(76576001)(92566002)(86362001)(33656002)(74316001)(102836002)(15975445007)(107886001)(2501003)(46102003)(2950100001)(54356999)(19580395003)(2900100001)(19580405001)(99286002)(76176999)(66066001)(2656002)(87936001)(5890100001)(77096005); DIR:OUT; SFP:1102; SCL:1; SRVR:BLUPR05MB564; H:BLUPR05MB562.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BLUPR05MB564F85913DD448D4D90C334C7090@BLUPR05MB564.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BLUPR05MB564; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB564; 
x-forefront-prvs: 0528942FD8
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-OriginatorOrg: juniper.net
X-MS-Exchange-CrossTenant-originalarrivaltime: 27 Mar 2015 14:40:09.2074 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BLUPR05MB564
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/977dUb_CnrrwaycRRxfwcGMaIHw>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 14:40:30 -0000
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From nobody Fri Mar 27 07:46:04 2015
Return-Path: <kishoret@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 877E61ACE9C for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:46:02 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.301
X-Spam-Level: 
X-Spam-Status: No, score=-1.301 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_56=0.6, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LNKniUz-BrmX for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:45:56 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1bon0741.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::1:741]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2F0CF1ACEAE for <bess@ietf.org>; Fri, 27 Mar 2015 07:45:51 -0700 (PDT)
Received: from BY2PR05MB678.namprd05.prod.outlook.com (10.141.221.147) by BY2PR05MB679.namprd05.prod.outlook.com (10.141.221.150) with Microsoft SMTP Server (TLS) id 15.1.118.21; Fri, 27 Mar 2015 14:45:30 +0000
Received: from BY2PR05MB678.namprd05.prod.outlook.com ([10.141.221.147]) by BY2PR05MB678.namprd05.prod.outlook.com ([10.141.221.147]) with mapi id 15.01.0118.022; Fri, 27 Mar 2015 14:45:30 +0000
From: Kishore Tiruveedhula <kishoret@juniper.net>
To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, Haoweiguo <haoweiguo@huawei.com>, "EXT - thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ1HXGRQ8q1/wykW+c8ba93j8wp0uHgMAgACbTwCAACntAIAAAsqAgAAf4ACAACCzAIAAFSOAgAAPjACAAAZ+AIAABLkAgAAvJQCAAB1OgIAAD/cAgAADPACAAAKbAIAAAO2AgAAMO4CAAATFgIAATC0A
Date: Fri, 27 Mar 2015 14:45:29 +0000
Message-ID: <D13AD4DB.2527B%kishoret@juniper.net>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A32A3.2CC08%satyamoh@cisco.com>
In-Reply-To: <D13A32A3.2CC08%satyamoh@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-ms-exchange-messagesentrepresentingtype: 1
x-originating-ip: [66.129.241.13]
authentication-results: cisco.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BY2PR05MB679;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(479174004)(69234005)(377454003)(24454002)(36304003)(377424004)(164054003)(99286002)(561944003)(122556002)(106116001)(87936001)(16236675004)(2656002)(66066001)(36756003)(93886004)(62966003)(86362001)(50986999)(77156002)(19580405001)(83506001)(19580395003)(107886001)(2950100001)(76176999)(92566002)(46102003)(2900100001)(2501003)(15975445007)(102836002)(5890100001)(19617315012)(54356999)(77096005)(579004); DIR:OUT; SFP:1102; SCL:1; SRVR:BY2PR05MB679; H:BY2PR05MB678.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BY2PR05MB6796C0E1728B8B448337683C1090@BY2PR05MB679.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BY2PR05MB679; BCL:0; PCL:0; RULEID:; SRVR:BY2PR05MB679; 
x-forefront-prvs: 0528942FD8
Content-Type: multipart/alternative; boundary="_000_D13AD4DB2527Bkishoretjunipernet_"
MIME-Version: 1.0
X-OriginatorOrg: juniper.net
X-MS-Exchange-CrossTenant-originalarrivaltime: 27 Mar 2015 14:45:29.4988 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BY2PR05MB679
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/Uunxv0frLgqCduk04wxvjUVwTNk>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 14:46:02 -0000

--_000_D13AD4DB2527Bkishoretjunipernet_
Content-Type: text/plain; charset="euc-kr"
Content-Transfer-Encoding: base64
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--_000_D13AD4DB2527Bkishoretjunipernet_
Content-Type: text/html; charset="euc-kr"
Content-ID: <4A3EFF9E123F72458399CD4AACC167AD@namprd05.prod.outlook.com>
Content-Transfer-Encoding: base64

PGh0bWw+DQo8aGVhZD4NCjxtZXRhIGh0dHAtZXF1aXY9IkNvbnRlbnQtVHlwZSIgY29udGVudD0i
dGV4dC9odG1sOyBjaGFyc2V0PWV1Yy1rciI+DQo8L2hlYWQ+DQo8Ym9keSBzdHlsZT0id29yZC13
cmFwOiBicmVhay13b3JkOyAtd2Via2l0LW5ic3AtbW9kZTogc3BhY2U7IC13ZWJraXQtbGluZS1i
cmVhazogYWZ0ZXItd2hpdGUtc3BhY2U7IGNvbG9yOiByZ2IoMCwgMCwgMCk7IGZvbnQtc2l6ZTog
MTRweDsgZm9udC1mYW1pbHk6IENhbGlicmksIHNhbnMtc2VyaWY7Ij4NCjxkaXY+SSBhZ3JlZSB3
aXRoIFNhdHlhLiAmbmJzcDtJZiB5b3UgY29tcHV0ZSB0aGUgREYgcm9sZSBhbmQgYWN0IHVwb24g
anVzdCBpbW1lZGlhdGx5IGFmdGVyIHJlY2VpdmluZyB0aGUgRVNJIHdpdGhkcmF3L2FkdmVydGlz
ZW1lbnQsIHlvdSBjYW4gYXZvaWQgdGhlIHN0YXRlIG9mIGJvdGggUEVzIGluIERGIG9yIE5vbi1E
Ri48L2Rpdj4NCjxkaXY+SnVzdCBhZnRlciBsb2NhbCBFU0kgZGlzY292ZXJ5LCBrZWVwIHRoZSBF
U0kgaW4gdGhlIE5vbi1ERiBzdGF0ZSwgYWR2ZXJ0aXNlIEVTSSByb3V0ZSBhbmQgc3RhcnQgdGhl
IDMgc2Vjb25kcyB0aGUgREYgZWxlY3Rpb24gdGltZXIuIEp1c3QgYWZ0ZXIgcmVjZWl2aW5nIHRo
ZSBFU0kgcm91dGUgZnJvbSB0aGUgcmVtb3RlIFBFLCBqdXN0IGNvbXB1dGUgdGhlIERGIHJvbGUg
aW1tZWRpYXRlbHkgYW5kIG5vIG5lZWQgdG8gd2FpdCBmb3IgMyBzZWNvbmRzLg0KIFRoZSAzIHNl
Y29uZHMgaXMgdGhlIG1heCB3YWl0IHRpbWUuIFlvdSBjb3VsZCBjaGFuZ2UgdGhpcyB0aW1lciB2
YWx1ZSBpZiB0aGUgcmVjZWl2aW5nIHRoZSBFU0kgcm91dGUgZnJvbSByZW1vdGUgUEUgdGFrZXMg
bG9uZ2VyIHRoYW4gMyBzZWNvbmRzLjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+VGhh
bmtzLDwvZGl2Pg0KPGRpdj5LaXNob3JlJm5ic3A7PC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0K
PHNwYW4gaWQ9Ik9MS19TUkNfQk9EWV9TRUNUSU9OIj4NCjxkaXYgc3R5bGU9ImZvbnQtZmFtaWx5
OkNhbGlicmk7IGZvbnQtc2l6ZToxMXB0OyB0ZXh0LWFsaWduOmxlZnQ7IGNvbG9yOmJsYWNrOyBC
T1JERVItQk9UVE9NOiBtZWRpdW0gbm9uZTsgQk9SREVSLUxFRlQ6IG1lZGl1bSBub25lOyBQQURE
SU5HLUJPVFRPTTogMGluOyBQQURESU5HLUxFRlQ6IDBpbjsgUEFERElORy1SSUdIVDogMGluOyBC
T1JERVItVE9QOiAjYjVjNGRmIDFwdCBzb2xpZDsgQk9SREVSLVJJR0hUOiBtZWRpdW0gbm9uZTsg
UEFERElORy1UT1A6IDNwdCI+DQo8c3BhbiBzdHlsZT0iZm9udC13ZWlnaHQ6Ym9sZCI+RnJvbTog
PC9zcGFuPiZxdW90O1NhdHlhIE1vaGFudHkgKHNhdHlhbW9oKSZxdW90OyAmbHQ7PGEgaHJlZj0i
bWFpbHRvOnNhdHlhbW9oQGNpc2NvLmNvbSI+c2F0eWFtb2hAY2lzY28uY29tPC9hPiZndDs8YnI+
DQo8c3BhbiBzdHlsZT0iZm9udC13ZWlnaHQ6Ym9sZCI+RGF0ZTogPC9zcGFuPkZyaWRheSwgTWFy
Y2ggMjcsIDIwMTUgYXQgMTI6MTIgQU08YnI+DQo8c3BhbiBzdHlsZT0iZm9udC13ZWlnaHQ6Ym9s
ZCI+VG86IDwvc3Bhbj5IYW93ZWlndW8gJmx0OzxhIGhyZWY9Im1haWx0bzpoYW93ZWlndW9AaHVh
d2VpLmNvbSI+aGFvd2VpZ3VvQGh1YXdlaS5jb208L2E+Jmd0OywgJnF1b3Q7RVhUIC0NCjxhIGhy
ZWY9Im1haWx0bzp0aG9tYXMubW9yaW5Ab3JhbmdlLmNvbSI+dGhvbWFzLm1vcmluQG9yYW5nZS5j
b208L2E+JnF1b3Q7ICZsdDs8YSBocmVmPSJtYWlsdG86dGhvbWFzLm1vcmluQG9yYW5nZS5jb20i
PnRob21hcy5tb3JpbkBvcmFuZ2UuY29tPC9hPiZndDssICZxdW90OzxhIGhyZWY9Im1haWx0bzpi
ZXNzQGlldGYub3JnIj5iZXNzQGlldGYub3JnPC9hPiZxdW90OyAmbHQ7PGEgaHJlZj0ibWFpbHRv
OmJlc3NAaWV0Zi5vcmciPmJlc3NAaWV0Zi5vcmc8L2E+Jmd0Ozxicj4NCjxzcGFuIHN0eWxlPSJm
b250LXdlaWdodDpib2xkIj5TdWJqZWN0OiA8L3NwYW4+UmU6IFtiZXNzXSBIYW5kc2hha2luZyBh
bW9uZyBQRXMgaW4gYW4gRVZQTiBFUyBiYXNlZCBvbiBQYXhvcyBhbGdvcml0aG08YnI+DQo8L2Rp
dj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pg0KPGRpdiBzdHlsZT0id29yZC13cmFwOiBicmVh
ay13b3JkOyAtd2Via2l0LW5ic3AtbW9kZTogc3BhY2U7IC13ZWJraXQtbGluZS1icmVhazogYWZ0
ZXItd2hpdGUtc3BhY2U7IGZvbnQtZmFtaWx5OiBDYWxpYnJpLCBzYW5zLXNlcmlmOyBmb250LXNp
emU6IDE0cHg7ICI+DQo8ZGl2IHN0eWxlPSJjb2xvcjogcmdiKDAsIDAsIDApOyAiPkhpIEhhb3dl
aWd1byw8L2Rpdj4NCjxkaXYgc3R5bGU9ImNvbG9yOiByZ2IoMCwgMCwgMCk7ICI+PGJyPg0KPC9k
aXY+DQo8ZGl2IHN0eWxlPSJjb2xvcjogcmdiKDAsIDAsIDApOyAiPk9uIDMvMjYvMTUgOTo1NSBQ
TSwgJnF1b3Q7SGFvd2VpZ3VvJnF1b3Q7ICZsdDs8YSBocmVmPSJtYWlsdG86aGFvd2VpZ3VvQGh1
YXdlaS5jb20iPmhhb3dlaWd1b0BodWF3ZWkuY29tPC9hPiZndDsgd3JvdGU6PC9kaXY+DQo8ZGl2
IHN0eWxlPSJjb2xvcjogcmdiKDAsIDAsIDApOyAiPjxicj4NCjwvZGl2Pg0KPGJsb2NrcXVvdGUg
aWQ9Ik1BQ19PVVRMT09LX0FUVFJJQlVUSU9OX0JMT0NLUVVPVEUiIHN0eWxlPSJjb2xvcjogcmdi
KDAsIDAsIDApOyBib3JkZXItbGVmdC1jb2xvcjogcmdiKDE4MSwgMTk2LCAyMjMpOyBib3JkZXIt
bGVmdC13aWR0aDogNXB4OyBib3JkZXItbGVmdC1zdHlsZTogc29saWQ7IHBhZGRpbmc6IDBweCAw
cHggMHB4IDVweDsgbWFyZ2luOiAwcHggMHB4IDBweCA1cHg7ICI+DQo8ZGl2PkhpIFNhdHlhLDwv
ZGl2Pg0KPGRpdj5QbHMgc2VlIGlubGluZS48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2
PlRoYW5rcyw8L2Rpdj4NCjxkaXY+d2VpZ3VvPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRp
dj5fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fPC9kaXY+DQo8ZGl2PkZy
b206IFNhdHlhIE1vaGFudHkgKHNhdHlhbW9oKSBbPGEgaHJlZj0ibWFpbHRvOnNhdHlhbW9oQGNp
c2NvLmNvbSI+c2F0eWFtb2hAY2lzY28uY29tPC9hPl08L2Rpdj4NCjxkaXY+U2VudDogRnJpZGF5
LCBNYXJjaCAyNywgMjAxNSAxMjoxMTwvZGl2Pg0KPGRpdj5UbzogSGFvd2VpZ3VvOyA8YSBocmVm
PSJtYWlsdG86dGhvbWFzLm1vcmluQG9yYW5nZS5jb20iPnRob21hcy5tb3JpbkBvcmFuZ2UuY29t
PC9hPjsNCjxhIGhyZWY9Im1haWx0bzpiZXNzQGlldGYub3JnIj5iZXNzQGlldGYub3JnPC9hPjwv
ZGl2Pg0KPGRpdj5TdWJqZWN0OiBSZTogW2Jlc3NdIEhhbmRzaGFraW5nIGFtb25nIFBFcyBpbiBh
biBFVlBOIEVTIGJhc2VkIG9uIFBheG9zIGFsZ29yaXRobTwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rp
dj4NCjxkaXY+VGhvbWFzLDwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+VGhhbmtzIGZv
ciB5b3VyIHN1bW1hcnkuIEl0IGNhcHR1cmVzIHRoZSBlc3NlbmNlIG9mIHRoZSBkaXNjdXNzaW9u
LjwvZGl2Pg0KPGRpdj5bd2VpZ3VvXTogWWVzLCBpIGFncmVlLjwvZGl2Pg0KPGRpdj48YnI+DQo8
L2Rpdj4NCjxkaXY+T25lIGZlYXR1cmUgb2YgdGhlIEhSVyBzY2hlbWUgaXMgdGhhdCBpdCBjb21w
dXRlcyB0aGUgQWN0aXZlIGFuZCB0aGU8L2Rpdj4NCjxkaXY+QmFja3VwIERGIHVwZnJvbnQuPC9k
aXY+DQo8ZGl2PldoZW5ldmVyIHRoZSBBY3RpdmUgUEUsIHNheSBQRTEsIGRpZXM7IDwvZGl2Pg0K
PGRpdj50aGUgQmFja3VwIFBFLCBQRTIsIGNhbiBiZWNvbWUgdGhlPC9kaXY+DQo8ZGl2PkFjdGl2
ZSBpbW1lZGlhdGVseSB1cG9uIHJlY2VpdmluZyB0aGUgd2l0aGRyYXcgb2YgdGhlIEVTIHJvdXRl
IGZyb20gUEUxLjwvZGl2Pg0KPGRpdj5JdCBuZWVkIG5vdCBldmVuIGhhdmUgdG8gd2FpdCBmb3Ig
dGhlIDMgc2Vjb25kcyB0aW1lLjwvZGl2Pg0KPGRpdj5bd2VpZ3VvXTogSSB0aGluayBtYXliZSB5
b3VyIG5lZ2xpZ2VuY2UsIGRpZXMgUEUxIGNhbid0IHNlbmQgd2l0aGRyYXcgbWVzc2FnZTopIEJ1
dCB3aGVuIFBFMSBsZWF2ZXMgRVNJLCBpdCBjYW4gd2l0aGRyYXcgdGhlIEVTIHJvdXRlIGltbWVk
aWF0ZWx5LjwvZGl2Pg0KPGRpdj5XaGVuIFBFMSBkaWVzIGFuZCByZWNvdmVycywgYXNzdW1pbmcg
UEUxIGlzIERGIFBFLCB0aGVuIFBFMSBhbmQgUEUyIHdpbGwgYm90aCBhY3QgYXMgUEUgZm9yIHNv
bWUgdHJhbnNpZXQgdGltZS48L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjxkaXYgc3R5bGU9ImNvbG9y
OiByZ2IoMCwgMCwgMCk7ICI+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxmb250IGNvbG9yPSIjMDAwMGZm
Ij5bU2F0eWFdIE5vdCByZWFsbHkgbmVnbGlnZW5jZS4gV2hhdCBJIG1lYW50IGlzIHRoZSBmb2xs
b3dpbmcgd2l0aG91dCBnZXR0aW5nIGludG8gZGV0YWlscy4gSW4gbW9zdCB0b3BvbG9naWVzIFBF
cyB3aWxsIGJlIGNvbm5lY3RlZCB2aWEgYW4gUlIuIFNvIHdoZW4gdGhlIFBFMSBkaWVzLCB0aGUg
cm91dGVzIHRoYXQgaGF2ZSBiZWVuIGFkdmVydGlzZWQgYnkgdGhlIFJSIHRvIG90aGVyIFBFcyB3
aWxsDQogYmUgd2l0aGRyYXduIGJ5IHRoZSBSUiBvbmNlIHRoZSBQRTEgc2Vzc2lvbiB3aXRoIFJS
IGlzIGJyb2tlbi4gQWx0ZXJuYXRpdmVseSwgaWYgdGhlcmUgaXMgbm8gUlIgYW5kIFBFMSBhbmQg
cEUyIGhhdmUgYSBwZWVyaW5nLCBQRTIgaXMgZ29pbmcgdG8gcHVyZ2UgdGhlIHJvdXRlcyBsZWFy
bnQgZnJvbSBQRTEgb25jZSB0aGUgaG9sZCB0aW1lciBleHBpcmVzLiBOZXQgZWZmZWN0IGFzIGZh
ciBhcyBERiBlbGVjdGlvbiBpbiBQRTIgaXMgY29uY2VybmVkDQogaXMgYSAmcXVvdDt3aXRoZHJh
dyZxdW90OyBvZiB0aGUgRVMgcm91dGUuPC9mb250PjwvZGl2Pg0KPGJsb2NrcXVvdGUgaWQ9Ik1B
Q19PVVRMT09LX0FUVFJJQlVUSU9OX0JMT0NLUVVPVEUiIHN0eWxlPSJjb2xvcjogcmdiKDAsIDAs
IDApOyBib3JkZXItbGVmdC1jb2xvcjogcmdiKDE4MSwgMTk2LCAyMjMpOyBib3JkZXItbGVmdC13
aWR0aDogNXB4OyBib3JkZXItbGVmdC1zdHlsZTogc29saWQ7IHBhZGRpbmc6IDBweCAwcHggMHB4
IDVweDsgbWFyZ2luOiAwcHggMHB4IDBweCA1cHg7ICI+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRp
dj5PbiB0aGUgb3RoZXIgaGFuZCwgdGhlIDMgc2Vjb25kIHRpbWVyIHdpbGwgc2VydmUgYXMgYSBi
dWxrIGFuZCBiYXRjaDwvZGl2Pg0KPGRpdj5tZWNoYW5pc20gZm9yIHRoZSBjYXNlIHdoZW4gbmV3
IFBFKHMpIGlzKGFyZSkgY29taW5nIHVwLCBzbyB3ZSBkb26p9nQgaGF2ZTwvZGl2Pg0KPGRpdj50
byBkbyBhIERGIGVsZWN0aW9uIGV2ZXIgc28gb2Z0ZW4uPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2
Pg0KPGRpdj5IYW93ZWlndW8sJm5ic3A7Jm5ic3A7SSB0aGluayB0aGUgJnF1b3Q7bG9vcCZxdW90
OyBzY2VuYXJpbyBpcyBzcGVjdWxhdGl2ZS48L2Rpdj4NCjxkaXY+SSB0aGluayBldmVyeWJvZHkg
dW5kZXJzdGFuZHMgdGhhdCwgeWVzLCB0aGVyZSBtYXkgYmUgc29tZSB0cmFuc2llbnQ8L2Rpdj4N
CjxkaXY+ZHVwbGljYXRpb24gb2YgYnVtIHRyYWZmaWM7IGhvd2V2ZXIsIHRoZXJlIGlzIG5vICZx
dW90O2xvb3AmcXVvdDsgcHJvYmxlbSB0byBiZTwvZGl2Pg0KPGRpdj5zb2x2ZWQuPC9kaXY+DQo8
ZGl2Plt3ZWlndW9dOiBUcmFmZmljIGR1cGxpY2F0aW9uIGFuZCBsb29wIGlzc3VlIGJvdGggZXhp
c3QgaW4gZHVhbCBERiBQRSBjYXNlLiBJbiBtdWx0aS1ob21lZCBkZXZpY2UgY2FzZSwgc2F5IENF
MSBpcyBtdWx0aS1ob21lZCB0byBQRTEgYW5kIFBFMiBhbmQgYm90aCBQRSBhcmUgREYgUEUuIFdo
ZW4gQ0UxIHNlbmRzIHRyYWZmaWMgdG8gUEUxLCB0aGUgdHJhZmZpYyB3aWxsIGxvb3AgYmFjayB0
byBDRTEgdGhyb3VnaCBQRTIuIElmIENFMSBjYW4NCiBmaWx0ZXIgYW5kIGRyb3AgdGhlIHRyYWZm
aWMgLCBpdCBpcyBmb3J0dW5hdGUsIG1heWJlIG5vIGxvb3AuIDwvZGl2Pg0KPGRpdj5Gb3IgbXVs
dGktaG9tZWQgbmV0d29yayBzY2VuYXJpbywgc2F5IGxvY2FsIG5ldHdvcmsxKENFMSBhbmQgQ0Uy
KSBpcyBtdWx0aS1ob21lZCB0byBQRTEgYW5kIFBFMi4gU2ltaWxhcmx5LCB0aGUgdHJhZmZpYyBm
cm9tIENFMSB3aWxsIGxvb3AgYmFjayB0aHJvdWdoIFBFMS0tJmd0O1BFMi0tJmd0O0NFMiwgaXQg
aXMgdmVyeSBzZXJpb3VzLiBJdCBpcyBhYnNvbHV0ZWx5IG5vdCBhbGxvd2VkLjwvZGl2Pg0KPC9i
bG9ja3F1b3RlPg0KPGRpdj48Zm9udCBjb2xvcj0iIzAwMDBmZiI+W1NhdHlhXSBTaG91bGRuJ3Qg
dGhlIHNwbGl0LWhvcml6b24gcHJvY2VkdXJlIG5vdCBraWNrIGluIGhlcmU/PC9mb250PjwvZGl2
Pg0KPGJsb2NrcXVvdGUgaWQ9Ik1BQ19PVVRMT09LX0FUVFJJQlVUSU9OX0JMT0NLUVVPVEUiIHN0
eWxlPSJjb2xvcjogcmdiKDAsIDAsIDApOyBib3JkZXItbGVmdC1jb2xvcjogcmdiKDE4MSwgMTk2
LCAyMjMpOyBib3JkZXItbGVmdC13aWR0aDogNXB4OyBib3JkZXItbGVmdC1zdHlsZTogc29saWQ7
IHBhZGRpbmc6IDBweCAwcHggMHB4IDVweDsgbWFyZ2luOiAwcHggMHB4IDBweCA1cHg7ICI+DQo8
ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5UaGFua3MsPC9kaXY+DQo8ZGl2Pi0tU2F0eWE8L2Rpdj4N
CjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4N
CjxkaXY+T24gMy8yNi8xNSA5OjA4IFBNLCAmcXVvdDtIYW93ZWlndW8mcXVvdDsgJmx0OzxhIGhy
ZWY9Im1haWx0bzpoYW93ZWlndW9AaHVhd2VpLmNvbSI+aGFvd2VpZ3VvQGh1YXdlaS5jb208L2E+
Jmd0OyB3cm90ZTo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8YmxvY2txdW90ZSBpZD0iTUFD
X09VVExPT0tfQVRUUklCVVRJT05fQkxPQ0tRVU9URSIgc3R5bGU9IkJPUkRFUi1MRUZUOiAjYjVj
NGRmIDUgc29saWQ7IFBBRERJTkc6MCAwIDAgNTsgTUFSR0lOOjAgMCAwIDU7Ij4NCjxkaXY+SGkg
VGhvbWFzLDwvZGl2Pg0KPGRpdj5JIGtub3cgeW91IGFyZSBuZXV0cmFsLCBzb3JyeSBmb3IgbXkg
cmVwbHkgdnVsbmVyYWJsZSB0byBiZSBtaXVuZGVyc3Rvb2QuPC9kaXY+DQo8ZGl2Pjxicj4NCjwv
ZGl2Pg0KPGRpdj5UaGFua3MsPC9kaXY+DQo8ZGl2PndlaWd1bzwvZGl2Pg0KPGRpdj48YnI+DQo8
L2Rpdj4NCjxkaXY+X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXzwvZGl2
Pg0KPGRpdj5Gcm9tOiA8YSBocmVmPSJtYWlsdG86dGhvbWFzLm1vcmluQG9yYW5nZS5jb20iPnRo
b21hcy5tb3JpbkBvcmFuZ2UuY29tPC9hPiBbPGEgaHJlZj0ibWFpbHRvOnRob21hcy5tb3JpbkBv
cmFuZ2UuY29tIj50aG9tYXMubW9yaW5Ab3JhbmdlLmNvbTwvYT5dPC9kaXY+DQo8ZGl2PlNlbnQ6
IEZyaWRheSwgTWFyY2ggMjcsIDIwMTUgMTE6NTk8L2Rpdj4NCjxkaXY+VG86IEhhb3dlaWd1bzsg
PGEgaHJlZj0ibWFpbHRvOmJlc3NAaWV0Zi5vcmciPmJlc3NAaWV0Zi5vcmc8L2E+PC9kaXY+DQo8
ZGl2PlN1YmplY3Q6IFJlOiBbYmVzc10gSGFuZHNoYWtpbmcgYW1vbmcgUEVzIGluIGFuIEVWUE4g
RVMgYmFzZWQgb24gUGF4b3M8L2Rpdj4NCjxkaXY+YWxnb3JpdGhtPC9kaXY+DQo8ZGl2Pjxicj4N
CjwvZGl2Pg0KPGRpdj5IaSBXZWlndW8sPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4y
MDE1LTAzLTI2LCBIYW93ZWlndW86PC9kaXY+DQo8ZGl2PiZndDsgVGhvbWFzOjwvZGl2Pg0KPGJs
b2NrcXVvdGUgaWQ9Ik1BQ19PVVRMT09LX0FUVFJJQlVUSU9OX0JMT0NLUVVPVEUiIHN0eWxlPSJC
T1JERVItTEVGVDogI2I1YzRkZiA1IHNvbGlkOyBQQURESU5HOjAgMCAwIDU7IE1BUkdJTjowIDAg
MCA1OyI+DQo8ZGl2PiZndDsgLSBSRkM3NDMyIG1heSBoYXZlIHRyYW5zaWVudCBwZXJpb2RzIHdo
ZXJlIHRoZSBERiBlbGVjdGlvbiBzdGF0ZSBpczwvZGl2Pg0KPGRpdj4mZ3Q7IG5vdCB5ZXQgc3lu
Y2hyb25pemVkIGJldHdlZW4gdGhlIHR3byBwZWVyczo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjxk
aXY+Jmd0OzwvZGl2Pg0KPGJsb2NrcXVvdGUgaWQ9Ik1BQ19PVVRMT09LX0FUVFJJQlVUSU9OX0JM
T0NLUVVPVEUiIHN0eWxlPSJCT1JERVItTEVGVDogI2I1YzRkZiA1IHNvbGlkOyBQQURESU5HOjAg
MCAwIDU7IE1BUkdJTjowIDAgMCA1OyI+DQo8ZGl2Plt3ZWlndW9dOiBZZXMsIGkgdGhpbmsgUkZD
IDc0MzIgaGFzIHRyYW5zaWVudCBwZXJpb2RzIG9mIHRyYWZmaWMgbG9vcDwvZGl2Pg0KPC9ibG9j
a3F1b3RlPg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Tm90ZSB3ZWxsIHRoYXQgSSBkaWRuJ3Qg
d3JpdGUgdGhhdCB0aGVyZSBjYW4gYmUgdHJhbnNpZW50IF9sb29wc18gLjwvZGl2Pg0KPGRpdj5J
IHRyaWVkIHRvIGNhcHR1cmUgdGhlIGV4Y2hhbmdlIHlvdSBoYWQgYWxsLCBhbmQgd2hhdCBJIGdh
dGhlciBpcyB0aGF0PC9kaXY+DQo8ZGl2PnRoZSBzcGxpdC1ob3Jpem9uIHByb2NlZHVyZSBfcHJl
dmVudHNfbG9vcHNfLCBpbmRlcGVuZGVudGx5IG9mIGFueTwvZGl2Pg0KPGRpdj50cmFuc2llbnQg
cGVyaW9kIHdoZXJlIERGIHN0YXRlIGlzIG5vdCBzeW5jaHJvbml6ZWQgYW5kIHdoaWNoIG1heSBs
ZWFkPC9kaXY+DQo8ZGl2PnRvIHRyYW5zaWVudCBfZHVwbGljYXRlc18uPC9kaXY+DQo8ZGl2Pjxi
cj4NCjwvZGl2Pg0KPGRpdj4tVGhvbWFzPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGJsb2Nr
cXVvdGUgaWQ9Ik1BQ19PVVRMT09LX0FUVFJJQlVUSU9OX0JMT0NLUVVPVEUiIHN0eWxlPSJCT1JE
RVItTEVGVDogI2I1YzRkZiA1IHNvbGlkOyBQQURESU5HOjAgMCAwIDU7IE1BUkdJTjowIDAgMCA1
OyI+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4yNi8wMy8yMDE1IDIwOjA1LCBKb2huIEUgRHJh
a2UgOjwvZGl2Pg0KPGJsb2NrcXVvdGUgaWQ9Ik1BQ19PVVRMT09LX0FUVFJJQlVUSU9OX0JMT0NL
UVVPVEUiIHN0eWxlPSJCT1JERVItTEVGVDogI2I1YzRkZiA1IHNvbGlkOyBQQURESU5HOjAgMCAw
IDU7IE1BUkdJTjowIDAgMCA1OyI+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5XZWlndW8sPC9k
aXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5JIGd1ZXNzIEkgd2Fzbqn2dCBjbGVhci4mbmJz
cDsmbmJzcDtJIHRoaW5rIHlvdSBkcmFmdCwgZm9yIHRoZSByZWFzb25zIEkgaGF2ZTwvZGl2Pg0K
PGRpdj5kZXRhaWxlZCwgaXMgYSBub24tc29sdXRpb24gdG8gYSBub24tcHJvYmxlbSB3aXRoIHRy
ZW1lbmRvdXMgY29udHJvbDwvZGl2Pg0KPGRpdj5wbGFuZSBjb3N0LjwvZGl2Pg0KPGRpdj48YnI+
DQo8L2Rpdj4NCjxkaXY+WW91cnMgSXJyZXNwZWN0aXZlbHksPC9kaXY+DQo8ZGl2Pjxicj4NCjwv
ZGl2Pg0KPGRpdj5Kb2huPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4qRnJvbToqSGFv
d2VpZ3VvIFs8YSBocmVmPSJtYWlsdG86aGFvd2VpZ3VvQGh1YXdlaS5jb20iPm1haWx0bzpoYW93
ZWlndW9AaHVhd2VpLmNvbTwvYT5dPC9kaXY+DQo8ZGl2PipTZW50OiogVGh1cnNkYXksIE1hcmNo
IDI2LCAyMDE1IDY6MTcgUE08L2Rpdj4NCjxkaXY+KlRvOiogSm9obiBFIERyYWtlOyBQYXRyaWNl
IEJyaXNzZXR0ZSAocGJyaXNzZXQpPC9kaXY+DQo8ZGl2PipDYzoqIEFsaSBTYWphc3NpIChzYWph
c3NpKTsgPGEgaHJlZj0ibWFpbHRvOmJlc3NAaWV0Zi5vcmciPmJlc3NAaWV0Zi5vcmc8L2E+PC9k
aXY+DQo8ZGl2PipTdWJqZWN0OiogUkU6IFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQRXMgaW4g
YW4gRVZQTiBFUyBiYXNlZCBvbjwvZGl2Pg0KPGRpdj5QYXhvcyBhbGdvcml0aG08L2Rpdj4NCjxk
aXY+PGJyPg0KPC9kaXY+DQo8ZGl2PlBscyBzZWUgYmVsb3cuPC9kaXY+DQo8ZGl2Pjxicj4NCjwv
ZGl2Pg0KPGRpdj5UaGFua3MsPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj53ZWlndW88
L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t
LS0tLS0tLS08L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PipGcm9tOipKb2huIEUgRHJh
a2UgWzxhIGhyZWY9Im1haWx0bzpqZHJha2VAanVuaXBlci5uZXQiPmpkcmFrZUBqdW5pcGVyLm5l
dDwvYT5dPC9kaXY+DQo8ZGl2PipTZW50OiogRnJpZGF5LCBNYXJjaCAyNywgMjAxNSA2OjAwPC9k
aXY+DQo8ZGl2PipUbzoqIEhhb3dlaWd1bzsgUGF0cmljZSBCcmlzc2V0dGUgKHBicmlzc2V0KTwv
ZGl2Pg0KPGRpdj4qQ2M6KiBBbGkgU2FqYXNzaSAoc2FqYXNzaSk7IDxhIGhyZWY9Im1haWx0bzpi
ZXNzQGlldGYub3JnIj5iZXNzQGlldGYub3JnPC9hPiAmbHQ7PGEgaHJlZj0ibWFpbHRvOmJlc3NA
aWV0Zi5vcmciPm1haWx0bzpiZXNzQGlldGYub3JnPC9hPiZndDs8L2Rpdj4NCjxkaXY+KlN1Ympl
Y3Q6KiBSRTogW2Jlc3NdIEhhbmRzaGFraW5nIGFtb25nIFBFcyBpbiBhbiBFVlBOIEVTIGJhc2Vk
IG9uPC9kaXY+DQo8ZGl2PlBheG9zIGFsZ29yaXRobTwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4N
CjxkaXY+VG8gcmVjYXAsPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5XZSBoYXZlIGVz
dGFibGlzaGVkIHRoYXQgeW91ciBwcm9wb3NhbCBpcyB1bnRlbmFibGUgYmVjYXVzZSBvZiBpdHM8
L2Rpdj4NCjxkaXY+Y29udHJvbCBwbGFuZSBsb2FkLjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4N
CjxkaXY+V2UgaGF2ZSBlc3RhYmxpc2hlZCB0aGF0IHlvdXIgcHJvcG9zYWwgaXMgYmFzZWQgdXBv
biBhIGZsYXdlZDwvZGl2Pg0KPGRpdj51bmRlcnN0YW5kaW5nIG9mIHRoZSBERiBlbGVjdGlvbiBp
biBSRkMgNzQzMi48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Plt3ZWlndW9dOiBJbiBl
dGhlcm5ldCB3b3JsZCwgdHJhZmZpYyBsb29wIGlzIHNlcmlvdXMgdGhhbiBzaG9ydCB0aW1lcjwv
ZGl2Pg0KPGRpdj50cmFmZmljIGRpc3J1cHRpb24uIElmIHlvdSB3YW50IHRvIGltcGxlbWVudCZu
YnNwOyZuYnNwO3RyYW5zaWV0IHRyYWZmaWMgbG9vcDwvZGl2Pg0KPGRpdj5wcm9jZXNzLCBpIHdp
bGwgbW9kaWZ5IG15IGRyYWZ0IHRvIHNvbHZlIHlvdXIgaXNzdWUuPC9kaXY+DQo8ZGl2Pjxicj4N
CjwvZGl2Pg0KPGRpdj5JZiBpIGFtIHRoZSBkZXZlbG9wZXIsIGkgd2lsbCBwcmVmZXIgc2hvcnQg
dGltZXIgdHJhZmZpYyBkaXNydXB0aW9uPC9kaXY+DQo8ZGl2PmJhc2VkIG9uIGN1cnJlbnQgRVZQ
TiBwcm90b2NvbC48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PldoYXQgeW91IGFyZSBu
b3cgYXJndWluZyBpcyB0aGF0IHlvdXIgZHJhZnQgcHJldmVudHMgdHdvIG9yIG1vcmUgUEVzPC9k
aXY+DQo8ZGl2PmZyb20gYmVpbmcgREYgc2ltdWx0YW5lb3VzbHkuIFRoaXMgaXMgY2xlYXJseSBu
b25zZW5zZS48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Plt3ZWlndW9dOiBJIHdpbGwg
bW9kaWZ5IHRoZSBkcmFmdCBwcm9ibGVtIHN0YXRlbWVudHMsIGFuZCB1c2UgdGhlIHNhbWU8L2Rp
dj4NCjxkaXY+aGFuZHNoYWtpbmcgc29sdXRpb24gdG8gc29sdmUgaXQuPC9kaXY+DQo8ZGl2Pjxi
cj4NCjwvZGl2Pg0KPGRpdj5GdXJ0aGVybW9yZSwgd2UgaGF2ZSBlc3RhYmxpc2hlZCB0aGF0IGhh
dmluZyB0d28gb3IgbW9yZSBERnMgZm9yIHdoYXQ8L2Rpdj4NCjxkaXY+ZXZlbiB5b3UgYWRtaXQg
aXMgYSBicmllZiB0cmFuc2llbnQgbGVhZHMgdG8gZHVwbGljYXRlIHRyYWZmaWMsIHdoaWNoPC9k
aXY+DQo8ZGl2PmlzIGFjY2VwdGFibGUsIGJ1dCBub3QgbG9vcHMsIHlvdXIgYXNzZXJ0aW9uIHRv
IHRoZSBjb250cmFyeS48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Plt3ZWlndW9dOiBJ
dCBpcyB0cmFuc2llbnQgbG9vcCBhbmQgdHJhZmZpYyBkdXBsaWNhdGlvbiBpc3N1ZS48L2Rpdj4N
CjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PllvdXJzIElycmVzcGVjdGl2ZWx5LDwvZGl2Pg0KPGRp
dj48YnI+DQo8L2Rpdj4NCjxkaXY+Sm9objwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+
KkZyb206Kkhhb3dlaWd1byBbPGEgaHJlZj0ibWFpbHRvOmhhb3dlaWd1b0BodWF3ZWkuY29tIj5t
YWlsdG86aGFvd2VpZ3VvQGh1YXdlaS5jb208L2E+XTwvZGl2Pg0KPGRpdj4qU2VudDoqIFRodXJz
ZGF5LCBNYXJjaCAyNiwgMjAxNSA1OjM3IFBNPC9kaXY+DQo8ZGl2PipUbzoqIEpvaG4gRSBEcmFr
ZTsgUGF0cmljZSBCcmlzc2V0dGUgKHBicmlzc2V0KTwvZGl2Pg0KPGRpdj4qQ2M6KiBBbGkgU2Fq
YXNzaSAoc2FqYXNzaSk7IDxhIGhyZWY9Im1haWx0bzpiZXNzQGlldGYub3JnIj5iZXNzQGlldGYu
b3JnPC9hPiAmbHQ7PGEgaHJlZj0ibWFpbHRvOmJlc3NAaWV0Zi5vcmciPm1haWx0bzpiZXNzQGll
dGYub3JnPC9hPiZndDs8L2Rpdj4NCjxkaXY+KlN1YmplY3Q6KiBSRTogW2Jlc3NdIEhhbmRzaGFr
aW5nIGFtb25nIFBFcyBpbiBhbiBFVlBOIEVTIGJhc2VkIG9uPC9kaXY+DQo8ZGl2PlBheG9zIGFs
Z29yaXRobTwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Sm9obiw8L2Rpdj4NCjxkaXY+
PGJyPg0KPC9kaXY+DQo8ZGl2PkFzIHlvdXIgdW5kZXJzdGFuZGluZyBvZiB0aGUgRVZQTiBkcmFm
dCwgdGhlIERGIGVsZWN0aW9uIG1lY2hhbmlzbSBoYXM8L2Rpdj4NCjxkaXY+bW9yZSBzZXJpb3Vz
IHNpZGUgZWZmZWN0LCBpdCB3aWxsIGhhdmUgc2hvcnQgdGltZSB0cmFmZmljIGxvb3AsaS5lLiw8
L2Rpdj4NCjxkaXY+ZHVhbCBERiBQRXMgd2lsbCBleGlzdCBmb3IgYSBzaG9ydCB0aW1lLiBJIHRo
aW5rIGR1YWwgREYgUEVzIGlzPC9kaXY+DQo8ZGl2PmFic29sdXRlbHkgbm90IHRvbGVyYXRlZCwg
YmVjYXVzZSBuYXRpdmUgZXRoZXJuZXQgaGVhZGVyIGhhcyBubyBUVEwsPC9kaXY+DQo8ZGl2PnVw
IHRvIHNldmVyYWwgaHVuZHJlZCBtcyB0cmFmZmljIGxvb3Agbm9ybWFsbHkgbm90IHRvbGVyYXRl
ZCBpbjwvZGl2Pg0KPGRpdj5jb21tZXJ0aWFsIG5ldHdvcmtzLjwvZGl2Pg0KPGRpdj48YnI+DQo8
L2Rpdj4NCjxkaXY+QXMgeW91ciB1bmRlcnN0YW5kaW5nLCB0aGUgUEVzIHNob3VsZCBkbyBhcyBm
b2xsb3dpbmc6PC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4xLiBBY2N1cmF0ZSB0aW1l
ciBzeW5jLiBOVFAgYWNjdXJhY3kgaXMgYmFkLCAxNTg4djIgaXMgZ29vZCBidXQgaGF2ZTwvZGl2
Pg0KPGRpdj5yYXJlbHkgZGVwbG95bWVudC48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2
PkFzc3VtaW5nIFBFMSxQRTIgYW5kIFBFMyBoYXZlIGNvbnNpc3RlbnQgdGltZXIgY2xvY2ssIHdo
ZW4gUEUzIGpvaW5zPC9kaXY+DQo8ZGl2PkVTSSBhbmQgdHJpZ2dlciBERiByZS1lbGVjdGlvbi4g
V2hlbiByZWNlcHRpb24gdGltZXIgZXhwaXJlczo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8
ZGl2PlBFMSB1cGdyYWRlcyB0byBERiBQRS48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2
PkFmdGVyIHJlY2VwdGlvbiB0aW1lciYjNDM7IEVTIHJvdXRlIHRyYW5zbWlzc2lvbiB0aW1lcjo8
L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PlBFMiBkb3dubG9hZHMgdG8gbm9uLURGIFBF
LjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+U28gaW4gdGltZXIgY2xvY2sgc3luYyBj
YXNlLCBkdWFsIERGIFBFcyB3aWxsIGV4aXN0IGF0IGxlYXN0PC9kaXY+DQo8ZGl2PnRyYW5zbWlz
c2lvbiB0aW1lci48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PklmIE5UUCBpcyB1c2Vk
IGZvciB0aW1lciBzeW5jLCBiZWNhdXNlIGl0IGhhcyBiYWQgYWNjdXJhY3ksIGR1YWwgREY8L2Rp
dj4NCjxkaXY+UEVzIHdpbGwgZXhpc3QgbW9yZSBsb25nZXIgdGltZXIuPC9kaXY+DQo8ZGl2Pjxi
cj4NCjwvZGl2Pg0KPGRpdj5TbyBhcyB5b3VyIHVuZGVyc3RhbmRpbmcgZm9yIERGIGVsZWN0aW9u
LCB0aGUgZHJhd2JhY2sgaXMgbW9yZSBjbGVhci48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8
ZGl2PlRoYW5rcyw8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PndlaWd1bzwvZGl2Pg0K
PGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pi0tLS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t
LTwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+KkZyb206KkpvaG4gRSBEcmFrZSBbPGEg
aHJlZj0ibWFpbHRvOmpkcmFrZUBqdW5pcGVyLm5ldCI+amRyYWtlQGp1bmlwZXIubmV0PC9hPl08
L2Rpdj4NCjxkaXY+KlNlbnQ6KiBGcmlkYXksIE1hcmNoIDI3LCAyMDE1IDQ6NDE8L2Rpdj4NCjxk
aXY+KlRvOiogSGFvd2VpZ3VvOyBQYXRyaWNlIEJyaXNzZXR0ZSAocGJyaXNzZXQpPC9kaXY+DQo8
ZGl2PipDYzoqIEFsaSBTYWphc3NpIChzYWphc3NpKTsgPGEgaHJlZj0ibWFpbHRvOmJlc3NAaWV0
Zi5vcmciPmJlc3NAaWV0Zi5vcmc8L2E+ICZsdDs8YSBocmVmPSJtYWlsdG86YmVzc0BpZXRmLm9y
ZyI+bWFpbHRvOmJlc3NAaWV0Zi5vcmc8L2E+Jmd0OzwvZGl2Pg0KPGRpdj4qU3ViamVjdDoqIFJF
OiBbYmVzc10gSGFuZHNoYWtpbmcgYW1vbmcgUEVzIGluIGFuIEVWUE4gRVMgYmFzZWQgb248L2Rp
dj4NCjxkaXY+UGF4b3MgYWxnb3JpdGhtPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5X
ZWlndW8sPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5XZSBoYXZlIGFscmVhZHkgZXN0
YWJsaXNoZWQgdGhhdCB5b3VyIHByb3Bvc2FsIGlzIHVudGVuYWJsZSBiZWNhdXNlIG9mPC9kaXY+
DQo8ZGl2Pml0cyBjb250cm9sIHBsYW5lIGxvYWQuPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0K
PGRpdj5XaGF0IHdlIGFyZSBub3cgZGlzY3Vzc2luZyBpcyB0aGF0IHlvdXIgcHJvcG9zYWwgaXMg
YmFzZWQgdXBvbiBhPC9kaXY+DQo8ZGl2Pm1pc3VuZGVyc3RhbmRpbmcgb2YgdGhlIGFsZ29yaXRo
bSBpbiBSRkMgNzQzMi4mbmJzcDsmbmJzcDtZb3UgYXJlIGFzc3VtaW5nIHRoYXQ8L2Rpdj4NCjxk
aXY+UEUxIHdpbGwgYWR2ZXJ0aXNlIGFuIEVTIHJvdXRlIGFuZCB0aGVuIHdhaXQgZm9yIHRoZSBj
b25maWd1cmVkPC9kaXY+DQo8ZGl2PmludGVydmFsIGJlZm9yZSBwZXJmb3JtaW5nIHRoZSBERiBl
bGVjdGlvbiB3aGlsZSBQRTIgYW5kIFBFMyB3aWxsPC9kaXY+DQo8ZGl2PnBlcmZvcm0gdGhlIERG
IGVsZWN0aW9uIGFzIHNvb24gYXMgdGhleSByZWNlaXZlIHRoZSBFUyByb3V0ZSBmcm9tIFBFMS48
L2Rpdj4NCjxkaXY+VGhpcyBpcyBub3Qgd2hhdCBSRkMgNzQzMiBzYXlzLjwvZGl2Pg0KPGRpdj48
YnI+DQo8L2Rpdj4NCjxkaXY+UmF0aGVyLCB3aGF0IGlzIHNheXMgaXMgdGhhdCB0aGUgYWR2ZXJ0
aXNlbWVudCBvZiB0aGUgRVMgcm91dGUgYnkgUEUxPC9kaXY+DQo8ZGl2PmFuZCBpdHMgcmVjZWlw
dCBieSBQRTIgYW5kIFBFMyBjYXVzZXMgYWxsIHRocmVlIFBFcyB0byBzdGFydCB0aGU8L2Rpdj4N
CjxkaXY+Y29uZmlndXJlZCBpbnRlcnZhbCB0aW1lciZuYnNwOyZuYnNwOy0gqfgzLiBXaGVuIHRo
ZSB0aW1lciBleHBpcmVzLCBlYWNoIFBFPC9kaXY+DQo8ZGl2PmJ1aWxkcyBhbiBvcmRlcmVkIGxp
c3Qgb2YgdGhlIElQIGFkZHJlc3NlcyBvZiBhbGwgdGhlIFBFIG5vZGVzPC9kaXY+DQo8ZGl2PmNv
bm5lY3RlZCB0byB0aGUgRXRoZXJuZXQgc2VnbWVudCAoaW5jbHVkaW5nIGl0c2VsZiksIGluIGlu
Y3JlYXNpbmc8L2Rpdj4NCjxkaXY+bnVtZXJpYyB2YWx1ZS6p9zwvZGl2Pg0KPGRpdj48YnI+DQo8
L2Rpdj4NCjxkaXY+WW91cnMgSXJyZXNwZWN0aXZlbHksPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2
Pg0KPGRpdj5Kb2huPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4qRnJvbToqSGFvd2Vp
Z3VvIFs8YSBocmVmPSJtYWlsdG86aGFvd2VpZ3VvQGh1YXdlaS5jb20iPm1haWx0bzpoYW93ZWln
dW9AaHVhd2VpLmNvbTwvYT5dPC9kaXY+DQo8ZGl2PipTZW50OiogVGh1cnNkYXksIE1hcmNoIDI2
LCAyMDE1IDM6MjYgUE08L2Rpdj4NCjxkaXY+KlRvOiogSm9obiBFIERyYWtlOyBQYXRyaWNlIEJy
aXNzZXR0ZSAocGJyaXNzZXQpPC9kaXY+DQo8ZGl2PipDYzoqIEFsaSBTYWphc3NpIChzYWphc3Np
KTsgPGEgaHJlZj0ibWFpbHRvOmJlc3NAaWV0Zi5vcmciPmJlc3NAaWV0Zi5vcmc8L2E+ICZsdDs8
YSBocmVmPSJtYWlsdG86YmVzc0BpZXRmLm9yZyI+bWFpbHRvOmJlc3NAaWV0Zi5vcmc8L2E+Jmd0
OzwvZGl2Pg0KPGRpdj4qU3ViamVjdDoqIFJFOiBbYmVzc10gSGFuZHNoYWtpbmcgYW1vbmcgUEVz
IGluIGFuIEVWUE4gRVMgYmFzZWQgb248L2Rpdj4NCjxkaXY+UGF4b3MgYWxnb3JpdGhtPC9kaXY+
DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5QbHMgcmVhZCBteSBkZXRhaWwgcmVwbGllcyB0byBT
YXR5YS4gSWYgeW91IHN0aWxsIGNhbid0IGNhdGNoIGl0LCBwbHM8L2Rpdj4NCjxkaXY+cmVhZCBt
eSBkcmFmdCBhbmQgRVZQTiBiYXNlIHByb3RvY29sLCZuYnNwOyZuYnNwO3RoYW5rczwvZGl2Pg0K
PGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+d2VpZ3VvPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0K
PGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tPC9kaXY+DQo8ZGl2Pjxicj4N
CjwvZGl2Pg0KPGRpdj4qRnJvbToqSm9obiBFIERyYWtlIFs8YSBocmVmPSJtYWlsdG86amRyYWtl
QGp1bmlwZXIubmV0Ij5qZHJha2VAanVuaXBlci5uZXQ8L2E+XTwvZGl2Pg0KPGRpdj4qU2VudDoq
IEZyaWRheSwgTWFyY2ggMjcsIDIwMTUgMToyODwvZGl2Pg0KPGRpdj4qVG86KiBQYXRyaWNlIEJy
aXNzZXR0ZSAocGJyaXNzZXQpPC9kaXY+DQo8ZGl2PipDYzoqIEhhb3dlaWd1bzsgQWxpIFNhamFz
c2kgKHNhamFzc2kpOyA8YSBocmVmPSJtYWlsdG86YmVzc0BpZXRmLm9yZyI+YmVzc0BpZXRmLm9y
ZzwvYT48L2Rpdj4NCjxkaXY+Jmx0OzxhIGhyZWY9Im1haWx0bzpiZXNzQGlldGYub3JnIj5tYWls
dG86YmVzc0BpZXRmLm9yZzwvYT4mZ3Q7PC9kaXY+DQo8ZGl2PipTdWJqZWN0OiogUmU6IFtiZXNz
XSBIYW5kc2hha2luZyBhbW9uZyBQRXMgaW4gYW4gRVZQTiBFUyBiYXNlZCBvbjwvZGl2Pg0KPGRp
dj5QYXhvcyBhbGdvcml0aG08L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PkkgdGhpbmsg
UGF0cmljZSBpcyBjb3JyZWN0LiZuYnNwOyZuYnNwO1lvdXIgcHJvcG9zYWwgZG9lc24ndCBzb2x2
ZSB0aGUgcHJvYmxlbTwvZGl2Pg0KPGRpdj5hbmQgaXQgZG9lcyBzbyBhdCBodWdlIGNvc3QuPC9k
aXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5TZW50IGZyb20gbXkgaVBob25lPC9kaXY+DQo8
ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+T24gTWFyIDI2LCAyMDE1
LCBhdCAxMDozNCBBTSwgUGF0cmljZSBCcmlzc2V0dGUgKHBicmlzc2V0KTwvZGl2Pg0KPGRpdj4m
bHQ7PGEgaHJlZj0ibWFpbHRvOnBicmlzc2V0QGNpc2NvLmNvbSI+cGJyaXNzZXRAY2lzY28uY29t
PC9hPiAmbHQ7PGEgaHJlZj0ibWFpbHRvOnBicmlzc2V0QGNpc2NvLmNvbSZndDsiPm1haWx0bzpw
YnJpc3NldEBjaXNjby5jb20mZ3Q7PC9hPiZndDsgd3JvdGU6PC9kaXY+DQo8ZGl2Pjxicj4NCjwv
ZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtXZWlndW8sPC9k
aXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDtJqfZtIG5vdCBzdXJlIEmp9m0gZm9sbG93aW5nIGhlcmUuPC9kaXY+DQo8ZGl2Pjxi
cj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtEb26p
9nQgeW91IGhhdmUgdGhlIHNhbWUgaXNzdWUgd2l0aCB5b3VyIGhhbmRzaGFraW5nIG1lY2hhbmlz
bT88L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwO0lmIHlvdSBkb26p9nQga25vdyB5b3VyIHBlZXIsIGhvdyBjYW4geW91IGhh
bmRzaGFrZT88L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwO1JlZ2FyZHMsPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRp
dj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtQYXRyaWNlPC9kaXY+DQo8ZGl2
Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtJ
bWFnZSByZW1vdmVkIGJ5IHNlbmRlci48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOypQYXRyaWNlIEJyaXNzZXR0ZSo8L2Rp
dj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7VEVDSE5JQ0FMIExF
QURFUi5FTkdJTkVFUklORzwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7PGEgaHJlZj0ibWFpbHRvOnBicmlzc2V0QGNpc2Nv
LmNvbSI+cGJyaXNzZXRAY2lzY28uY29tPC9hPiAmbHQ7PGEgaHJlZj0ibWFpbHRvOnBicmlzc2V0
QGNpc2NvLmNvbSI+bWFpbHRvOnBicmlzc2V0QGNpc2NvLmNvbTwvYT4mZ3Q7PC9kaXY+DQo8ZGl2
PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO1Bob25lOiAqJiM0MzsxIDYxMyAy
NTQgMzMzNio8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRp
dj48YnI+DQo8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
KkNpc2NvIFN5c3RlbXMgQ2FuYWRhIENvLiAvIExlcyBTeXN0ZW1lcyBDaXNjbyBDYW5hZGEgQ0lF
KjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtDYW5hZGE8
L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Q2lzY28uY29t
ICZsdDs8YSBocmVmPSJodHRwOi8vd3d3LmNpc2NvLmNvbS9nbG9iYWwvQ0EvIj5odHRwOi8vd3d3
LmNpc2NvLmNvbS9nbG9iYWwvQ0EvPC9hPiZndDs8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8
ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7SW1hZ2UgcmVtb3ZlZCBieSBzZW5kZXIuVGhpbmsgYmVmb3Jl
IHlvdSBwcmludC48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwO1RoaXMgZW1haWwgbWF5IGNvbnRhaW4gY29uZmlkZW50aWFs
IGFuZCBwcml2aWxlZ2VkIG1hdGVyaWFsIGZvcjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDt0aGUgc29sZSB1c2Ugb2YgdGhlIGludGVuZGVkIHJlY2lwaWVu
dC4gQW55IHJldmlldywgdXNlLDwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDtkaXN0cmlidXRpb24gb3IgZGlzY2xvc3VyZSBieSBvdGhlcnMgaXMgc3RyaWN0
bHkgcHJvaGliaXRlZC4gSWY8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7eW91IGFyZSBub3QgdGhlIGludGVuZGVkIHJlY2lwaWVudCAob3IgYXV0aG9yaXpl
ZCB0byByZWNlaXZlIGZvcjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDt0aGUgcmVjaXBpZW50KSwgcGxlYXNlIGNvbnRhY3QgdGhlIHNlbmRlciBieSByZXBs
eSBlbWFpbCBhbmQ8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7ZGVsZXRlIGFsbCBjb3BpZXMgb2YgdGhpcyBtZXNzYWdlLjwvZGl2Pg0KPGRpdj48YnI+DQo8
L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7UGxlYXNlIGNs
aWNrIGhlcmU8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZsdDs8YSBocmVmPSJodHRw
Oi8vd3d3LmNpc2NvLmNvbS93ZWIvYWJvdXQvZG9pbmdfYnVzaW5lc3MvbGVnYWwvY3JpL2luZGV4
Lmh0bWwiPmh0dHA6Ly93d3cuY2lzY28uY29tL3dlYi9hYm91dC9kb2luZ19idXNpbmVzcy9sZWdh
bC9jcmkvaW5kZXguaHRtbDwvYT4mZ3Q7PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwO2ZvciBDb21wYW55IFJlZ2lzdHJhdGlvbiBJbmZvcm1hdGlvbi48L2Rp
dj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsqRnJvbTogKkhhb3dlaWd1byAmbHQ7PGEgaHJlZj0i
bWFpbHRvOmhhb3dlaWd1b0BodWF3ZWkuY29tIj5oYW93ZWlndW9AaHVhd2VpLmNvbTwvYT48L2Rp
dj4NCjxkaXY+Jmx0OzxhIGhyZWY9Im1haWx0bzpoYW93ZWlndW9AaHVhd2VpLmNvbSZndDsiPm1h
aWx0bzpoYW93ZWlndW9AaHVhd2VpLmNvbSZndDs8L2E+Jmd0OzwvZGl2Pg0KPGRpdj4mbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsqRGF0ZTogKlRodXJzZGF5LCBNYXJjaCAyNiwg
MjAxNSBhdCAxMDoyNCBBTTwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsqVG86ICpQYXRyaWNlIEJyaXNzZXR0ZSAmbHQ7PGEgaHJlZj0ibWFpbHRvOnBicmlz
c2V0QGNpc2NvLmNvbSI+cGJyaXNzZXRAY2lzY28uY29tPC9hPjwvZGl2Pg0KPGRpdj4mbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbHQ7PGEgaHJlZj0ibWFpbHRvOnBicmlzc2V0
QGNpc2NvLmNvbSZndDsiPm1haWx0bzpwYnJpc3NldEBjaXNjby5jb20mZ3Q7PC9hPiZndDssIEpv
aG4gRSBEcmFrZSAmbHQ7PGEgaHJlZj0ibWFpbHRvOmpkcmFrZUBqdW5pcGVyLm5ldCI+amRyYWtl
QGp1bmlwZXIubmV0PC9hPjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbHQ7PGEgaHJlZj0ibWFpbHRvOmpkcmFrZUBqdW5pcGVyLm5ldCZndDsiPm1haWx0
bzpqZHJha2VAanVuaXBlci5uZXQmZ3Q7PC9hPiZndDssIEFsaSBTYWphc3NpICZsdDs8YSBocmVm
PSJtYWlsdG86c2FqYXNzaUBjaXNjby5jb20iPnNhamFzc2lAY2lzY28uY29tPC9hPjwvZGl2Pg0K
PGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbHQ7PGEgaHJlZj0ibWFp
bHRvOnNhamFzc2lAY2lzY28uY29tJmd0OyI+bWFpbHRvOnNhamFzc2lAY2lzY28uY29tJmd0Ozwv
YT4mZ3Q7LCAmcXVvdDs8YSBocmVmPSJtYWlsdG86YmVzc0BpZXRmLm9yZyI+YmVzc0BpZXRmLm9y
ZzwvYT48L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jmx0
OzxhIGhyZWY9Im1haWx0bzpiZXNzQGlldGYub3JnJmd0OyI+bWFpbHRvOmJlc3NAaWV0Zi5vcmcm
Z3Q7PC9hPiZxdW90OyAmbHQ7PGEgaHJlZj0ibWFpbHRvOmJlc3NAaWV0Zi5vcmciPmJlc3NAaWV0
Zi5vcmc8L2E+ICZsdDs8YSBocmVmPSJtYWlsdG86YmVzc0BpZXRmLm9yZyZndDsiPm1haWx0bzpi
ZXNzQGlldGYub3JnJmd0OzwvYT4mZ3Q7PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOypTdWJqZWN0OiAqUkU6IFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQ
RXMgaW4gYW4gRVZQTiBFUyBiYXNlZCBvbjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDtQYXhvcyBhbGdvcml0aG08L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+
DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwO0hpIFBhdHJpY2UsPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtV
cCB0byByZWNlcHRpb24gdGltZXIgdHJhZmZpYyBkaXNydXB0aW9uIGluIHRyYW5zaWVudCBwaGFz
ZSBpczwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDtvbmUgb2YgdGhlIGlzc3Vlcy48L2Rpdj4NCjxkaXY+PGJyPg0K
PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwO1RoYW5rcyw8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
O3dlaWd1bzwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2
Pi0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLTwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7KkZyb206
KlBhdHJpY2UgQnJpc3NldHRlIChwYnJpc3NldCkgWzxhIGhyZWY9Im1haWx0bzpwYnJpc3NldEBj
aXNjby5jb20iPnBicmlzc2V0QGNpc2NvLmNvbTwvYT48L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jmx0OzxhIGhy
ZWY9Im1haWx0bzpwYnJpc3NldEBjaXNjby5jb20mZ3Q7Ij5tYWlsdG86cGJyaXNzZXRAY2lzY28u
Y29tJmd0OzwvYT5dPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOypTZW50OiogVGh1cnNkYXksIE1hcmNoIDI2LCAy
MDE1IDIwOjU0PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOypUbzoqIEhhb3dlaWd1bzsgSm9obiBFIERyYWtlOyBB
bGkgU2FqYXNzaSAoc2FqYXNzaSk7PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOzxhIGhyZWY9Im1haWx0bzpiZXNz
QGlldGYub3JnIj5iZXNzQGlldGYub3JnPC9hPiAmbHQ7PGEgaHJlZj0ibWFpbHRvOmJlc3NAaWV0
Zi5vcmciPm1haWx0bzpiZXNzQGlldGYub3JnPC9hPiZndDs8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7KlN1Ympl
Y3Q6KiBSZTogW2Jlc3NdIEhhbmRzaGFraW5nIGFtb25nIFBFcyBpbiBhbiBFVlBOIEVTPC9kaXY+
DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwO2Jhc2VkIG9uIFBheG9zIGFsZ29yaXRobTwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rp
dj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7V2VpZ3VvLDwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7WW91
IG1lbnRpb24gJnF1b3Q7QnV0IGlmIHlvdXIgZHJhZnQgaGF2ZSBub3Qgc29sdmVkIGFsbCBpc3N1
ZXOp9yw8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO0NhbiB5b3UgZXhwbGFpbiB3
aGF0IFNhdHlhqfZzIGRyYWZ0IGlzIG5vdCBzb2x2aW5nPzwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rp
dj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7UmVnYXJkcyw8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO1Bh
dHJpY2U8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO0ltYWdlIHJlbW92ZWQgYnkg
c2VuZGVyLjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7KlBhdHJpY2UgQnJpc3Nl
dHRlKjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDtURUNITklDQUwgTEVBREVSLkVOR0lORUVSSU5HPC9kaXY+DQo8
ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDs8YSBocmVmPSJtYWlsdG86cGJyaXNzZXRAY2lzY28u
Y29tIj5wYnJpc3NldEBjaXNjby5jb208L2E+ICZsdDs8YSBocmVmPSJtYWlsdG86cGJyaXNzZXRA
Y2lzY28uY29tIj5tYWlsdG86cGJyaXNzZXRAY2lzY28uY29tPC9hPiZndDs8L2Rpdj4NCjxkaXY+
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7UGhvbmU6IComIzQzOzEgNjEzIDI1NCAzMzM2KjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4N
CjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsqQ2lzY28gU3lz
dGVtcyBDYW5hZGEgQ28uIC8gTGVzIFN5c3RlbWVzIENpc2NvIENhbmFkYSBDSUUqPC9kaXY+DQo8
ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwO0NhbmFkYTwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtDaXNjby5jb20gJmx0OzxhIGhyZWY9Imh0dHA6
Ly93d3cuY2lzY28uY29tL2dsb2JhbC9DQS8iPmh0dHA6Ly93d3cuY2lzY28uY29tL2dsb2JhbC9D
QS88L2E+Jmd0OzwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8
ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtJbWFnZSByZW1vdmVkIGJ5IHNlbmRlci5UaGluayBi
ZWZvcmUgeW91IHByaW50LjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7VGhpcyBl
bWFpbCBtYXkgY29udGFpbiBjb25maWRlbnRpYWwgYW5kIHByaXZpbGVnZWQgbWF0ZXJpYWw8L2Rp
dj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Zm9yIHRoZSBzb2xlIHVzZSBvZiB0aGUgaW50ZW5kZWQgcmVjaXBpZW50LiBB
bnkgcmV2aWV3LCB1c2UsPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO2Rpc3RyaWJ1dGlvbiBvciBkaXNjbG9zdXJl
IGJ5IG90aGVycyBpcyBzdHJpY3RseSBwcm9oaWJpdGVkLjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtJZiB5b3Ug
YXJlIG5vdCB0aGUgaW50ZW5kZWQgcmVjaXBpZW50IChvciBhdXRob3JpemVkIHRvPC9kaXY+DQo8
ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwO3JlY2VpdmUgZm9yIHRoZSByZWNpcGllbnQpLCBwbGVhc2UgY29udGFjdCB0aGUgc2Vu
ZGVyIGJ5IHJlcGx5PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO2VtYWlsIGFuZCBkZWxldGUgYWxsIGNvcGllcyBv
ZiB0aGlzIG1lc3NhZ2UuPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtQbGVhc2Ug
Y2xpY2sgaGVyZTwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Jmx0OzxhIGhyZWY9Imh0
dHA6Ly93d3cuY2lzY28uY29tL3dlYi9hYm91dC9kb2luZ19idXNpbmVzcy9sZWdhbC9jcmkvaW5k
ZXguaHRtbCI+aHR0cDovL3d3dy5jaXNjby5jb20vd2ViL2Fib3V0L2RvaW5nX2J1c2luZXNzL2xl
Z2FsL2NyaS9pbmRleC5odG1sPC9hPiZndDs8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Zm9yIENvbXBhbnkgUmVn
aXN0cmF0aW9uIEluZm9ybWF0aW9uLjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+PGJy
Pg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOypGcm9tOiAqSGFvd2VpZ3VvICZsdDs8YSBocmVmPSJtYWlsdG86
aGFvd2VpZ3VvQGh1YXdlaS5jb20iPmhhb3dlaWd1b0BodWF3ZWkuY29tPC9hPjwvZGl2Pg0KPGRp
dj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbHQ7PGEgaHJlZj0ibWFpbHRvOmhhb3dlaWd1b0BodWF3ZWkuY29tJmd0OyI+bWFpbHRv
Omhhb3dlaWd1b0BodWF3ZWkuY29tJmd0OzwvYT4mZ3Q7PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOypEYXRlOiAq
V2VkbmVzZGF5LCBNYXJjaCAyNSwgMjAxNSBhdCAxMDozOCBQTTwvZGl2Pg0KPGRpdj4mbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsqVG86
ICpKb2huIEUgRHJha2UgJmx0OzxhIGhyZWY9Im1haWx0bzpqZHJha2VAanVuaXBlci5uZXQiPmpk
cmFrZUBqdW5pcGVyLm5ldDwvYT48L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jmx0OzxhIGhyZWY9Im1haWx0bzpq
ZHJha2VAanVuaXBlci5uZXQmZ3Q7Ij5tYWlsdG86amRyYWtlQGp1bmlwZXIubmV0Jmd0OzwvYT4m
Z3Q7LCBBbGkgU2FqYXNzaSAmbHQ7PGEgaHJlZj0ibWFpbHRvOnNhamFzc2lAY2lzY28uY29tIj5z
YWphc3NpQGNpc2NvLmNvbTwvYT48L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jmx0OzxhIGhyZWY9Im1haWx0bzpz
YWphc3NpQGNpc2NvLmNvbSZndDsiPm1haWx0bzpzYWphc3NpQGNpc2NvLmNvbSZndDs8L2E+Jmd0
OywgJnF1b3Q7PGEgaHJlZj0ibWFpbHRvOmJlc3NAaWV0Zi5vcmciPmJlc3NAaWV0Zi5vcmc8L2E+
PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZsdDs8YSBocmVmPSJtYWlsdG86YmVzc0BpZXRmLm9yZyZndDsiPm1h
aWx0bzpiZXNzQGlldGYub3JnJmd0OzwvYT4mcXVvdDsgJmx0OzxhIGhyZWY9Im1haWx0bzpiZXNz
QGlldGYub3JnIj5iZXNzQGlldGYub3JnPC9hPiAmbHQ7PGEgaHJlZj0ibWFpbHRvOmJlc3NAaWV0
Zi5vcmcmZ3Q7Ij5tYWlsdG86YmVzc0BpZXRmLm9yZyZndDs8L2E+Jmd0OzwvZGl2Pg0KPGRpdj4m
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsqU3ViamVjdDogKlJlOiBbYmVzc10gSGFuZHNoYWtpbmcgYW1vbmcgUEVzIGluIGFuIEVWUE4g
RVM8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7YmFzZWQgb24gUGF4b3MgYWxnb3JpdGhtPC9kaXY+DQo8ZGl2Pjxi
cj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtIaSBKb2huLDwvZGl2
Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7IEZp
cnN0bHkgaSB0aGluayBFVlBOIGNvbW11bml0eSBzaG91bGQgcmVhY2ggY29uc2Vuc3VzIG9uPC9k
aXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO3RoZSBpc3N1ZXMgb2YgY3VycmVu
dCBERiBlbGVjdGlvbiBtZWNoYW5pc20uIEFsbCB0aGVzZTwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDtpc3N1ZXMgc2hvdWxkIGJlIHJlc29sdmVkIGluIGEgc2luZ2xlIG5l
dyBERiBlbGVjdGlvbjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtkcmFm
dCxyYXRoZXIgdGhhbiBpbiBtdWx0aXBsZSBzZXBhcmF0ZSBkcmFmdHMuIElmIHlvdXI8L2Rpdj4N
CjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7ZHJhZnQgY2FuIHNvbHZlIGFsbCB0aGVz
ZSBpc3N1ZXMgYW5kIHN0YWJsZSwgaSBoYXZlIG5vPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwO3F1ZXN0aW9uIGZvciBpdHMgcHJvZ3Jlc3NpbmcuIEJ1dCBpZiB5b3VyIGRy
YWZ0IGhhdmUgbm90PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO3NvbHZl
ZCBhbGwgaXNzdWVzLCBpIHRoaW5rIGl0IGhhZCBiZXR0ZXIgY29tYmluZSB3aXRoPC9kaXY+DQo8
ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO290aGVyIGRyYWZ0cyB0byBwcm92aWRlIGEg
Y29tcHJlaGVuc2l2ZSBzb2x1dGlvbi4gSSB0aGluazwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDt0aGUgaXNzdWVzIGxpc3RlZCBpbiBkcmFmdC1yYWJhZGFuLWJlc3MtZXZw
bi1hYy1kZi0wMSBhbmQ8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7ZHJh
ZnQtaGFvLWJlc3MtZXZwbi1kZi1oYW5kc2hha2luZy0wMCBpcyB2YWxpZCwgaXQgc2hvdWxkPC9k
aXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO2JlIHJlc29sdmVkLiBTbyBpIHRo
aW5rIGFsdGhvdWdoIHlvdXIgbmV3IEhhc2ggYWxnb3JpdGhtPC9kaXY+DQo8ZGl2PiZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwO2ZvciBERiBlbGVjdGlvbiBpcyBnb29kLCBpdCBvbmx5IGluY2x1
ZGVzIHBhcnRpYWw8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7ZW5oYW5j
ZW1lbnRzLCBtYXliZSBpdCBzdGlsbCBuZWVkcyBzb21lIHRpbWUgZm9yPC9kaXY+DQo8ZGl2PmNv
bnNlbnN1cy48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwO1RoYW5rcyw8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwO3dlaWd1bzwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+
PGJyPg0KPC9kaXY+DQo8ZGl2Pi0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLTwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rp
dj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7KkZyb206KkpvaG4gRSBEcmFrZSBb
PGEgaHJlZj0ibWFpbHRvOmpkcmFrZUBqdW5pcGVyLm5ldCI+amRyYWtlQGp1bmlwZXIubmV0PC9h
PjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbHQ7PGEgaHJlZj0ibWFp
bHRvOmpkcmFrZUBqdW5pcGVyLm5ldCZndDsiPm1haWx0bzpqZHJha2VAanVuaXBlci5uZXQmZ3Q7
PC9hPl08L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7KlNlbnQ6KiBUaHVy
c2RheSwgTWFyY2ggMjYsIDIwMTUgNzoxNjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsqVG86KiBIYW93ZWlndW87IEFsaSBTYWphc3NpIChzYWphc3NpKTsgPGEgaHJlZj0i
bWFpbHRvOmJlc3NAaWV0Zi5vcmciPg0KYmVzc0BpZXRmLm9yZzwvYT48L2Rpdj4NCjxkaXY+Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jmx0OzxhIGhyZWY9Im1haWx0bzpiZXNzQGlldGYub3Jn
Ij5tYWlsdG86YmVzc0BpZXRmLm9yZzwvYT4mZ3Q7PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOypTdWJqZWN0OiogUkU6IFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQRXMg
aW4gYW4gRVZQTiBFUzwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtiYXNl
ZCBvbiBQYXhvcyBhbGdvcml0aG08L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwO1dlaWd1byw8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8
ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO1lvdXIgcHJvcG9zYWwgaW50cm9kdWNlcyBh
IGNvbnRyb2wgcGxhbmUgcHJvY2Vzc2luZyBsb2FkPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwO3RoYXQgaXMgTygjRVZJcyAqIFBFcykgcGVyIERGIGVsZWN0aW9uIGFuZCBn
aXZlbiB0aGF0PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO3RoZXJlIGNh
biBiZSA0SyBFVklzIHBlciBFUywgdGhpcyBsb29rcyBsaWtlIGE8L2Rpdj4NCjxkaXY+Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7KipzdWJzdGFudGlhbCoqIGxvYWQuIEZ1cnRoZXJtb3JlLDwv
ZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
eW91IGNhbqn2dCZuYnNwOyZuYnNwO3VzZSB0aGUgRVMgcm91dGUgdG8gY28tb3JkaW5hdGUgREYg
ZWxlY3Rpb248L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7YmVjYXVzZSB5
b3Ugd291bGQgbmVlZCB0byBjYXJyeSB5b3VyIG5ldyBleHRlbmRlZDwvZGl2Pg0KPGRpdj4mbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtjb21tdW5pdHkgZm9yIGVhY2ggRVZJIGFuZCB0aGV5IHdv
dWxkIG5vdCBhbGwgZml0LiZuYnNwOyZuYnNwO1lvdTwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDthbHNvIGNhbqn2dCB1c2UgdGhlIFBlciBFVkkgRXRoZXJuZXQgQUQgcm91
dGUgYmVjYXVzZSB0aGF0PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO2lz
IHByb2Nlc3NlZCBieSBhbGwgUEVzIGluIHRoZSBFVkkuPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2
Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtJIHRoaW5rIHRoYXQgZnJvbSBhIHBy
YWN0aWNhbCBwZXJzcGVjdGl2ZSB0aGUgbmV3IERGPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwO2VsZWN0aW9uIHByb3Bvc2VkIGluIFNhdHlhqfZzIGRyYWZ0IGlzIHN1ZmZp
Y2llbnRseSBzdGFibGU8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7dGhh
dCBpdCByZW5kZXJzIHlvdXIgZHJhZnQgbW9vdCwgZXZlbiBpZiBpdCBjb3VsZCBiZSBtYWRlPC9k
aXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO3RvIHdvcmsuPC9kaXY+DQo8ZGl2
Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtZb3VycyBJcnJl
c3BlY3RpdmVseSw8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwO0pvaG48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOypGcm9tOipCRVNTIFs8YSBocmVmPSJtYWlsdG86YmVzcy1ib3Vu
Y2VzQGlldGYub3JnIj5tYWlsdG86YmVzcy1ib3VuY2VzQGlldGYub3JnPC9hPl0gKk9uIEJlaGFs
ZiBPZjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsqSGFvd2VpZ3VvPC9k
aXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOypTZW50OiogV2VkbmVzZGF5LCBN
YXJjaCAyNSwgMjAxNSA1OjM0IFBNPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOypUbzoqIEFsaSBTYWphc3NpIChzYWphc3NpKTsgPGEgaHJlZj0ibWFpbHRvOmJlc3NAaWV0
Zi5vcmciPmJlc3NAaWV0Zi5vcmc8L2E+PC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZsdDs8YSBocmVmPSJtYWlsdG86YmVzc0BpZXRmLm9yZyI+bWFpbHRvOmJlc3NAaWV0
Zi5vcmc8L2E+Jmd0OzwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsqU3Vi
amVjdDoqIFJlOiBbYmVzc10gSGFuZHNoYWtpbmcgYW1vbmcgUEVzIGluIGFuIEVWUE4gRVM8L2Rp
dj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7YmFzZWQgb24gUGF4b3MgYWxnb3Jp
dGhtPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDtIaSBBbGksPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDtUaGFua3MgZm9yIHlvdXIgaW5mb3JtYXRpb24uIEkgc2Nhbm5lZCB0aHJv
dWdoIHRoaXMgZHJhZnQsPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO2l0
IHJlYWxseSBpbnRyb2R1Y2VzIGludGVyLWNoYXNzaXMgbWVzc2FnZSBmb3IgREYgZWxlY3Rpb248
L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7aGFuZHNoYWtpbmcsIHRoZSBy
ZXF1aXJlbWVudHMgaW4gdGhpcyBkcmFmdCBpcyB0bzwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDtlbGltaW5hdGUgdHJhbnNpZXQgTG9vcCBhbmQgdHJhZmZpYyBkdXBsaWNh
dGlvbi4gQ3VycmVudDwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtFVlBO
IERGIGVsZWN0aW9uIG1lY2hhbmlzbSBhbHJlYWR5IGVsaW1pbmF0ZWQgbG9vcCBhbmQ8L2Rpdj4N
CjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7dHJhZmZpYyBkdXBsaWNhdGlvbiBieSBj
b25maWd1cmluZyBsb25nIHJlY2VwdGlvbiB0aW1lciBvbjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDtlYWNoIG11bHRpLWhvbWVkIFBFLCBidXQgdXAgdG8gcmVjZXB0aW9u
IHRpbWVyIHRyYWZmaWM8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7ZGlz
cnVwdGlvbiBpc3N1ZSBzdGlsbCBleGlzdC4gRVZQTiBmb3IgRENJIGlzIGFuIGltcG9ydGFudDwv
ZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDt1c2UgY2FzZSBmb3IgRVZQTiwg
dXAgdG8gcmVjZXB0aW9uIHRpbWVyIHRyYWZmaWM8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7ZGlzcnVwdGlvbiBjYW4ndCBiZSB0b2xlcmF0ZWQgZm9yIHNlcnZpY2UgcHJv
dmlkZXJzLCBpdDwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtzaG91bGQg
YmUgaW1wcm92ZWQuPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDtBbHNvIGZvciBhY2N1cmFjeSwgaSB0aGluayBoYW5kc2hha2luZyBzdGF0
ZSBtYWNoaW5lIG9uPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO2VhY2gg
bXVsdGktaG9tZWQgUEUgaXMgYWxzbyBuZWVkZWQuIEZyb20gc29sdXRpb248L2Rpdj4NCjxkaXY+
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7cGVyc3BlY3RpdmUsIGluIG15IGRyYWZ0LCBubyBp
bnRlci1jaGFzc2lzIG1lc3NhZ2UgaXM8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7aW50cm9kdWNlZCwgb25seSBvbmUgbmV3IGV4dGVuZGVkIGNvbW11bml0eSBpcyBpbnRy
b2R1Y2VkLDwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtpIHRoaW5rIHRo
ZSBwcm9jZXNzIGlzIGNvbXBhcmF0aXZlbHkgc2ltcGxlIHRoYW4geW91cjwvZGl2Pg0KPGRpdj4m
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtmb2xsb3dpbmcgZHJhZnQuPC9kaXY+DQo8ZGl2Pjxi
cj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtDdXJyZW50IEVWUE4g
REYgZWxlY3Rpb24gaGFzIHNvbWUgZHJhd2JhY2tzLCBzbyB0aGVyZSBhcmU8L2Rpdj4NCjxkaXY+
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7dGhyZWUgbmV3IGRyYWZ0cyBhYm91dCBERiBlbGVj
dGlvbiBlbWVyZ2VkLiBJIHRoaW5rIEJFU1M8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7V0cgY2FuIGNvbnNpZGVyIHRoZXNlIHRocmVlIGRyYWZ0cyBpbiBnbG9iYWwgdmll
dywgYTwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtzaW5nbGUsY29tcHJl
aGVuc2l2ZSBuZXcgREYgZWxlY3Rpb24gZHJhZnQgaXMgaG9wZWQuPC9kaXY+DQo8ZGl2Pjxicj4N
CjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDt0aGFua3MuPC9kaXY+DQo8
ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDtUaGFua3Ms
PC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDt3ZWlndW88L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRp
dj4tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS08L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOypGcm9tOipCRVNTIFs8YSBocmVmPSJtYWlsdG86YmVzcy1ib3Vu
Y2VzQGlldGYub3JnIj5iZXNzLWJvdW5jZXNAaWV0Zi5vcmc8L2E+PC9kaXY+DQo8ZGl2PiZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZsdDs8YSBocmVmPSJtYWlsdG86YmVzcy1ib3VuY2VzQGll
dGYub3JnJmd0OyI+bWFpbHRvOmJlc3MtYm91bmNlc0BpZXRmLm9yZyZndDs8L2E+XSBvbiBiZWhh
bGYgb2YgQWxpIFNhamFzc2k8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
KHNhamFzc2kpIFs8YSBocmVmPSJtYWlsdG86c2FqYXNzaUBjaXNjby5jb20iPnNhamFzc2lAY2lz
Y28uY29tPC9hPiAmbHQ7PGEgaHJlZj0ibWFpbHRvOnNhamFzc2lAY2lzY28uY29tJmd0OyI+bWFp
bHRvOnNhamFzc2lAY2lzY28uY29tJmd0OzwvYT5dPC9kaXY+DQo8ZGl2PiZuYnNwOyZuYnNwOyZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOypTZW50OiogVGh1cnNkYXksIE1hcmNoIDI2LCAyMDE1IDA6MjE8L2Rpdj4N
CjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7KlRvOiogPGEgaHJlZj0ibWFpbHRvOmJl
c3NAaWV0Zi5vcmciPmJlc3NAaWV0Zi5vcmc8L2E+ICZsdDs8YSBocmVmPSJtYWlsdG86YmVzc0Bp
ZXRmLm9yZyI+bWFpbHRvOmJlc3NAaWV0Zi5vcmc8L2E+Jmd0OzwvZGl2Pg0KPGRpdj4mbmJzcDsm
bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsqU3ViamVjdDoqIFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQ
RXMgaW4gYW4gRVZQTiBFUzwvZGl2Pg0KPGRpdj4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDti
YXNlZCBvbiBQYXhvcyBhbGdvcml0aG08L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PiZu
YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw
OyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwO0ZZSS0gRmlyc3QgcHVibGlzaGVkIEp1bHkgNCwgMjAx
MTwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PjxhIGhy
ZWY9Imh0dHBzOi8vZGF0YXRyYWNrZXIuaWV0Zi5vcmcvZG9jL2RyYWZ0LXNhamFzc2ktbDJ2cG4t
ZXZwbi1zZWdtZW50LXJvdXRlLyI+aHR0cHM6Ly9kYXRhdHJhY2tlci5pZXRmLm9yZy9kb2MvZHJh
ZnQtc2FqYXNzaS1sMnZwbi1ldnBuLXNlZ21lbnQtcm91dGUvPC9hPjwvZGl2Pg0KPGRpdj48YnI+
DQo8L2Rpdj4NCjxkaXY+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7LUFsaTwvZGl2Pg0KPGRp
dj48YnI+DQo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRp
dj5fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXzwvZGl2Pg0K
PGRpdj5CRVNTIG1haWxpbmcgbGlzdDwvZGl2Pg0KPGRpdj48YSBocmVmPSJtYWlsdG86QkVTU0Bp
ZXRmLm9yZyI+QkVTU0BpZXRmLm9yZzwvYT48L2Rpdj4NCjxkaXY+PGEgaHJlZj0iaHR0cHM6Ly93
d3cuaWV0Zi5vcmcvbWFpbG1hbi9saXN0aW5mby9iZXNzIj5odHRwczovL3d3dy5pZXRmLm9yZy9t
YWlsbWFuL2xpc3RpbmZvL2Jlc3M8L2E+PC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8ZGl2Pjxicj4N
CjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4N
CjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4N
CjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4N
CjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4N
CjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+X19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXzwvZGl2Pg0K
PGRpdj5fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX188L2Rp
dj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2PkNlIG1lc3NhZ2UgZXQgc2VzIHBpZWNlcyBqb2lu
dGVzIHBldXZlbnQgY29udGVuaXIgZGVzIGluZm9ybWF0aW9uczwvZGl2Pg0KPGRpdj5jb25maWRl
bnRpZWxsZXMgb3UgcHJpdmlsZWdpZWVzIGV0IG5lIGRvaXZlbnQgZG9uYzwvZGl2Pg0KPGRpdj5w
YXMgZXRyZSBkaWZmdXNlcywgZXhwbG9pdGVzIG91IGNvcGllcyBzYW5zIGF1dG9yaXNhdGlvbi4g
U2kgdm91cyBhdmV6PC9kaXY+DQo8ZGl2PnJlY3UgY2UgbWVzc2FnZSBwYXIgZXJyZXVyLCB2ZXVp
bGxleiBsZSBzaWduYWxlcjwvZGl2Pg0KPGRpdj5hIGwnZXhwZWRpdGV1ciBldCBsZSBkZXRydWly
ZSBhaW5zaSBxdWUgbGVzIHBpZWNlcyBqb2ludGVzLiBMZXM8L2Rpdj4NCjxkaXY+bWVzc2FnZXMg
ZWxlY3Ryb25pcXVlcyBldGFudCBzdXNjZXB0aWJsZXMgZCdhbHRlcmF0aW9uLDwvZGl2Pg0KPGRp
dj5PcmFuZ2UgZGVjbGluZSB0b3V0ZSByZXNwb25zYWJpbGl0ZSBzaSBjZSBtZXNzYWdlIGEgZXRl
IGFsdGVyZSwgZGVmb3JtZTwvZGl2Pg0KPGRpdj5vdSBmYWxzaWZpZS4gTWVyY2kuPC9kaXY+DQo8
ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5UaGlzIG1lc3NhZ2UgYW5kIGl0cyBhdHRhY2htZW50cyBt
YXkgY29udGFpbiBjb25maWRlbnRpYWwgb3IgcHJpdmlsZWdlZDwvZGl2Pg0KPGRpdj5pbmZvcm1h
dGlvbiB0aGF0IG1heSBiZSBwcm90ZWN0ZWQgYnkgbGF3OzwvZGl2Pg0KPGRpdj50aGV5IHNob3Vs
ZCBub3QgYmUgZGlzdHJpYnV0ZWQsIHVzZWQgb3IgY29waWVkIHdpdGhvdXQgYXV0aG9yaXNhdGlv
bi48L2Rpdj4NCjxkaXY+SWYgeW91IGhhdmUgcmVjZWl2ZWQgdGhpcyBlbWFpbCBpbiBlcnJvciwg
cGxlYXNlIG5vdGlmeSB0aGUgc2VuZGVyIGFuZDwvZGl2Pg0KPGRpdj5kZWxldGUgdGhpcyBtZXNz
YWdlIGFuZCBpdHMgYXR0YWNobWVudHMuPC9kaXY+DQo8ZGl2PkFzIGVtYWlscyBtYXkgYmUgYWx0
ZXJlZCwgT3JhbmdlIGlzIG5vdCBsaWFibGUgZm9yIG1lc3NhZ2VzIHRoYXQgaGF2ZTwvZGl2Pg0K
PGRpdj5iZWVuIG1vZGlmaWVkLCBjaGFuZ2VkIG9yIGZhbHNpZmllZC48L2Rpdj4NCjxkaXY+VGhh
bmsgeW91LjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+X19fX19fX19fX19fX19fX19f
X19fX19fX19fX19fX19fX19fX19fX19fX19fX188L2Rpdj4NCjxkaXY+QkVTUyBtYWlsaW5nIGxp
c3Q8L2Rpdj4NCjxkaXY+PGEgaHJlZj0ibWFpbHRvOkJFU1NAaWV0Zi5vcmciPkJFU1NAaWV0Zi5v
cmc8L2E+PC9kaXY+DQo8ZGl2PjxhIGhyZWY9Imh0dHBzOi8vd3d3LmlldGYub3JnL21haWxtYW4v
bGlzdGluZm8vYmVzcyI+aHR0cHM6Ly93d3cuaWV0Zi5vcmcvbWFpbG1hbi9saXN0aW5mby9iZXNz
PC9hPjwvZGl2Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjxkaXY+PGJyPg0K
PC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5fX19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXzwvZGl2Pg0K
PGRpdj5fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXzwvZGl2
Pg0KPGRpdj48YnI+DQo8L2Rpdj4NCjxkaXY+Q2UgbWVzc2FnZSBldCBzZXMgcGllY2VzIGpvaW50
ZXMgcGV1dmVudCBjb250ZW5pciBkZXMgaW5mb3JtYXRpb25zPC9kaXY+DQo8ZGl2PmNvbmZpZGVu
dGllbGxlcyBvdSBwcml2aWxlZ2llZXMgZXQgbmUgZG9pdmVudCBkb25jPC9kaXY+DQo8ZGl2PnBh
cyBldHJlIGRpZmZ1c2VzLCBleHBsb2l0ZXMgb3UgY29waWVzIHNhbnMgYXV0b3Jpc2F0aW9uLiBT
aSB2b3VzIGF2ZXo8L2Rpdj4NCjxkaXY+cmVjdSBjZSBtZXNzYWdlIHBhciBlcnJldXIsIHZldWls
bGV6IGxlIHNpZ25hbGVyPC9kaXY+DQo8ZGl2PmEgbCdleHBlZGl0ZXVyIGV0IGxlIGRldHJ1aXJl
IGFpbnNpIHF1ZSBsZXMgcGllY2VzIGpvaW50ZXMuIExlcyBtZXNzYWdlczwvZGl2Pg0KPGRpdj5l
bGVjdHJvbmlxdWVzIGV0YW50IHN1c2NlcHRpYmxlcyBkJ2FsdGVyYXRpb24sPC9kaXY+DQo8ZGl2
Pk9yYW5nZSBkZWNsaW5lIHRvdXRlIHJlc3BvbnNhYmlsaXRlIHNpIGNlIG1lc3NhZ2UgYSBldGUg
YWx0ZXJlLCBkZWZvcm1lPC9kaXY+DQo8ZGl2Pm91IGZhbHNpZmllLiBNZXJjaS48L2Rpdj4NCjxk
aXY+PGJyPg0KPC9kaXY+DQo8ZGl2PlRoaXMgbWVzc2FnZSBhbmQgaXRzIGF0dGFjaG1lbnRzIG1h
eSBjb250YWluIGNvbmZpZGVudGlhbCBvciBwcml2aWxlZ2VkPC9kaXY+DQo8ZGl2PmluZm9ybWF0
aW9uIHRoYXQgbWF5IGJlIHByb3RlY3RlZCBieSBsYXc7PC9kaXY+DQo8ZGl2PnRoZXkgc2hvdWxk
IG5vdCBiZSBkaXN0cmlidXRlZCwgdXNlZCBvciBjb3BpZWQgd2l0aG91dCBhdXRob3Jpc2F0aW9u
LjwvZGl2Pg0KPGRpdj5JZiB5b3UgaGF2ZSByZWNlaXZlZCB0aGlzIGVtYWlsIGluIGVycm9yLCBw
bGVhc2Ugbm90aWZ5IHRoZSBzZW5kZXIgYW5kPC9kaXY+DQo8ZGl2PmRlbGV0ZSB0aGlzIG1lc3Nh
Z2UgYW5kIGl0cyBhdHRhY2htZW50cy48L2Rpdj4NCjxkaXY+QXMgZW1haWxzIG1heSBiZSBhbHRl
cmVkLCBPcmFuZ2UgaXMgbm90IGxpYWJsZSBmb3IgbWVzc2FnZXMgdGhhdCBoYXZlPC9kaXY+DQo8
ZGl2PmJlZW4gbW9kaWZpZWQsIGNoYW5nZWQgb3IgZmFsc2lmaWVkLjwvZGl2Pg0KPGRpdj5UaGFu
ayB5b3UuPC9kaXY+DQo8ZGl2Pl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fX19fPC9kaXY+DQo8ZGl2PkJFU1MgbWFpbGluZyBsaXN0PC9kaXY+DQo8ZGl2PjxhIGhy
ZWY9Im1haWx0bzpCRVNTQGlldGYub3JnIj5CRVNTQGlldGYub3JnPC9hPjwvZGl2Pg0KPGRpdj48
YSBocmVmPSJodHRwczovL3d3dy5pZXRmLm9yZy9tYWlsbWFuL2xpc3RpbmZvL2Jlc3MiPmh0dHBz
Oi8vd3d3LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vYmVzczwvYT48L2Rpdj4NCjwvYmxvY2tx
dW90ZT4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPC9kaXY+DQo8L3NwYW4+DQo8L2JvZHk+DQo8
L2h0bWw+DQo=

--_000_D13AD4DB2527Bkishoretjunipernet_--


From nobody Fri Mar 27 07:53:42 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 092191ACE82 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:53:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.611
X-Spam-Level: 
X-Spam-Status: No, score=-3.611 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7hxA8F5mAGKN for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:53:36 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 61B0D1ACE75 for <bess@ietf.org>; Fri, 27 Mar 2015 07:53:34 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT91981; Fri, 27 Mar 2015 14:53:33 +0000 (GMT)
Received: from nkgeml405-hub.china.huawei.com (10.98.56.36) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 14:53:31 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml405-hub.china.huawei.com ([10.98.56.36]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 22:53:27 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HYAABS9Z9AAAQdUUA==
Date: Fri, 27 Mar 2015 14:53:26 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857752@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com>, <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com>, <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.144.149]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/wRvUE9KVH4uWeQH7EdupdkZ22O8>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 14:53:41 -0000

Hi Jorge,
Sorry for sending again to correct some word error.

In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE3 is s=
ingle homed to remote PE3. CE1 and CE2 forms one layer 2 network, so this i=
s multi-homed network scenario. The picture is as follows:
           CE3
             |
            PE3

PE1                  PE2
  |                       |
CE1-----------CE2
In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends BUM=
 traffic to CE1, the traffic will reach PE1 and PE2, CE1 will receive two c=
opy of the traffic from CE3, also loop is formed, the Loop path is:

CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3

I think the traffic loop is absolutely not allowed. In MHN case, the loop w=
ill form due to dual-DF PE in transiet period.
Thanks,
weiguo



________________________________________
From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
Sent: Friday, March 27, 2015 20:26
To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com; bess@ietf=
.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Hi Weiguo,

About this:

[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
loop back to CE1 through PE2. If CE1 can filter and drop the traffic , it
is fortunate, maybe no loop.
For multi-homed network scenario, say local network1(CE1 and CE2) is
multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop back
through PE1-->PE2-->CE2, it is very serious. It is absolutely not allowed.


[JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2 will
not send the traffic back to CE1 due to the ESI-label and split-horizon
procedures.
In MH network case PE2 will NOT send the traffic to CE2 for the same
reason.

Let me know if you disagree please.

Thanks.
Jorge

-----Original Message-----
From: Haoweiguo <haoweiguo@huawei.com>
Date: Thursday, March 26, 2015 at 9:55 PM
To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
"thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org"
<bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
algorithm

>Hi Satya,
>Pls see inline.
>
>Thanks,
>weiguo
>
>________________________________________
>From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
>Sent: Friday, March 27, 2015 12:11
>To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>algorithm
>
>Thomas,
>
>Thanks for your summary. It captures the essence of the discussion.
>[weiguo]: Yes, i agree.
>
>One feature of the HRW scheme is that it computes the Active and the
>Backup DF upfront.
>Whenever the Active PE, say PE1, dies;
>the Backup PE, PE2, can become the
>Active immediately upon receiving the withdraw of the ES route from PE1.
>It need not even have to wait for the 3 seconds time.
>[weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
>message:) But when PE1 leaves ESI, it can withdraw the ES route
>immediately.
>When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2 will
>both act as PE for some transiet time.
>
>On the other hand, the 3 second timer will serve as a bulk and batch
>mechanism for the case when new PE(s) is(are) coming up, so we don=B9t hav=
e
>to do a DF election ever so often.
>
>Haoweiguo,  I think the "loop" scenario is speculative.
>I think everybody understands that, yes, there may be some transient
>duplication of bum traffic; however, there is no "loop" problem to be
>solved.
>[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
>and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>loop back to CE1 through PE2. If CE1 can filter and drop the traffic , it
>is fortunate, maybe no loop.
>For multi-homed network scenario, say local network1(CE1 and CE2) is
>multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
>allowed.
>
>Thanks,
>--Satya
>
>
>
>On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
>
>>Hi Thomas,
>>I know you are neutral, sorry for my reply vulnerable to be miunderstood.
>>
>>Thanks,
>>weiguo
>>
>>________________________________________
>>From: thomas.morin@orange.com [thomas.morin@orange.com]
>>Sent: Friday, March 27, 2015 11:59
>>To: Haoweiguo; bess@ietf.org
>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>algorithm
>>
>>Hi Weiguo,
>>
>>2015-03-26, Haoweiguo:
>> > Thomas:
>>> > - RFC7432 may have transient periods where the DF election state is
>>> > not yet synchronized between the two peers:
>> >
>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic loop
>>
>>Note well that I didn't write that there can be transient _loops_ .
>>I tried to capture the exchange you had all, and what I gather is that
>>the split-horizon procedure _prevents_loops_, independently of any
>>transient period where DF state is not synchronized and which may lead
>>to transient _duplicates_.
>>
>>-Thomas
>>
>>>
>>> 26/03/2015 20:05, John E Drake :
>>>>
>>>> Weiguo,
>>>>
>>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I have
>>>> detailed, is a non-solution to a non-problem with tremendous control
>>>> plane cost.
>>>>
>>>> Yours Irrespectively,
>>>>
>>>> John
>>>>
>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> Pls see below.
>>>>
>>>> Thanks,
>>>>
>>>> weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>> *From:*John E Drake [jdrake@juniper.net]
>>>> *Sent:* Friday, March 27, 2015 6:00
>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> To recap,
>>>>
>>>> We have established that your proposal is untenable because of its
>>>> control plane load.
>>>>
>>>> We have established that your proposal is based upon a flawed
>>>> understanding of the DF election in RFC 7432.
>>>>
>>>> [weiguo]: In ethernet world, traffic loop is serious than short timer
>>>> traffic disruption. If you want to implement  transiet traffic loop
>>>> process, i will modify my draft to solve your issue.
>>>>
>>>> If i am the developer, i will prefer short timer traffic disruption
>>>> based on current EVPN protocol.
>>>>
>>>> What you are now arguing is that your draft prevents two or more PEs
>>>> from being DF simultaneously. This is clearly nonsense.
>>>>
>>>> [weiguo]: I will modify the draft problem statements, and use the same
>>>> handshaking solution to solve it.
>>>>
>>>> Furthermore, we have established that having two or more DFs for what
>>>> even you admit is a brief transient leads to duplicate traffic, which
>>>> is acceptable, but not loops, your assertion to the contrary.
>>>>
>>>> [weiguo]: It is transient loop and traffic duplication issue.
>>>>
>>>> Yours Irrespectively,
>>>>
>>>> John
>>>>
>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> John,
>>>>
>>>> As your understanding of the EVPN draft, the DF election mechanism has
>>>> more serious side effect, it will have short time traffic loop,i.e.,
>>>> dual DF PEs will exist for a short time. I think dual DF PEs is
>>>> absolutely not tolerated, because native ethernet header has no TTL,
>>>> up to several hundred ms traffic loop normally not tolerated in
>>>> commertial networks.
>>>>
>>>> As your understanding, the PEs should do as following:
>>>>
>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have
>>>> rarely deployment.
>>>>
>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins
>>>> ESI and trigger DF re-election. When reception timer expires:
>>>>
>>>> PE1 upgrades to DF PE.
>>>>
>>>> After reception timer+ ES route transmission timer:
>>>>
>>>> PE2 downloads to non-DF PE.
>>>>
>>>> So in timer clock sync case, dual DF PEs will exist at least
>>>> transmission timer.
>>>>
>>>> If NTP is used for timer sync, because it has bad accuracy, dual DF
>>>> PEs will exist more longer timer.
>>>>
>>>> So as your understanding for DF election, the drawback is more clear.
>>>>
>>>> Thanks,
>>>>
>>>> weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>> *From:*John E Drake [jdrake@juniper.net]
>>>> *Sent:* Friday, March 27, 2015 4:41
>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> Weiguo,
>>>>
>>>> We have already established that your proposal is untenable because of
>>>> its control plane load.
>>>>
>>>> What we are now discussing is that your proposal is based upon a
>>>> misunderstanding of the algorithm in RFC 7432.  You are assuming that
>>>> PE1 will advertise an ES route and then wait for the configured
>>>> interval before performing the DF election while PE2 and PE3 will
>>>> perform the DF election as soon as they receive the ES route from PE1.
>>>> This is not what RFC 7432 says.
>>>>
>>>> Rather, what is says is that the advertisement of the ES route by PE1
>>>> and its receipt by PE2 and PE3 causes all three PEs to start the
>>>> configured interval timer  - =B33. When the timer expires, each PE
>>>> builds an ordered list of the IP addresses of all the PE nodes
>>>> connected to the Ethernet segment (including itself), in increasing
>>>> numeric value.=B2
>>>>
>>>> Yours Irrespectively,
>>>>
>>>> John
>>>>
>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> Pls read my detail replies to Satya. If you still can't catch it, pls
>>>> read my draft and EVPN base protocol,  thanks
>>>>
>>>> weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>> *From:*John E Drake [jdrake@juniper.net]
>>>> *Sent:* Friday, March 27, 2015 1:28
>>>> *To:* Patrice Brissette (pbrisset)
>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>> <mailto:bess@ietf.org>
>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>>>> Paxos algorithm
>>>>
>>>> I think Patrice is correct.  Your proposal doesn't solve the problem
>>>> and it does so at huge cost.
>>>>
>>>> Sent from my iPhone
>>>>
>>>>
>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>>>
>>>>      Weiguo,
>>>>
>>>>      I=B9m not sure I=B9m following here.
>>>>
>>>>      Don=B9t you have the same issue with your handshaking mechanism?
>>>>
>>>>      If you don=B9t know your peer, how can you handshake?
>>>>
>>>>      Regards,
>>>>
>>>>      Patrice
>>>>
>>>>      Image removed by sender.
>>>>
>>>>      *Patrice Brissette*
>>>>      TECHNICAL LEADER.ENGINEERING
>>>>
>>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>      Phone: *+1 613 254 3336*
>>>>
>>>>
>>>>
>>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>      Canada
>>>>      Cisco.com <http://www.cisco.com/global/CA/>
>>>>
>>>>
>>>>
>>>>      Image removed by sender.Think before you print.
>>>>
>>>>      This email may contain confidential and privileged material for
>>>>      the sole use of the intended recipient. Any review, use,
>>>>      distribution or disclosure by others is strictly prohibited. If
>>>>      you are not the intended recipient (or authorized to receive for
>>>>      the recipient), please contact the sender by reply email and
>>>>      delete all copies of this message.
>>>>
>>>>      Please click here
>>>>
>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>>      for Company Registration Information.
>>>>
>>>>
>>>>      *From: *Haoweiguo <haoweiguo@huawei.com
>>>><mailto:haoweiguo@huawei.com>>
>>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
>>>>      *To: *Patrice Brissette <pbrisset@cisco.com
>>>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES based
>>>>on
>>>>      Paxos algorithm
>>>>
>>>>          Hi Patrice,
>>>>
>>>>          Up to reception timer traffic disruption in transient phase
>>>>is
>>>>          one of the issues.
>>>>
>>>>          Thanks,
>>>>
>>>>          weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>>>          <mailto:pbrisset@cisco.com>]
>>>>          *Sent:* Thursday, March 26, 2015 20:54
>>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>>>          bess@ietf.org <mailto:bess@ietf.org>
>>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>>          based on Paxos algorithm
>>>>
>>>>          Weiguo,
>>>>
>>>>          You mention "But if your draft have not solved all issues=B2,
>>>>
>>>>          Can you explain what Satya=B9s draft is not solving?
>>>>
>>>>          Regards,
>>>>
>>>>          Patrice
>>>>
>>>>          Image removed by sender.
>>>>
>>>>          *Patrice Brissette*
>>>>          TECHNICAL LEADER.ENGINEERING
>>>>
>>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>          Phone: *+1 613 254 3336*
>>>>
>>>>
>>>>
>>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>          Canada
>>>>          Cisco.com <http://www.cisco.com/global/CA/>
>>>>
>>>>
>>>>
>>>>          Image removed by sender.Think before you print.
>>>>
>>>>          This email may contain confidential and privileged material
>>>>          for the sole use of the intended recipient. Any review, use,
>>>>          distribution or disclosure by others is strictly prohibited.
>>>>          If you are not the intended recipient (or authorized to
>>>>          receive for the recipient), please contact the sender by
>>>>reply
>>>>          email and delete all copies of this message.
>>>>
>>>>          Please click here
>>>>
>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>>          for Company Registration Information.
>>>>
>>>>
>>>>          *From: *Haoweiguo <haoweiguo@huawei.com
>>>>          <mailto:haoweiguo@huawei.com>>
>>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>>>          *To: *John E Drake <jdrake@juniper.net
>>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>          <mailto:bess@ietf.org>" <bess@ietf.org
>>>><mailto:bess@ietf.org>>
>>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>>>          based on Paxos algorithm
>>>>
>>>>              Hi John,
>>>>
>>>>               Firstly i think EVPN community should reach consensus on
>>>>              the issues of current DF election mechanism. All these
>>>>              issues should be resolved in a single new DF election
>>>>              draft,rather than in multiple separate drafts. If your
>>>>              draft can solve all these issues and stable, i have no
>>>>              question for its progressing. But if your draft have not
>>>>              solved all issues, i think it had better combine with
>>>>              other drafts to provide a comprehensive solution. I think
>>>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01 and
>>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it should
>>>>              be resolved. So i think although your new Hash algorithm
>>>>              for DF election is good, it only includes partial
>>>>              enhancements, maybe it still needs some time for
>>>>consensus.
>>>>
>>>>              Thanks,
>>>>
>>>>              weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>>              *From:*John E Drake [jdrake@juniper.net
>>>>              <mailto:jdrake@juniper.net>]
>>>>              *Sent:* Thursday, March 26, 2015 7:16
>>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>              <mailto:bess@ietf.org>
>>>>              *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES
>>>>              based on Paxos algorithm
>>>>
>>>>              Weiguo,
>>>>
>>>>              Your proposal introduces a control plane processing load
>>>>              that is O(#EVIs * PEs) per DF election and given that
>>>>              there can be 4K EVIs per ES, this looks like a
>>>>              **substantial** load. Furthermore,
>>>>
>>>>              you can=B9t  use the ES route to co-ordinate DF election
>>>>              because you would need to carry your new extended
>>>>              community for each EVI and they would not all fit.  You
>>>>              also can=B9t use the Per EVI Ethernet AD route because th=
at
>>>>              is processed by all PEs in the EVI.
>>>>
>>>>              I think that from a practical perspective the new DF
>>>>              election proposed in Satya=B9s draft is sufficiently stab=
le
>>>>              that it renders your draft moot, even if it could be made
>>>>              to work.
>>>>
>>>>              Yours Irrespectively,
>>>>
>>>>              John
>>>>
>>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf Of
>>>>              *Haoweiguo
>>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
>>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
>>>>              <mailto:bess@ietf.org>
>>>>              *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>>              based on Paxos algorithm
>>>>
>>>>              Hi Ali,
>>>>
>>>>              Thanks for your information. I scanned through this
>>>>draft,
>>>>              it really introduces inter-chassis message for DF
>>>>election
>>>>              handshaking, the requirements in this draft is to
>>>>              eliminate transiet Loop and traffic duplication. Current
>>>>              EVPN DF election mechanism already eliminated loop and
>>>>              traffic duplication by configuring long reception timer
>>>>on
>>>>              each multi-homed PE, but up to reception timer traffic
>>>>              disruption issue still exist. EVPN for DCI is an
>>>>important
>>>>              use case for EVPN, up to reception timer traffic
>>>>              disruption can't be tolerated for service providers, it
>>>>              should be improved.
>>>>
>>>>              Also for accuracy, i think handshaking state machine on
>>>>              each multi-homed PE is also needed. From solution
>>>>              perspective, in my draft, no inter-chassis message is
>>>>              introduced, only one new extended community is
>>>>introduced,
>>>>              i think the process is comparatively simple than your
>>>>              following draft.
>>>>
>>>>              Current EVPN DF election has some drawbacks, so there are
>>>>              three new drafts about DF election emerged. I think BESS
>>>>              WG can consider these three drafts in global view, a
>>>>              single,comprehensive new DF election draft is hoped.
>>>>
>>>>              thanks.
>>>>
>>>>              Thanks,
>>>>
>>>>              weiguo
>>>>
>>>>
>>>>-----------------------------------------------------------------------
>>>>-
>>>>
>>>>              *From:*BESS [bess-bounces@ietf.org
>>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajassi
>>>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>]
>>>>              *Sent:* Thursday, March 26, 2015 0:21
>>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
>>>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>>>              based on Paxos algorithm
>>>>
>>>>              FYI- First published July 4, 2011
>>>>
>>>>
>>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-route
>>>>/
>>>>
>>>>              -Ali
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> BESS mailing list
>>>> BESS@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/bess
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>________________________________________________________________________
>>>_
>>>________________________________________________
>>>
>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>confidentielles ou privilegiees et ne doivent donc
>>> pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>>>recu ce message par erreur, veuillez le signaler
>>> a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>>messages electroniques etant susceptibles d'alteration,
>>> Orange decline toute responsabilite si ce message a ete altere, deforme
>>>ou falsifie. Merci.
>>>
>>> This message and its attachments may contain confidential or privileged
>>>information that may be protected by law;
>>> they should not be distributed, used or copied without authorisation.
>>> If you have received this email in error, please notify the sender and
>>>delete this message and its attachments.
>>> As emails may be altered, Orange is not liable for messages that have
>>>been modified, changed or falsified.
>>> Thank you.
>>>
>>> _______________________________________________
>>> BESS mailing list
>>> BESS@ietf.org
>>> https://www.ietf.org/mailman/listinfo/bess
>>>
>>
>>
>>_________________________________________________________________________
>>_
>>_______________________________________________
>>
>>Ce message et ses pieces jointes peuvent contenir des informations
>>confidentielles ou privilegiees et ne doivent donc
>>pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>>recu ce message par erreur, veuillez le signaler
>>a l'expediteur et le detruire ainsi que les pieces jointes. Les messages
>>electroniques etant susceptibles d'alteration,
>>Orange decline toute responsabilite si ce message a ete altere, deforme
>>ou falsifie. Merci.
>>
>>This message and its attachments may contain confidential or privileged
>>information that may be protected by law;
>>they should not be distributed, used or copied without authorisation.
>>If you have received this email in error, please notify the sender and
>>delete this message and its attachments.
>>As emails may be altered, Orange is not liable for messages that have
>>been modified, changed or falsified.
>>Thank you.
>>_______________________________________________
>>BESS mailing list
>>BESS@ietf.org
>>https://www.ietf.org/mailman/listinfo/bess
>_______________________________________________
>BESS mailing list
>BESS@ietf.org
>https://www.ietf.org/mailman/listinfo/bess=


From nobody Fri Mar 27 07:55:50 2015
Return-Path: <jdrake@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 434D51ACEB5 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:55:49 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.302
X-Spam-Level: 
X-Spam-Status: No, score=-1.302 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hplJPN8g0fMa for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:55:43 -0700 (PDT)
Received: from na01-by2-obe.outbound.protection.outlook.com (mail-by2on0112.outbound.protection.outlook.com [207.46.100.112]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8CFEB1ACE8F for <bess@ietf.org>; Fri, 27 Mar 2015 07:55:39 -0700 (PDT)
Received: from BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) by BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) with Microsoft SMTP Server (TLS) id 15.1.118.21; Fri, 27 Mar 2015 14:55:36 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) by BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) with mapi id 15.01.0118.022; Fri, 27 Mar 2015 14:55:36 +0000
From: John E Drake <jdrake@juniper.net>
To: Haoweiguo <haoweiguo@huawei.com>, "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "EXT - thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8P/o146fJTTukSNC2GolBinQZ0u4lsAgAACyoCAAB/gjYAAILMAgAAMEtCAABidAIAAA48AgAAHqACAAC468IAAHjiAgAAP+ACAAAM7AIAAApwAgAAA7YCAAAw6gIAAfgsAgAAiUICAAAZsYA==
Date: Fri, 27 Mar 2015 14:55:35 +0000
Message-ID: <BLUPR05MB562A8E28D580CBCD24031C0C7090@BLUPR05MB562.namprd05.prod.outlook.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com>, <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [66.129.239.10]
authentication-results: huawei.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BLUPR05MB562;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(377424004)(24454002)(51704005)(479174004)(36304003)(69234005)(13464003)(164054003)(377454003)(50986999)(122556002)(86362001)(2900100001)(5890100001)(561944003)(2501003)(33656002)(76176999)(54356999)(46102003)(2950100001)(93886004)(107886001)(102836002)(15975445007)(77096005)(77156002)(62966003)(2656002)(74316001)(66066001)(92566002)(106116001)(19580395003)(19580405001)(76576001)(87936001)(99286002)(579004)(559001); DIR:OUT; SFP:1102; SCL:1; SRVR:BLUPR05MB562; H:BLUPR05MB562.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BLUPR05MB5626D3A409BC86E3D68CE29C7090@BLUPR05MB562.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BLUPR05MB562; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB562; 
x-forefront-prvs: 0528942FD8
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: juniper.net
X-MS-Exchange-CrossTenant-originalarrivaltime: 27 Mar 2015 14:55:36.0485 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BLUPR05MB562
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/3GVcH0p-fx0LPlKunH2BMxqeSY8>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 14:55:49 -0000

There's no loop.  Any BUM packet received from either CE1 or CE2 will not b=
e forwarded back to them because of the ES label.

Yours Irrespectively,

John

> -----Original Message-----
> From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
> Sent: Friday, March 27, 2015 10:30 AM
> To: Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT -
> thomas.morin@orange.com; bess@ietf.org
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>=20
> Hi Jorge,
> Sorry, maybe i have not explain clearly, pls allow me to clarify it more =
clearly.
> In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2 is
> single homed to remote PE3. CE1 and CE2 forms one layer 2 network, so it
> multi-homed network scenario.
>            CE3
>              |
>             PE3
>=20
>=20
> PE1                  PE2
>   |                       |
> CE1-----------CE2
>=20
> In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends B=
UM
> traffic to CE1, the traffic will reach PE1 and PE2, CE1 will receive two =
copy of
> the traffic from CE3, also loop is formed, the form path:
>=20
> CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
>=20
> I think the traffic loop is absolutely not allowed. In MHN case, the loop=
 will
> formed due to dual-DF PE in transiet period.
>=20
> Thanks,
> weiguo
>=20
>=20
>=20
> ________________________________________
> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
> Sent: Friday, March 27, 2015 20:26
> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
> bess@ietf.org
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>=20
> Hi Weiguo,
>=20
> About this:
>=20
> [weiguo]: Traffic duplication and loop issue both exist in dual DF PE cas=
e. In
> multi-homed device case, say CE1 is multi-homed to PE1 and PE2 and both P=
E
> are DF PE. When CE1 sends traffic to PE1, the traffic will loop back to C=
E1
> through PE2. If CE1 can filter and drop the traffic , it is fortunate, ma=
ybe no
> loop.
> For multi-homed network scenario, say local network1(CE1 and CE2) is mult=
i-
> homed to PE1 and PE2. Similarly, the traffic from CE1 will loop back thro=
ugh
> PE1-->PE2-->CE2, it is very serious. It is absolutely not allowed.
>=20
>=20
> [JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2
> will not send the traffic back to CE1 due to the ESI-label and split-hori=
zon
> procedures.
> In MH network case PE2 will NOT send the traffic to CE2 for the same reas=
on.
>=20
> Let me know if you disagree please.
>=20
> Thanks.
> Jorge
>=20
> -----Original Message-----
> From: Haoweiguo <haoweiguo@huawei.com>
> Date: Thursday, March 26, 2015 at 9:55 PM
> To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
> "thomas.morin@orange.com" <thomas.morin@orange.com>,
> "bess@ietf.org"
> <bess@ietf.org>
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>=20
> >Hi Satya,
> >Pls see inline.
> >
> >Thanks,
> >weiguo
> >
> >________________________________________
> >From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
> >Sent: Friday, March 27, 2015 12:11
> >To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
> >Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> >algorithm
> >
> >Thomas,
> >
> >Thanks for your summary. It captures the essence of the discussion.
> >[weiguo]: Yes, i agree.
> >
> >One feature of the HRW scheme is that it computes the Active and the
> >Backup DF upfront.
> >Whenever the Active PE, say PE1, dies;
> >the Backup PE, PE2, can become the
> >Active immediately upon receiving the withdraw of the ES route from PE1.
> >It need not even have to wait for the 3 seconds time.
> >[weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
> >message:) But when PE1 leaves ESI, it can withdraw the ES route
> >immediately.
> >When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2
> >will both act as PE for some transiet time.
> >
> >On the other hand, the 3 second timer will serve as a bulk and batch
> >mechanism for the case when new PE(s) is(are) coming up, so we don=B9t
> >have to do a DF election ever so often.
> >
> >Haoweiguo,  I think the "loop" scenario is speculative.
> >I think everybody understands that, yes, there may be some transient
> >duplication of bum traffic; however, there is no "loop" problem to be
> >solved.
> >[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
> >case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
> >and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
> >loop back to CE1 through PE2. If CE1 can filter and drop the traffic ,
> >it is fortunate, maybe no loop.
> >For multi-homed network scenario, say local network1(CE1 and CE2) is
> >multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
> >back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
> >allowed.
> >
> >Thanks,
> >--Satya
> >
> >
> >
> >On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
> >
> >>Hi Thomas,
> >>I know you are neutral, sorry for my reply vulnerable to be miunderstoo=
d.
> >>
> >>Thanks,
> >>weiguo
> >>
> >>________________________________________
> >>From: thomas.morin@orange.com [thomas.morin@orange.com]
> >>Sent: Friday, March 27, 2015 11:59
> >>To: Haoweiguo; bess@ietf.org
> >>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> >>algorithm
> >>
> >>Hi Weiguo,
> >>
> >>2015-03-26, Haoweiguo:
> >> > Thomas:
> >>> > - RFC7432 may have transient periods where the DF election state
> >>> > is not yet synchronized between the two peers:
> >> >
> >>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic
> >>> loop
> >>
> >>Note well that I didn't write that there can be transient _loops_ .
> >>I tried to capture the exchange you had all, and what I gather is that
> >>the split-horizon procedure _prevents_loops_, independently of any
> >>transient period where DF state is not synchronized and which may lead
> >>to transient _duplicates_.
> >>
> >>-Thomas
> >>
> >>>
> >>> 26/03/2015 20:05, John E Drake :
> >>>>
> >>>> Weiguo,
> >>>>
> >>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I have
> >>>> detailed, is a non-solution to a non-problem with tremendous
> >>>> control plane cost.
> >>>>
> >>>> Yours Irrespectively,
> >>>>
> >>>> John
> >>>>
> >>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>> *Sent:* Thursday, March 26, 2015 6:17 PM
> >>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> >>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> Pls see below.
> >>>>
> >>>> Thanks,
> >>>>
> >>>> weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>> *From:*John E Drake [jdrake@juniper.net]
> >>>> *Sent:* Friday, March 27, 2015 6:00
> >>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> >>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> To recap,
> >>>>
> >>>> We have established that your proposal is untenable because of its
> >>>> control plane load.
> >>>>
> >>>> We have established that your proposal is based upon a flawed
> >>>> understanding of the DF election in RFC 7432.
> >>>>
> >>>> [weiguo]: In ethernet world, traffic loop is serious than short
> >>>> timer traffic disruption. If you want to implement  transiet
> >>>> traffic loop process, i will modify my draft to solve your issue.
> >>>>
> >>>> If i am the developer, i will prefer short timer traffic disruption
> >>>> based on current EVPN protocol.
> >>>>
> >>>> What you are now arguing is that your draft prevents two or more
> >>>> PEs from being DF simultaneously. This is clearly nonsense.
> >>>>
> >>>> [weiguo]: I will modify the draft problem statements, and use the
> >>>> same handshaking solution to solve it.
> >>>>
> >>>> Furthermore, we have established that having two or more DFs for
> >>>> what even you admit is a brief transient leads to duplicate
> >>>> traffic, which is acceptable, but not loops, your assertion to the
> contrary.
> >>>>
> >>>> [weiguo]: It is transient loop and traffic duplication issue.
> >>>>
> >>>> Yours Irrespectively,
> >>>>
> >>>> John
> >>>>
> >>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>> *Sent:* Thursday, March 26, 2015 5:37 PM
> >>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> John,
> >>>>
> >>>> As your understanding of the EVPN draft, the DF election mechanism
> >>>> has more serious side effect, it will have short time traffic
> >>>> loop,i.e., dual DF PEs will exist for a short time. I think dual DF
> >>>> PEs is absolutely not tolerated, because native ethernet header has
> >>>> no TTL, up to several hundred ms traffic loop normally not
> >>>> tolerated in commertial networks.
> >>>>
> >>>> As your understanding, the PEs should do as following:
> >>>>
> >>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but
> >>>> have rarely deployment.
> >>>>
> >>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3
> >>>> joins ESI and trigger DF re-election. When reception timer expires:
> >>>>
> >>>> PE1 upgrades to DF PE.
> >>>>
> >>>> After reception timer+ ES route transmission timer:
> >>>>
> >>>> PE2 downloads to non-DF PE.
> >>>>
> >>>> So in timer clock sync case, dual DF PEs will exist at least
> >>>> transmission timer.
> >>>>
> >>>> If NTP is used for timer sync, because it has bad accuracy, dual DF
> >>>> PEs will exist more longer timer.
> >>>>
> >>>> So as your understanding for DF election, the drawback is more clear=
.
> >>>>
> >>>> Thanks,
> >>>>
> >>>> weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>> *From:*John E Drake [jdrake@juniper.net]
> >>>> *Sent:* Friday, March 27, 2015 4:41
> >>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> >>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> Weiguo,
> >>>>
> >>>> We have already established that your proposal is untenable because
> >>>> of its control plane load.
> >>>>
> >>>> What we are now discussing is that your proposal is based upon a
> >>>> misunderstanding of the algorithm in RFC 7432.  You are assuming
> >>>> that
> >>>> PE1 will advertise an ES route and then wait for the configured
> >>>> interval before performing the DF election while PE2 and PE3 will
> >>>> perform the DF election as soon as they receive the ES route from PE=
1.
> >>>> This is not what RFC 7432 says.
> >>>>
> >>>> Rather, what is says is that the advertisement of the ES route by
> >>>> PE1 and its receipt by PE2 and PE3 causes all three PEs to start
> >>>> the configured interval timer  - =B33. When the timer expires, each
> >>>> PE builds an ordered list of the IP addresses of all the PE nodes
> >>>> connected to the Ethernet segment (including itself), in increasing
> >>>> numeric value.=B2
> >>>>
> >>>> Yours Irrespectively,
> >>>>
> >>>> John
> >>>>
> >>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>> *Sent:* Thursday, March 26, 2015 3:26 PM
> >>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> Pls read my detail replies to Satya. If you still can't catch it,
> >>>> pls read my draft and EVPN base protocol,  thanks
> >>>>
> >>>> weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>> *From:*John E Drake [jdrake@juniper.net]
> >>>> *Sent:* Friday, March 27, 2015 1:28
> >>>> *To:* Patrice Brissette (pbrisset)
> >>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
> >>>> <mailto:bess@ietf.org>
> >>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> I think Patrice is correct.  Your proposal doesn't solve the
> >>>> problem and it does so at huge cost.
> >>>>
> >>>> Sent from my iPhone
> >>>>
> >>>>
> >>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
> >>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
> >>>>
> >>>>      Weiguo,
> >>>>
> >>>>      I=B9m not sure I=B9m following here.
> >>>>
> >>>>      Don=B9t you have the same issue with your handshaking mechanism=
?
> >>>>
> >>>>      If you don=B9t know your peer, how can you handshake?
> >>>>
> >>>>      Regards,
> >>>>
> >>>>      Patrice
> >>>>
> >>>>      Image removed by sender.
> >>>>
> >>>>      *Patrice Brissette*
> >>>>      TECHNICAL LEADER.ENGINEERING
> >>>>
> >>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
> >>>>      Phone: *+1 613 254 3336*
> >>>>
> >>>>
> >>>>
> >>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
> >>>>      Canada
> >>>>      Cisco.com <http://www.cisco.com/global/CA/>
> >>>>
> >>>>
> >>>>
> >>>>      Image removed by sender.Think before you print.
> >>>>
> >>>>      This email may contain confidential and privileged material for
> >>>>      the sole use of the intended recipient. Any review, use,
> >>>>      distribution or disclosure by others is strictly prohibited. If
> >>>>      you are not the intended recipient (or authorized to receive fo=
r
> >>>>      the recipient), please contact the sender by reply email and
> >>>>      delete all copies of this message.
> >>>>
> >>>>      Please click here
> >>>>
> >>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.ht
> ml>
> >>>>      for Company Registration Information.
> >>>>
> >>>>
> >>>>      *From: *Haoweiguo <haoweiguo@huawei.com
> >>>><mailto:haoweiguo@huawei.com>>
> >>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
> >>>>      *To: *Patrice Brissette <pbrisset@cisco.com
> >>>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
> >>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
> >>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
> >>>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
> >>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES
> >>>>based on
> >>>>      Paxos algorithm
> >>>>
> >>>>          Hi Patrice,
> >>>>
> >>>>          Up to reception timer traffic disruption in transient
> >>>>phase is
> >>>>          one of the issues.
> >>>>
> >>>>          Thanks,
> >>>>
> >>>>          weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
> >>>>          <mailto:pbrisset@cisco.com>]
> >>>>          *Sent:* Thursday, March 26, 2015 20:54
> >>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
> >>>>          bess@ietf.org <mailto:bess@ietf.org>
> >>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
> >>>>          based on Paxos algorithm
> >>>>
> >>>>          Weiguo,
> >>>>
> >>>>          You mention "But if your draft have not solved all
> >>>> issues=B2,
> >>>>
> >>>>          Can you explain what Satya=B9s draft is not solving?
> >>>>
> >>>>          Regards,
> >>>>
> >>>>          Patrice
> >>>>
> >>>>          Image removed by sender.
> >>>>
> >>>>          *Patrice Brissette*
> >>>>          TECHNICAL LEADER.ENGINEERING
> >>>>
> >>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
> >>>>          Phone: *+1 613 254 3336*
> >>>>
> >>>>
> >>>>
> >>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
> >>>>          Canada
> >>>>          Cisco.com <http://www.cisco.com/global/CA/>
> >>>>
> >>>>
> >>>>
> >>>>          Image removed by sender.Think before you print.
> >>>>
> >>>>          This email may contain confidential and privileged material
> >>>>          for the sole use of the intended recipient. Any review, use=
,
> >>>>          distribution or disclosure by others is strictly prohibited=
.
> >>>>          If you are not the intended recipient (or authorized to
> >>>>          receive for the recipient), please contact the sender by
> >>>>reply
> >>>>          email and delete all copies of this message.
> >>>>
> >>>>          Please click here
> >>>>
> >>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.ht
> ml>
> >>>>          for Company Registration Information.
> >>>>
> >>>>
> >>>>          *From: *Haoweiguo <haoweiguo@huawei.com
> >>>>          <mailto:haoweiguo@huawei.com>>
> >>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
> >>>>          *To: *John E Drake <jdrake@juniper.net
> >>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.co=
m
> >>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
> >>>>          <mailto:bess@ietf.org>" <bess@ietf.org
> >>>><mailto:bess@ietf.org>>
> >>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
> >>>>          based on Paxos algorithm
> >>>>
> >>>>              Hi John,
> >>>>
> >>>>               Firstly i think EVPN community should reach consensus =
on
> >>>>              the issues of current DF election mechanism. All these
> >>>>              issues should be resolved in a single new DF election
> >>>>              draft,rather than in multiple separate drafts. If your
> >>>>              draft can solve all these issues and stable, i have no
> >>>>              question for its progressing. But if your draft have no=
t
> >>>>              solved all issues, i think it had better combine with
> >>>>              other drafts to provide a comprehensive solution. I thi=
nk
> >>>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01 a=
nd
> >>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it shou=
ld
> >>>>              be resolved. So i think although your new Hash algorith=
m
> >>>>              for DF election is good, it only includes partial
> >>>>              enhancements, maybe it still needs some time for
> >>>>consensus.
> >>>>
> >>>>              Thanks,
> >>>>
> >>>>              weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>>              *From:*John E Drake [jdrake@juniper.net
> >>>>              <mailto:jdrake@juniper.net>]
> >>>>              *Sent:* Thursday, March 26, 2015 7:16
> >>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
> >>>>              <mailto:bess@ietf.org>
> >>>>              *Subject:* RE: [bess] Handshaking among PEs in an EVPN =
ES
> >>>>              based on Paxos algorithm
> >>>>
> >>>>              Weiguo,
> >>>>
> >>>>              Your proposal introduces a control plane processing loa=
d
> >>>>              that is O(#EVIs * PEs) per DF election and given that
> >>>>              there can be 4K EVIs per ES, this looks like a
> >>>>              **substantial** load. Furthermore,
> >>>>
> >>>>              you can=B9t  use the ES route to co-ordinate DF electio=
n
> >>>>              because you would need to carry your new extended
> >>>>              community for each EVI and they would not all fit.  You
> >>>>              also can=B9t use the Per EVI Ethernet AD route because =
that
> >>>>              is processed by all PEs in the EVI.
> >>>>
> >>>>              I think that from a practical perspective the new DF
> >>>>              election proposed in Satya=B9s draft is sufficiently st=
able
> >>>>              that it renders your draft moot, even if it could be ma=
de
> >>>>              to work.
> >>>>
> >>>>              Yours Irrespectively,
> >>>>
> >>>>              John
> >>>>
> >>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf O=
f
> >>>>              *Haoweiguo
> >>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
> >>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
> >>>>              <mailto:bess@ietf.org>
> >>>>              *Subject:* Re: [bess] Handshaking among PEs in an EVPN =
ES
> >>>>              based on Paxos algorithm
> >>>>
> >>>>              Hi Ali,
> >>>>
> >>>>              Thanks for your information. I scanned through this
> >>>>draft,
> >>>>              it really introduces inter-chassis message for DF
> >>>>election
> >>>>              handshaking, the requirements in this draft is to
> >>>>              eliminate transiet Loop and traffic duplication. Curren=
t
> >>>>              EVPN DF election mechanism already eliminated loop and
> >>>>              traffic duplication by configuring long reception
> >>>>timer on
> >>>>              each multi-homed PE, but up to reception timer traffic
> >>>>              disruption issue still exist. EVPN for DCI is an
> >>>>important
> >>>>              use case for EVPN, up to reception timer traffic
> >>>>              disruption can't be tolerated for service providers, it
> >>>>              should be improved.
> >>>>
> >>>>              Also for accuracy, i think handshaking state machine on
> >>>>              each multi-homed PE is also needed. From solution
> >>>>              perspective, in my draft, no inter-chassis message is
> >>>>              introduced, only one new extended community is
> >>>>introduced,
> >>>>              i think the process is comparatively simple than your
> >>>>              following draft.
> >>>>
> >>>>              Current EVPN DF election has some drawbacks, so there a=
re
> >>>>              three new drafts about DF election emerged. I think BES=
S
> >>>>              WG can consider these three drafts in global view, a
> >>>>              single,comprehensive new DF election draft is hoped.
> >>>>
> >>>>              thanks.
> >>>>
> >>>>              Thanks,
> >>>>
> >>>>              weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>>              *From:*BESS [bess-bounces@ietf.org
> >>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajass=
i
> >>>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>=
]
> >>>>              *Sent:* Thursday, March 26, 2015 0:21
> >>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
> >>>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
> >>>>              based on Paxos algorithm
> >>>>
> >>>>              FYI- First published July 4, 2011
> >>>>
> >>>>
> >>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-ro
> >>>>ute
> >>>>/
> >>>>
> >>>>              -Ali
> >>>>
> >>>>
> >>>>
> >>>> _______________________________________________
> >>>> BESS mailing list
> >>>> BESS@ietf.org
> >>>> https://www.ietf.org/mailman/listinfo/bess
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>_______________________________________________________
> ______________
> >>>___
> >>>_
> >>>________________________________________________
> >>>
> >>> Ce message et ses pieces jointes peuvent contenir des informations
> >>>confidentielles ou privilegiees et ne doivent donc  pas etre
> >>>diffuses, exploites ou copies sans autorisation. Si vous avez recu ce
> >>>message par erreur, veuillez le signaler  a l'expediteur et le
> >>>detruire ainsi que les pieces jointes. Les messages electroniques
> >>>etant susceptibles d'alteration,  Orange decline toute responsabilite
> >>>si ce message a ete altere, deforme ou falsifie. Merci.
> >>>
> >>> This message and its attachments may contain confidential or
> >>>privileged information that may be protected by law;  they should not
> >>>be distributed, used or copied without authorisation.
> >>> If you have received this email in error, please notify the sender
> >>>and delete this message and its attachments.
> >>> As emails may be altered, Orange is not liable for messages that
> >>>have been modified, changed or falsified.
> >>> Thank you.
> >>>
> >>> _______________________________________________
> >>> BESS mailing list
> >>> BESS@ietf.org
> >>> https://www.ietf.org/mailman/listinfo/bess
> >>>
> >>
> >>
> >>________________________________________________________
> ______________
> >>___
> >>_
> >>_______________________________________________
> >>
> >>Ce message et ses pieces jointes peuvent contenir des informations
> >>confidentielles ou privilegiees et ne doivent donc pas etre diffuses,
> >>exploites ou copies sans autorisation. Si vous avez recu ce message
> >>par erreur, veuillez le signaler a l'expediteur et le detruire ainsi
> >>que les pieces jointes. Les messages electroniques etant susceptibles
> >>d'alteration, Orange decline toute responsabilite si ce message a ete
> >>altere, deforme ou falsifie. Merci.
> >>
> >>This message and its attachments may contain confidential or
> >>privileged information that may be protected by law; they should not
> >>be distributed, used or copied without authorisation.
> >>If you have received this email in error, please notify the sender and
> >>delete this message and its attachments.
> >>As emails may be altered, Orange is not liable for messages that have
> >>been modified, changed or falsified.
> >>Thank you.
> >>_______________________________________________
> >>BESS mailing list
> >>BESS@ietf.org
> >>https://www.ietf.org/mailman/listinfo/bess
> >_______________________________________________
> >BESS mailing list
> >BESS@ietf.org
> >https://www.ietf.org/mailman/listinfo/bess
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess


From nobody Fri Mar 27 08:00:00 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1A87D1ACE8F for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:59:51 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5bu8KZOBbqbM for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 07:59:46 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9E7721ACE8A for <bess@ietf.org>; Fri, 27 Mar 2015 07:59:40 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id 2E7C5D6415C92; Fri, 27 Mar 2015 14:59:36 +0000 (GMT)
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2RExcVw023120 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 15:59:38 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR712WXCHHUB03.zeu.alcatel-lucent.com ([135.239.2.74]) with mapi id 14.03.0195.001; Fri, 27 Mar 2015 15:59:38 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Haoweiguo <haoweiguo@huawei.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8QfEaq8IGmBqka8wmUCt0qdaZ0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HOAABL1oIAADZ/DgA==
Date: Fri, 27 Mar 2015 14:59:37 +0000
Message-ID: <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.40]
Content-Type: text/plain; charset="utf-8"
Content-ID: <75F8FEBDE000804AA191B5EC2BA7A351@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/FjNAkW0GYc6xrBDWsWt0hNXgH3I>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 14:59:51 -0000
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From nobody Fri Mar 27 08:11:48 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 464401A0282 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:11:46 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.611
X-Spam-Level: 
X-Spam-Status: No, score=-3.611 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id CG3--t3u4IUe for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:11:41 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id DA3291A00DF for <bess@ietf.org>; Fri, 27 Mar 2015 08:11:40 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml405-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT93536; Fri, 27 Mar 2015 15:11:39 +0000 (GMT)
Received: from nkgeml409-hub.china.huawei.com (10.98.56.40) by lhreml405-hub.china.huawei.com (10.201.5.242) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 15:11:37 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml409-hub.china.huawei.com ([10.98.56.40]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 23:11:31 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HYAABS9Z9AAD2dSgAAN/i8V
Date: Fri, 27 Mar 2015 15:11:30 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>, <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com>
In-Reply-To: <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.144.149]
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/cEs1nNi3wJ2PEf40dAFVGbta6eo>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 15:11:46 -0000

Hi Jorge,
Let give you another indefinitely forwarding loop case, the picture is as f=
ollows:

CE3------------CE4
  |                        |
PE3                   PE4



PE1                   PE2
 |                         |
CE1------------CE2

If PE1 and PE2 are dual DF PE in transiet time, assuing PE3 and PE4 are in =
stable state ,PE3 is DF PE, PE4 is non-DF PE, the BUM traffic from CE3 to C=
E1, the forwarding loop path is as follows;
CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->PE4---->=
forever......

Do you think the harm is enough?
Thanks,
weiguo
________________________________________
From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
Sent: Friday, March 27, 2015 22:59
To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com; bess@ietf=
.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Hi Weiguo,

OK, thanks for clarifying, that is the part I did not understand in your
loop :-)

To me a loop means you send a frame and that frame keeps being forwarded
by the same nodes indefinitely till you take a corrective action.
In this case, CE3 might get a few packets during the transient state, but
those packets will have its own MAC SA so CE3 will discard them. I don=92t
think that is harmful for CE3 in most of the cases. And if it is, you just
need to make sure the timers are long enough and the DF election properly
implemented to avoid the transient DF-DF state.

The handshaking is definitely overkilling IMHO.

Thanks.
Jorge

-----Original Message-----
From: Haoweiguo <haoweiguo@huawei.com>
Date: Friday, March 27, 2015 at 7:29 AM
To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
(satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
<thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
algorithm

>Hi Jorge,
>Sorry, maybe i have not explain clearly, pls allow me to clarify it more
>clearly.
>In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2 is
>single homed to remote PE3. CE1 and CE2 forms one layer 2 network, so it
>multi-homed network scenario.
>           CE3
>             |
>            PE3
>
>
>PE1                  PE2
>  |                       |
>CE1-----------CE2
>
>In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends
>BUM traffic to CE1, the traffic will reach PE1 and PE2, CE1 will receive
>two copy of the traffic from CE3, also loop is formed, the form path:
>
>CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
>
>I think the traffic loop is absolutely not allowed. In MHN case, the loop
>will formed due to dual-DF PE in transiet period.
>
>Thanks,
>weiguo
>
>
>
>________________________________________
>From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>Sent: Friday, March 27, 2015 20:26
>To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>bess@ietf.org
>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>algorithm
>
>Hi Weiguo,
>
>About this:
>
>[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
>and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>loop back to CE1 through PE2. If CE1 can filter and drop the traffic , it
>is fortunate, maybe no loop.
>For multi-homed network scenario, say local network1(CE1 and CE2) is
>multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop back
>through PE1-->PE2-->CE2, it is very serious. It is absolutely not allowed.
>
>
>[JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2 will
>not send the traffic back to CE1 due to the ESI-label and split-horizon
>procedures.
>In MH network case PE2 will NOT send the traffic to CE2 for the same
>reason.
>
>Let me know if you disagree please.
>
>Thanks.
>Jorge
>
>-----Original Message-----
>From: Haoweiguo <haoweiguo@huawei.com>
>Date: Thursday, March 26, 2015 at 9:55 PM
>To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
>"thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org"
><bess@ietf.org>
>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>algorithm
>
>>Hi Satya,
>>Pls see inline.
>>
>>Thanks,
>>weiguo
>>
>>________________________________________
>>From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
>>Sent: Friday, March 27, 2015 12:11
>>To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>algorithm
>>
>>Thomas,
>>
>>Thanks for your summary. It captures the essence of the discussion.
>>[weiguo]: Yes, i agree.
>>
>>One feature of the HRW scheme is that it computes the Active and the
>>Backup DF upfront.
>>Whenever the Active PE, say PE1, dies;
>>the Backup PE, PE2, can become the
>>Active immediately upon receiving the withdraw of the ES route from PE1.
>>It need not even have to wait for the 3 seconds time.
>>[weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
>>message:) But when PE1 leaves ESI, it can withdraw the ES route
>>immediately.
>>When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2 will
>>both act as PE for some transiet time.
>>
>>On the other hand, the 3 second timer will serve as a bulk and batch
>>mechanism for the case when new PE(s) is(are) coming up, so we don=B9t ha=
ve
>>to do a DF election ever so often.
>>
>>Haoweiguo,  I think the "loop" scenario is speculative.
>>I think everybody understands that, yes, there may be some transient
>>duplication of bum traffic; however, there is no "loop" problem to be
>>solved.
>>[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
>>and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>>loop back to CE1 through PE2. If CE1 can filter and drop the traffic , it
>>is fortunate, maybe no loop.
>>For multi-homed network scenario, say local network1(CE1 and CE2) is
>>multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
>>allowed.
>>
>>Thanks,
>>--Satya
>>
>>
>>
>>On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
>>
>>>Hi Thomas,
>>>I know you are neutral, sorry for my reply vulnerable to be
>>>miunderstood.
>>>
>>>Thanks,
>>>weiguo
>>>
>>>________________________________________
>>>From: thomas.morin@orange.com [thomas.morin@orange.com]
>>>Sent: Friday, March 27, 2015 11:59
>>>To: Haoweiguo; bess@ietf.org
>>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>algorithm
>>>
>>>Hi Weiguo,
>>>
>>>2015-03-26, Haoweiguo:
>>> > Thomas:
>>>> > - RFC7432 may have transient periods where the DF election state is
>>>> > not yet synchronized between the two peers:
>>> >
>>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic loop
>>>
>>>Note well that I didn't write that there can be transient _loops_ .
>>>I tried to capture the exchange you had all, and what I gather is that
>>>the split-horizon procedure _prevents_loops_, independently of any
>>>transient period where DF state is not synchronized and which may lead
>>>to transient _duplicates_.
>>>
>>>-Thomas
>>>
>>>>
>>>> 26/03/2015 20:05, John E Drake :
>>>>>
>>>>> Weiguo,
>>>>>
>>>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I have
>>>>> detailed, is a non-solution to a non-problem with tremendous control
>>>>> plane cost.
>>>>>
>>>>> Yours Irrespectively,
>>>>>
>>>>> John
>>>>>
>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>> Paxos algorithm
>>>>>
>>>>> Pls see below.
>>>>>
>>>>> Thanks,
>>>>>
>>>>> weiguo
>>>>>
>>>>>
>>>>>----------------------------------------------------------------------
>>>>>-
>>>>>-
>>>>>
>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>> *Sent:* Friday, March 27, 2015 6:00
>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>> Paxos algorithm
>>>>>
>>>>> To recap,
>>>>>
>>>>> We have established that your proposal is untenable because of its
>>>>> control plane load.
>>>>>
>>>>> We have established that your proposal is based upon a flawed
>>>>> understanding of the DF election in RFC 7432.
>>>>>
>>>>> [weiguo]: In ethernet world, traffic loop is serious than short timer
>>>>> traffic disruption. If you want to implement  transiet traffic loop
>>>>> process, i will modify my draft to solve your issue.
>>>>>
>>>>> If i am the developer, i will prefer short timer traffic disruption
>>>>> based on current EVPN protocol.
>>>>>
>>>>> What you are now arguing is that your draft prevents two or more PEs
>>>>> from being DF simultaneously. This is clearly nonsense.
>>>>>
>>>>> [weiguo]: I will modify the draft problem statements, and use the
>>>>>same
>>>>> handshaking solution to solve it.
>>>>>
>>>>> Furthermore, we have established that having two or more DFs for what
>>>>> even you admit is a brief transient leads to duplicate traffic, which
>>>>> is acceptable, but not loops, your assertion to the contrary.
>>>>>
>>>>> [weiguo]: It is transient loop and traffic duplication issue.
>>>>>
>>>>> Yours Irrespectively,
>>>>>
>>>>> John
>>>>>
>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>> Paxos algorithm
>>>>>
>>>>> John,
>>>>>
>>>>> As your understanding of the EVPN draft, the DF election mechanism
>>>>>has
>>>>> more serious side effect, it will have short time traffic loop,i.e.,
>>>>> dual DF PEs will exist for a short time. I think dual DF PEs is
>>>>> absolutely not tolerated, because native ethernet header has no TTL,
>>>>> up to several hundred ms traffic loop normally not tolerated in
>>>>> commertial networks.
>>>>>
>>>>> As your understanding, the PEs should do as following:
>>>>>
>>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have
>>>>> rarely deployment.
>>>>>
>>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins
>>>>> ESI and trigger DF re-election. When reception timer expires:
>>>>>
>>>>> PE1 upgrades to DF PE.
>>>>>
>>>>> After reception timer+ ES route transmission timer:
>>>>>
>>>>> PE2 downloads to non-DF PE.
>>>>>
>>>>> So in timer clock sync case, dual DF PEs will exist at least
>>>>> transmission timer.
>>>>>
>>>>> If NTP is used for timer sync, because it has bad accuracy, dual DF
>>>>> PEs will exist more longer timer.
>>>>>
>>>>> So as your understanding for DF election, the drawback is more clear.
>>>>>
>>>>> Thanks,
>>>>>
>>>>> weiguo
>>>>>
>>>>>
>>>>>----------------------------------------------------------------------
>>>>>-
>>>>>-
>>>>>
>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>> *Sent:* Friday, March 27, 2015 4:41
>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>> Paxos algorithm
>>>>>
>>>>> Weiguo,
>>>>>
>>>>> We have already established that your proposal is untenable because
>>>>>of
>>>>> its control plane load.
>>>>>
>>>>> What we are now discussing is that your proposal is based upon a
>>>>> misunderstanding of the algorithm in RFC 7432.  You are assuming that
>>>>> PE1 will advertise an ES route and then wait for the configured
>>>>> interval before performing the DF election while PE2 and PE3 will
>>>>> perform the DF election as soon as they receive the ES route from
>>>>>PE1.
>>>>> This is not what RFC 7432 says.
>>>>>
>>>>> Rather, what is says is that the advertisement of the ES route by PE1
>>>>> and its receipt by PE2 and PE3 causes all three PEs to start the
>>>>> configured interval timer  - =B33. When the timer expires, each PE
>>>>> builds an ordered list of the IP addresses of all the PE nodes
>>>>> connected to the Ethernet segment (including itself), in increasing
>>>>> numeric value.=B2
>>>>>
>>>>> Yours Irrespectively,
>>>>>
>>>>> John
>>>>>
>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>> Paxos algorithm
>>>>>
>>>>> Pls read my detail replies to Satya. If you still can't catch it, pls
>>>>> read my draft and EVPN base protocol,  thanks
>>>>>
>>>>> weiguo
>>>>>
>>>>>
>>>>>----------------------------------------------------------------------
>>>>>-
>>>>>-
>>>>>
>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>> *Sent:* Friday, March 27, 2015 1:28
>>>>> *To:* Patrice Brissette (pbrisset)
>>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>> <mailto:bess@ietf.org>
>>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>>>>> Paxos algorithm
>>>>>
>>>>> I think Patrice is correct.  Your proposal doesn't solve the problem
>>>>> and it does so at huge cost.
>>>>>
>>>>> Sent from my iPhone
>>>>>
>>>>>
>>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>>>>
>>>>>      Weiguo,
>>>>>
>>>>>      I=B9m not sure I=B9m following here.
>>>>>
>>>>>      Don=B9t you have the same issue with your handshaking mechanism?
>>>>>
>>>>>      If you don=B9t know your peer, how can you handshake?
>>>>>
>>>>>      Regards,
>>>>>
>>>>>      Patrice
>>>>>
>>>>>      Image removed by sender.
>>>>>
>>>>>      *Patrice Brissette*
>>>>>      TECHNICAL LEADER.ENGINEERING
>>>>>
>>>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>      Phone: *+1 613 254 3336*
>>>>>
>>>>>
>>>>>
>>>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>      Canada
>>>>>      Cisco.com <http://www.cisco.com/global/CA/>
>>>>>
>>>>>
>>>>>
>>>>>      Image removed by sender.Think before you print.
>>>>>
>>>>>      This email may contain confidential and privileged material for
>>>>>      the sole use of the intended recipient. Any review, use,
>>>>>      distribution or disclosure by others is strictly prohibited. If
>>>>>      you are not the intended recipient (or authorized to receive for
>>>>>      the recipient), please contact the sender by reply email and
>>>>>      delete all copies of this message.
>>>>>
>>>>>      Please click here
>>>>>
>>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>>>      for Company Registration Information.
>>>>>
>>>>>
>>>>>      *From: *Haoweiguo <haoweiguo@huawei.com
>>>>><mailto:haoweiguo@huawei.com>>
>>>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
>>>>>      *To: *Patrice Brissette <pbrisset@cisco.com
>>>>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES based
>>>>>on
>>>>>      Paxos algorithm
>>>>>
>>>>>          Hi Patrice,
>>>>>
>>>>>          Up to reception timer traffic disruption in transient phase
>>>>>is
>>>>>          one of the issues.
>>>>>
>>>>>          Thanks,
>>>>>
>>>>>          weiguo
>>>>>
>>>>>
>>>>>----------------------------------------------------------------------
>>>>>-
>>>>>-
>>>>>
>>>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>>>>          <mailto:pbrisset@cisco.com>]
>>>>>          *Sent:* Thursday, March 26, 2015 20:54
>>>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>>>>          bess@ietf.org <mailto:bess@ietf.org>
>>>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>          based on Paxos algorithm
>>>>>
>>>>>          Weiguo,
>>>>>
>>>>>          You mention "But if your draft have not solved all issues=B2=
,
>>>>>
>>>>>          Can you explain what Satya=B9s draft is not solving?
>>>>>
>>>>>          Regards,
>>>>>
>>>>>          Patrice
>>>>>
>>>>>          Image removed by sender.
>>>>>
>>>>>          *Patrice Brissette*
>>>>>          TECHNICAL LEADER.ENGINEERING
>>>>>
>>>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>          Phone: *+1 613 254 3336*
>>>>>
>>>>>
>>>>>
>>>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>          Canada
>>>>>          Cisco.com <http://www.cisco.com/global/CA/>
>>>>>
>>>>>
>>>>>
>>>>>          Image removed by sender.Think before you print.
>>>>>
>>>>>          This email may contain confidential and privileged material
>>>>>          for the sole use of the intended recipient. Any review, use,
>>>>>          distribution or disclosure by others is strictly prohibited.
>>>>>          If you are not the intended recipient (or authorized to
>>>>>          receive for the recipient), please contact the sender by
>>>>>reply
>>>>>          email and delete all copies of this message.
>>>>>
>>>>>          Please click here
>>>>>
>>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>>>          for Company Registration Information.
>>>>>
>>>>>
>>>>>          *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>          <mailto:haoweiguo@huawei.com>>
>>>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>>>>          *To: *John E Drake <jdrake@juniper.net
>>>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>          <mailto:bess@ietf.org>" <bess@ietf.org
>>>>><mailto:bess@ietf.org>>
>>>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>          based on Paxos algorithm
>>>>>
>>>>>              Hi John,
>>>>>
>>>>>               Firstly i think EVPN community should reach consensus
>>>>>on
>>>>>              the issues of current DF election mechanism. All these
>>>>>              issues should be resolved in a single new DF election
>>>>>              draft,rather than in multiple separate drafts. If your
>>>>>              draft can solve all these issues and stable, i have no
>>>>>              question for its progressing. But if your draft have not
>>>>>              solved all issues, i think it had better combine with
>>>>>              other drafts to provide a comprehensive solution. I
>>>>>think
>>>>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01
>>>>>and
>>>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it
>>>>>should
>>>>>              be resolved. So i think although your new Hash algorithm
>>>>>              for DF election is good, it only includes partial
>>>>>              enhancements, maybe it still needs some time for
>>>>>consensus.
>>>>>
>>>>>              Thanks,
>>>>>
>>>>>              weiguo
>>>>>
>>>>>
>>>>>----------------------------------------------------------------------
>>>>>-
>>>>>-
>>>>>
>>>>>              *From:*John E Drake [jdrake@juniper.net
>>>>>              <mailto:jdrake@juniper.net>]
>>>>>              *Sent:* Thursday, March 26, 2015 7:16
>>>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>              <mailto:bess@ietf.org>
>>>>>              *Subject:* RE: [bess] Handshaking among PEs in an EVPN
>>>>>ES
>>>>>              based on Paxos algorithm
>>>>>
>>>>>              Weiguo,
>>>>>
>>>>>              Your proposal introduces a control plane processing load
>>>>>              that is O(#EVIs * PEs) per DF election and given that
>>>>>              there can be 4K EVIs per ES, this looks like a
>>>>>              **substantial** load. Furthermore,
>>>>>
>>>>>              you can=B9t  use the ES route to co-ordinate DF election
>>>>>              because you would need to carry your new extended
>>>>>              community for each EVI and they would not all fit.  You
>>>>>              also can=B9t use the Per EVI Ethernet AD route because
>>>>>that
>>>>>              is processed by all PEs in the EVI.
>>>>>
>>>>>              I think that from a practical perspective the new DF
>>>>>              election proposed in Satya=B9s draft is sufficiently
>>>>>stable
>>>>>              that it renders your draft moot, even if it could be
>>>>>made
>>>>>              to work.
>>>>>
>>>>>              Yours Irrespectively,
>>>>>
>>>>>              John
>>>>>
>>>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf Of
>>>>>              *Haoweiguo
>>>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
>>>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>              <mailto:bess@ietf.org>
>>>>>              *Subject:* Re: [bess] Handshaking among PEs in an EVPN
>>>>>ES
>>>>>              based on Paxos algorithm
>>>>>
>>>>>              Hi Ali,
>>>>>
>>>>>              Thanks for your information. I scanned through this
>>>>>draft,
>>>>>              it really introduces inter-chassis message for DF
>>>>>election
>>>>>              handshaking, the requirements in this draft is to
>>>>>              eliminate transiet Loop and traffic duplication. Current
>>>>>              EVPN DF election mechanism already eliminated loop and
>>>>>              traffic duplication by configuring long reception timer
>>>>>on
>>>>>              each multi-homed PE, but up to reception timer traffic
>>>>>              disruption issue still exist. EVPN for DCI is an
>>>>>important
>>>>>              use case for EVPN, up to reception timer traffic
>>>>>              disruption can't be tolerated for service providers, it
>>>>>              should be improved.
>>>>>
>>>>>              Also for accuracy, i think handshaking state machine on
>>>>>              each multi-homed PE is also needed. From solution
>>>>>              perspective, in my draft, no inter-chassis message is
>>>>>              introduced, only one new extended community is
>>>>>introduced,
>>>>>              i think the process is comparatively simple than your
>>>>>              following draft.
>>>>>
>>>>>              Current EVPN DF election has some drawbacks, so there
>>>>>are
>>>>>              three new drafts about DF election emerged. I think BESS
>>>>>              WG can consider these three drafts in global view, a
>>>>>              single,comprehensive new DF election draft is hoped.
>>>>>
>>>>>              thanks.
>>>>>
>>>>>              Thanks,
>>>>>
>>>>>              weiguo
>>>>>
>>>>>
>>>>>----------------------------------------------------------------------
>>>>>-
>>>>>-
>>>>>
>>>>>              *From:*BESS [bess-bounces@ietf.org
>>>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajassi
>>>>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>]
>>>>>              *Sent:* Thursday, March 26, 2015 0:21
>>>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
>>>>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>>>>              based on Paxos algorithm
>>>>>
>>>>>              FYI- First published July 4, 2011
>>>>>
>>>>>
>>>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-rout
>>>>>e
>>>>>/
>>>>>
>>>>>              -Ali
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> BESS mailing list
>>>>> BESS@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>_______________________________________________________________________
>>>>_
>>>>_
>>>>________________________________________________
>>>>
>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>>confidentielles ou privilegiees et ne doivent donc
>>>> pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>>>>recu ce message par erreur, veuillez le signaler
>>>> a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>>>messages electroniques etant susceptibles d'alteration,
>>>> Orange decline toute responsabilite si ce message a ete altere,
>>>>deforme
>>>>ou falsifie. Merci.
>>>>
>>>> This message and its attachments may contain confidential or
>>>>privileged
>>>>information that may be protected by law;
>>>> they should not be distributed, used or copied without authorisation.
>>>> If you have received this email in error, please notify the sender and
>>>>delete this message and its attachments.
>>>> As emails may be altered, Orange is not liable for messages that have
>>>>been modified, changed or falsified.
>>>> Thank you.
>>>>
>>>> _______________________________________________
>>>> BESS mailing list
>>>> BESS@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>
>>>
>>>
>>>________________________________________________________________________
>>>_
>>>_
>>>_______________________________________________
>>>
>>>Ce message et ses pieces jointes peuvent contenir des informations
>>>confidentielles ou privilegiees et ne doivent donc
>>>pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>>>recu ce message par erreur, veuillez le signaler
>>>a l'expediteur et le detruire ainsi que les pieces jointes. Les messages
>>>electroniques etant susceptibles d'alteration,
>>>Orange decline toute responsabilite si ce message a ete altere, deforme
>>>ou falsifie. Merci.
>>>
>>>This message and its attachments may contain confidential or privileged
>>>information that may be protected by law;
>>>they should not be distributed, used or copied without authorisation.
>>>If you have received this email in error, please notify the sender and
>>>delete this message and its attachments.
>>>As emails may be altered, Orange is not liable for messages that have
>>>been modified, changed or falsified.
>>>Thank you.
>>>_______________________________________________
>>>BESS mailing list
>>>BESS@ietf.org
>>>https://www.ietf.org/mailman/listinfo/bess
>>_______________________________________________
>>BESS mailing list
>>BESS@ietf.org
>>https://www.ietf.org/mailman/listinfo/bess
>_______________________________________________
>BESS mailing list
>BESS@ietf.org
>https://www.ietf.org/mailman/listinfo/bess=


From nobody Fri Mar 27 08:22:14 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 33F671A00DF for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:22:13 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.611
X-Spam-Level: 
X-Spam-Status: No, score=-3.611 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id nrfZ19Maeci9 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:22:08 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 68C7D1A0231 for <bess@ietf.org>; Fri, 27 Mar 2015 08:22:07 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml404-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT94423; Fri, 27 Mar 2015 15:22:06 +0000 (GMT)
Received: from NKGEML406-HUB.china.huawei.com (10.98.56.37) by lhreml404-hub.china.huawei.com (10.201.5.218) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 15:22:04 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml406-hub.china.huawei.com ([10.98.56.37]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 23:21:58 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: John E Drake <jdrake@juniper.net>, "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "EXT - thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HYAABS9Z9AAD4thgAANfcBr
Date: Fri, 27 Mar 2015 15:21:57 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F8577A5@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com>, <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562A8E28D580CBCD24031C0C7090@BLUPR05MB562.namprd05.prod.outlook.com>
In-Reply-To: <BLUPR05MB562A8E28D580CBCD24031C0C7090@BLUPR05MB562.namprd05.prod.outlook.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.144.149]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/l7eq0BPuIOWf1jjQ0KTY5O4Qfbk>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 15:22:13 -0000

Hi John,
Thanks for your explanation.
Yes, there is no local loop relying on ESI label filtering in transiet peri=
od. But the traffic from remote PE to local mutli-homed network side, the l=
oop will be formed. I have shown that in the relies to Jorge.

Thanks,
weiguo

________________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 22:55
To: Haoweiguo; Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT - thom=
as.morin@orange.com; bess@ietf.org
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

There's no loop.  Any BUM packet received from either CE1 or CE2 will not b=
e forwarded back to them because of the ES label.

Yours Irrespectively,

John

> -----Original Message-----
> From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
> Sent: Friday, March 27, 2015 10:30 AM
> To: Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT -
> thomas.morin@orange.com; bess@ietf.org
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>
> Hi Jorge,
> Sorry, maybe i have not explain clearly, pls allow me to clarify it more =
clearly.
> In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2 is
> single homed to remote PE3. CE1 and CE2 forms one layer 2 network, so it
> multi-homed network scenario.
>            CE3
>              |
>             PE3
>
>
> PE1                  PE2
>   |                       |
> CE1-----------CE2
>
> In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends B=
UM
> traffic to CE1, the traffic will reach PE1 and PE2, CE1 will receive two =
copy of
> the traffic from CE3, also loop is formed, the form path:
>
> CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
>
> I think the traffic loop is absolutely not allowed. In MHN case, the loop=
 will
> formed due to dual-DF PE in transiet period.
>
> Thanks,
> weiguo
>
>
>
> ________________________________________
> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
> Sent: Friday, March 27, 2015 20:26
> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
> bess@ietf.org
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>
> Hi Weiguo,
>
> About this:
>
> [weiguo]: Traffic duplication and loop issue both exist in dual DF PE cas=
e. In
> multi-homed device case, say CE1 is multi-homed to PE1 and PE2 and both P=
E
> are DF PE. When CE1 sends traffic to PE1, the traffic will loop back to C=
E1
> through PE2. If CE1 can filter and drop the traffic , it is fortunate, ma=
ybe no
> loop.
> For multi-homed network scenario, say local network1(CE1 and CE2) is mult=
i-
> homed to PE1 and PE2. Similarly, the traffic from CE1 will loop back thro=
ugh
> PE1-->PE2-->CE2, it is very serious. It is absolutely not allowed.
>
>
> [JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2
> will not send the traffic back to CE1 due to the ESI-label and split-hori=
zon
> procedures.
> In MH network case PE2 will NOT send the traffic to CE2 for the same reas=
on.
>
> Let me know if you disagree please.
>
> Thanks.
> Jorge
>
> -----Original Message-----
> From: Haoweiguo <haoweiguo@huawei.com>
> Date: Thursday, March 26, 2015 at 9:55 PM
> To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
> "thomas.morin@orange.com" <thomas.morin@orange.com>,
> "bess@ietf.org"
> <bess@ietf.org>
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>
> >Hi Satya,
> >Pls see inline.
> >
> >Thanks,
> >weiguo
> >
> >________________________________________
> >From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
> >Sent: Friday, March 27, 2015 12:11
> >To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
> >Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> >algorithm
> >
> >Thomas,
> >
> >Thanks for your summary. It captures the essence of the discussion.
> >[weiguo]: Yes, i agree.
> >
> >One feature of the HRW scheme is that it computes the Active and the
> >Backup DF upfront.
> >Whenever the Active PE, say PE1, dies;
> >the Backup PE, PE2, can become the
> >Active immediately upon receiving the withdraw of the ES route from PE1.
> >It need not even have to wait for the 3 seconds time.
> >[weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
> >message:) But when PE1 leaves ESI, it can withdraw the ES route
> >immediately.
> >When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2
> >will both act as PE for some transiet time.
> >
> >On the other hand, the 3 second timer will serve as a bulk and batch
> >mechanism for the case when new PE(s) is(are) coming up, so we don=B9t
> >have to do a DF election ever so often.
> >
> >Haoweiguo,  I think the "loop" scenario is speculative.
> >I think everybody understands that, yes, there may be some transient
> >duplication of bum traffic; however, there is no "loop" problem to be
> >solved.
> >[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
> >case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
> >and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
> >loop back to CE1 through PE2. If CE1 can filter and drop the traffic ,
> >it is fortunate, maybe no loop.
> >For multi-homed network scenario, say local network1(CE1 and CE2) is
> >multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
> >back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
> >allowed.
> >
> >Thanks,
> >--Satya
> >
> >
> >
> >On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
> >
> >>Hi Thomas,
> >>I know you are neutral, sorry for my reply vulnerable to be miunderstoo=
d.
> >>
> >>Thanks,
> >>weiguo
> >>
> >>________________________________________
> >>From: thomas.morin@orange.com [thomas.morin@orange.com]
> >>Sent: Friday, March 27, 2015 11:59
> >>To: Haoweiguo; bess@ietf.org
> >>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> >>algorithm
> >>
> >>Hi Weiguo,
> >>
> >>2015-03-26, Haoweiguo:
> >> > Thomas:
> >>> > - RFC7432 may have transient periods where the DF election state
> >>> > is not yet synchronized between the two peers:
> >> >
> >>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic
> >>> loop
> >>
> >>Note well that I didn't write that there can be transient _loops_ .
> >>I tried to capture the exchange you had all, and what I gather is that
> >>the split-horizon procedure _prevents_loops_, independently of any
> >>transient period where DF state is not synchronized and which may lead
> >>to transient _duplicates_.
> >>
> >>-Thomas
> >>
> >>>
> >>> 26/03/2015 20:05, John E Drake :
> >>>>
> >>>> Weiguo,
> >>>>
> >>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I have
> >>>> detailed, is a non-solution to a non-problem with tremendous
> >>>> control plane cost.
> >>>>
> >>>> Yours Irrespectively,
> >>>>
> >>>> John
> >>>>
> >>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>> *Sent:* Thursday, March 26, 2015 6:17 PM
> >>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> >>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> Pls see below.
> >>>>
> >>>> Thanks,
> >>>>
> >>>> weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>> *From:*John E Drake [jdrake@juniper.net]
> >>>> *Sent:* Friday, March 27, 2015 6:00
> >>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> >>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> To recap,
> >>>>
> >>>> We have established that your proposal is untenable because of its
> >>>> control plane load.
> >>>>
> >>>> We have established that your proposal is based upon a flawed
> >>>> understanding of the DF election in RFC 7432.
> >>>>
> >>>> [weiguo]: In ethernet world, traffic loop is serious than short
> >>>> timer traffic disruption. If you want to implement  transiet
> >>>> traffic loop process, i will modify my draft to solve your issue.
> >>>>
> >>>> If i am the developer, i will prefer short timer traffic disruption
> >>>> based on current EVPN protocol.
> >>>>
> >>>> What you are now arguing is that your draft prevents two or more
> >>>> PEs from being DF simultaneously. This is clearly nonsense.
> >>>>
> >>>> [weiguo]: I will modify the draft problem statements, and use the
> >>>> same handshaking solution to solve it.
> >>>>
> >>>> Furthermore, we have established that having two or more DFs for
> >>>> what even you admit is a brief transient leads to duplicate
> >>>> traffic, which is acceptable, but not loops, your assertion to the
> contrary.
> >>>>
> >>>> [weiguo]: It is transient loop and traffic duplication issue.
> >>>>
> >>>> Yours Irrespectively,
> >>>>
> >>>> John
> >>>>
> >>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>> *Sent:* Thursday, March 26, 2015 5:37 PM
> >>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> John,
> >>>>
> >>>> As your understanding of the EVPN draft, the DF election mechanism
> >>>> has more serious side effect, it will have short time traffic
> >>>> loop,i.e., dual DF PEs will exist for a short time. I think dual DF
> >>>> PEs is absolutely not tolerated, because native ethernet header has
> >>>> no TTL, up to several hundred ms traffic loop normally not
> >>>> tolerated in commertial networks.
> >>>>
> >>>> As your understanding, the PEs should do as following:
> >>>>
> >>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but
> >>>> have rarely deployment.
> >>>>
> >>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3
> >>>> joins ESI and trigger DF re-election. When reception timer expires:
> >>>>
> >>>> PE1 upgrades to DF PE.
> >>>>
> >>>> After reception timer+ ES route transmission timer:
> >>>>
> >>>> PE2 downloads to non-DF PE.
> >>>>
> >>>> So in timer clock sync case, dual DF PEs will exist at least
> >>>> transmission timer.
> >>>>
> >>>> If NTP is used for timer sync, because it has bad accuracy, dual DF
> >>>> PEs will exist more longer timer.
> >>>>
> >>>> So as your understanding for DF election, the drawback is more clear=
.
> >>>>
> >>>> Thanks,
> >>>>
> >>>> weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>> *From:*John E Drake [jdrake@juniper.net]
> >>>> *Sent:* Friday, March 27, 2015 4:41
> >>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> >>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> Weiguo,
> >>>>
> >>>> We have already established that your proposal is untenable because
> >>>> of its control plane load.
> >>>>
> >>>> What we are now discussing is that your proposal is based upon a
> >>>> misunderstanding of the algorithm in RFC 7432.  You are assuming
> >>>> that
> >>>> PE1 will advertise an ES route and then wait for the configured
> >>>> interval before performing the DF election while PE2 and PE3 will
> >>>> perform the DF election as soon as they receive the ES route from PE=
1.
> >>>> This is not what RFC 7432 says.
> >>>>
> >>>> Rather, what is says is that the advertisement of the ES route by
> >>>> PE1 and its receipt by PE2 and PE3 causes all three PEs to start
> >>>> the configured interval timer  - =B33. When the timer expires, each
> >>>> PE builds an ordered list of the IP addresses of all the PE nodes
> >>>> connected to the Ethernet segment (including itself), in increasing
> >>>> numeric value.=B2
> >>>>
> >>>> Yours Irrespectively,
> >>>>
> >>>> John
> >>>>
> >>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>> *Sent:* Thursday, March 26, 2015 3:26 PM
> >>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> Pls read my detail replies to Satya. If you still can't catch it,
> >>>> pls read my draft and EVPN base protocol,  thanks
> >>>>
> >>>> weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>> *From:*John E Drake [jdrake@juniper.net]
> >>>> *Sent:* Friday, March 27, 2015 1:28
> >>>> *To:* Patrice Brissette (pbrisset)
> >>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
> >>>> <mailto:bess@ietf.org>
> >>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
> >>>> Paxos algorithm
> >>>>
> >>>> I think Patrice is correct.  Your proposal doesn't solve the
> >>>> problem and it does so at huge cost.
> >>>>
> >>>> Sent from my iPhone
> >>>>
> >>>>
> >>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
> >>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
> >>>>
> >>>>      Weiguo,
> >>>>
> >>>>      I=B9m not sure I=B9m following here.
> >>>>
> >>>>      Don=B9t you have the same issue with your handshaking mechanism=
?
> >>>>
> >>>>      If you don=B9t know your peer, how can you handshake?
> >>>>
> >>>>      Regards,
> >>>>
> >>>>      Patrice
> >>>>
> >>>>      Image removed by sender.
> >>>>
> >>>>      *Patrice Brissette*
> >>>>      TECHNICAL LEADER.ENGINEERING
> >>>>
> >>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
> >>>>      Phone: *+1 613 254 3336*
> >>>>
> >>>>
> >>>>
> >>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
> >>>>      Canada
> >>>>      Cisco.com <http://www.cisco.com/global/CA/>
> >>>>
> >>>>
> >>>>
> >>>>      Image removed by sender.Think before you print.
> >>>>
> >>>>      This email may contain confidential and privileged material for
> >>>>      the sole use of the intended recipient. Any review, use,
> >>>>      distribution or disclosure by others is strictly prohibited. If
> >>>>      you are not the intended recipient (or authorized to receive fo=
r
> >>>>      the recipient), please contact the sender by reply email and
> >>>>      delete all copies of this message.
> >>>>
> >>>>      Please click here
> >>>>
> >>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.ht
> ml>
> >>>>      for Company Registration Information.
> >>>>
> >>>>
> >>>>      *From: *Haoweiguo <haoweiguo@huawei.com
> >>>><mailto:haoweiguo@huawei.com>>
> >>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
> >>>>      *To: *Patrice Brissette <pbrisset@cisco.com
> >>>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
> >>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
> >>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
> >>>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
> >>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES
> >>>>based on
> >>>>      Paxos algorithm
> >>>>
> >>>>          Hi Patrice,
> >>>>
> >>>>          Up to reception timer traffic disruption in transient
> >>>>phase is
> >>>>          one of the issues.
> >>>>
> >>>>          Thanks,
> >>>>
> >>>>          weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
> >>>>          <mailto:pbrisset@cisco.com>]
> >>>>          *Sent:* Thursday, March 26, 2015 20:54
> >>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
> >>>>          bess@ietf.org <mailto:bess@ietf.org>
> >>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
> >>>>          based on Paxos algorithm
> >>>>
> >>>>          Weiguo,
> >>>>
> >>>>          You mention "But if your draft have not solved all
> >>>> issues=B2,
> >>>>
> >>>>          Can you explain what Satya=B9s draft is not solving?
> >>>>
> >>>>          Regards,
> >>>>
> >>>>          Patrice
> >>>>
> >>>>          Image removed by sender.
> >>>>
> >>>>          *Patrice Brissette*
> >>>>          TECHNICAL LEADER.ENGINEERING
> >>>>
> >>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
> >>>>          Phone: *+1 613 254 3336*
> >>>>
> >>>>
> >>>>
> >>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
> >>>>          Canada
> >>>>          Cisco.com <http://www.cisco.com/global/CA/>
> >>>>
> >>>>
> >>>>
> >>>>          Image removed by sender.Think before you print.
> >>>>
> >>>>          This email may contain confidential and privileged material
> >>>>          for the sole use of the intended recipient. Any review, use=
,
> >>>>          distribution or disclosure by others is strictly prohibited=
.
> >>>>          If you are not the intended recipient (or authorized to
> >>>>          receive for the recipient), please contact the sender by
> >>>>reply
> >>>>          email and delete all copies of this message.
> >>>>
> >>>>          Please click here
> >>>>
> >>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.ht
> ml>
> >>>>          for Company Registration Information.
> >>>>
> >>>>
> >>>>          *From: *Haoweiguo <haoweiguo@huawei.com
> >>>>          <mailto:haoweiguo@huawei.com>>
> >>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
> >>>>          *To: *John E Drake <jdrake@juniper.net
> >>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.co=
m
> >>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
> >>>>          <mailto:bess@ietf.org>" <bess@ietf.org
> >>>><mailto:bess@ietf.org>>
> >>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
> >>>>          based on Paxos algorithm
> >>>>
> >>>>              Hi John,
> >>>>
> >>>>               Firstly i think EVPN community should reach consensus =
on
> >>>>              the issues of current DF election mechanism. All these
> >>>>              issues should be resolved in a single new DF election
> >>>>              draft,rather than in multiple separate drafts. If your
> >>>>              draft can solve all these issues and stable, i have no
> >>>>              question for its progressing. But if your draft have no=
t
> >>>>              solved all issues, i think it had better combine with
> >>>>              other drafts to provide a comprehensive solution. I thi=
nk
> >>>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01 a=
nd
> >>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it shou=
ld
> >>>>              be resolved. So i think although your new Hash algorith=
m
> >>>>              for DF election is good, it only includes partial
> >>>>              enhancements, maybe it still needs some time for
> >>>>consensus.
> >>>>
> >>>>              Thanks,
> >>>>
> >>>>              weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>>              *From:*John E Drake [jdrake@juniper.net
> >>>>              <mailto:jdrake@juniper.net>]
> >>>>              *Sent:* Thursday, March 26, 2015 7:16
> >>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
> >>>>              <mailto:bess@ietf.org>
> >>>>              *Subject:* RE: [bess] Handshaking among PEs in an EVPN =
ES
> >>>>              based on Paxos algorithm
> >>>>
> >>>>              Weiguo,
> >>>>
> >>>>              Your proposal introduces a control plane processing loa=
d
> >>>>              that is O(#EVIs * PEs) per DF election and given that
> >>>>              there can be 4K EVIs per ES, this looks like a
> >>>>              **substantial** load. Furthermore,
> >>>>
> >>>>              you can=B9t  use the ES route to co-ordinate DF electio=
n
> >>>>              because you would need to carry your new extended
> >>>>              community for each EVI and they would not all fit.  You
> >>>>              also can=B9t use the Per EVI Ethernet AD route because =
that
> >>>>              is processed by all PEs in the EVI.
> >>>>
> >>>>              I think that from a practical perspective the new DF
> >>>>              election proposed in Satya=B9s draft is sufficiently st=
able
> >>>>              that it renders your draft moot, even if it could be ma=
de
> >>>>              to work.
> >>>>
> >>>>              Yours Irrespectively,
> >>>>
> >>>>              John
> >>>>
> >>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf O=
f
> >>>>              *Haoweiguo
> >>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
> >>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
> >>>>              <mailto:bess@ietf.org>
> >>>>              *Subject:* Re: [bess] Handshaking among PEs in an EVPN =
ES
> >>>>              based on Paxos algorithm
> >>>>
> >>>>              Hi Ali,
> >>>>
> >>>>              Thanks for your information. I scanned through this
> >>>>draft,
> >>>>              it really introduces inter-chassis message for DF
> >>>>election
> >>>>              handshaking, the requirements in this draft is to
> >>>>              eliminate transiet Loop and traffic duplication. Curren=
t
> >>>>              EVPN DF election mechanism already eliminated loop and
> >>>>              traffic duplication by configuring long reception
> >>>>timer on
> >>>>              each multi-homed PE, but up to reception timer traffic
> >>>>              disruption issue still exist. EVPN for DCI is an
> >>>>important
> >>>>              use case for EVPN, up to reception timer traffic
> >>>>              disruption can't be tolerated for service providers, it
> >>>>              should be improved.
> >>>>
> >>>>              Also for accuracy, i think handshaking state machine on
> >>>>              each multi-homed PE is also needed. From solution
> >>>>              perspective, in my draft, no inter-chassis message is
> >>>>              introduced, only one new extended community is
> >>>>introduced,
> >>>>              i think the process is comparatively simple than your
> >>>>              following draft.
> >>>>
> >>>>              Current EVPN DF election has some drawbacks, so there a=
re
> >>>>              three new drafts about DF election emerged. I think BES=
S
> >>>>              WG can consider these three drafts in global view, a
> >>>>              single,comprehensive new DF election draft is hoped.
> >>>>
> >>>>              thanks.
> >>>>
> >>>>              Thanks,
> >>>>
> >>>>              weiguo
> >>>>
> >>>>
> >>>>--------------------------------------------------------------------
> >>>>---
> >>>>-
> >>>>
> >>>>              *From:*BESS [bess-bounces@ietf.org
> >>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajass=
i
> >>>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>=
]
> >>>>              *Sent:* Thursday, March 26, 2015 0:21
> >>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
> >>>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
> >>>>              based on Paxos algorithm
> >>>>
> >>>>              FYI- First published July 4, 2011
> >>>>
> >>>>
> >>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-ro
> >>>>ute
> >>>>/
> >>>>
> >>>>              -Ali
> >>>>
> >>>>
> >>>>
> >>>> _______________________________________________
> >>>> BESS mailing list
> >>>> BESS@ietf.org
> >>>> https://www.ietf.org/mailman/listinfo/bess
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>_______________________________________________________
> ______________
> >>>___
> >>>_
> >>>________________________________________________
> >>>
> >>> Ce message et ses pieces jointes peuvent contenir des informations
> >>>confidentielles ou privilegiees et ne doivent donc  pas etre
> >>>diffuses, exploites ou copies sans autorisation. Si vous avez recu ce
> >>>message par erreur, veuillez le signaler  a l'expediteur et le
> >>>detruire ainsi que les pieces jointes. Les messages electroniques
> >>>etant susceptibles d'alteration,  Orange decline toute responsabilite
> >>>si ce message a ete altere, deforme ou falsifie. Merci.
> >>>
> >>> This message and its attachments may contain confidential or
> >>>privileged information that may be protected by law;  they should not
> >>>be distributed, used or copied without authorisation.
> >>> If you have received this email in error, please notify the sender
> >>>and delete this message and its attachments.
> >>> As emails may be altered, Orange is not liable for messages that
> >>>have been modified, changed or falsified.
> >>> Thank you.
> >>>
> >>> _______________________________________________
> >>> BESS mailing list
> >>> BESS@ietf.org
> >>> https://www.ietf.org/mailman/listinfo/bess
> >>>
> >>
> >>
> >>________________________________________________________
> ______________
> >>___
> >>_
> >>_______________________________________________
> >>
> >>Ce message et ses pieces jointes peuvent contenir des informations
> >>confidentielles ou privilegiees et ne doivent donc pas etre diffuses,
> >>exploites ou copies sans autorisation. Si vous avez recu ce message
> >>par erreur, veuillez le signaler a l'expediteur et le detruire ainsi
> >>que les pieces jointes. Les messages electroniques etant susceptibles
> >>d'alteration, Orange decline toute responsabilite si ce message a ete
> >>altere, deforme ou falsifie. Merci.
> >>
> >>This message and its attachments may contain confidential or
> >>privileged information that may be protected by law; they should not
> >>be distributed, used or copied without authorisation.
> >>If you have received this email in error, please notify the sender and
> >>delete this message and its attachments.
> >>As emails may be altered, Orange is not liable for messages that have
> >>been modified, changed or falsified.
> >>Thank you.
> >>_______________________________________________
> >>BESS mailing list
> >>BESS@ietf.org
> >>https://www.ietf.org/mailman/listinfo/bess
> >_______________________________________________
> >BESS mailing list
> >BESS@ietf.org
> >https://www.ietf.org/mailman/listinfo/bess
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess=


From nobody Fri Mar 27 08:23:02 2015
Return-Path: <jdrake@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E1CD51A8A73 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:22:58 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.302
X-Spam-Level: 
X-Spam-Status: No, score=-1.302 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iSHChNUV-lvN for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:22:49 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1on0797.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::797]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8CBDA1ACEE7 for <bess@ietf.org>; Fri, 27 Mar 2015 08:22:37 -0700 (PDT)
Received: from BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) by BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) with Microsoft SMTP Server (TLS) id 15.1.118.21; Fri, 27 Mar 2015 15:22:20 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) by BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) with mapi id 15.01.0118.022; Fri, 27 Mar 2015 15:22:20 +0000
From: John E Drake <jdrake@juniper.net>
To: Haoweiguo <haoweiguo@huawei.com>, "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "EXT - thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8P/o146fJTTukSNC2GolBinQZ0u4lsAgAACyoCAAB/gjYAAILMAgAAMEtCAABidAIAAA48AgAAHqACAAC468IAAHjiAgAAP+ACAAAM7AIAAApwAgAAA7YCAAAw6gIAAfgsAgAAiUICAAAhggIAAA1IAgAACs5A=
Date: Fri, 27 Mar 2015 15:22:20 +0000
Message-ID: <BLUPR05MB562504AAF732AF476E5C218C7090@BLUPR05MB562.namprd05.prod.outlook.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>, <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [66.129.239.10]
authentication-results: huawei.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BLUPR05MB562;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(377424004)(24454002)(51704005)(479174004)(36304003)(69234005)(13464003)(164054003)(377454003)(52604005)(50986999)(122556002)(86362001)(2900100001)(5890100001)(561944003)(33656002)(2501003)(54356999)(76176999)(46102003)(2950100001)(93886004)(107886001)(102836002)(15975445007)(77096005)(77156002)(62966003)(2656002)(74316001)(66066001)(92566002)(106116001)(19580395003)(19580405001)(76576001)(87936001)(99286002)(579004)(559001); DIR:OUT; SFP:1102; SCL:1; SRVR:BLUPR05MB562; H:BLUPR05MB562.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BLUPR05MB562D266378DE82CD9F5349EC7090@BLUPR05MB562.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BLUPR05MB562; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB562; 
x-forefront-prvs: 0528942FD8
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: juniper.net
X-MS-Exchange-CrossTenant-originalarrivaltime: 27 Mar 2015 15:22:20.4682 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BLUPR05MB562
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/rAVPriSZ95-Lyz00veRRz44PwLk>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 15:22:59 -0000

I think we have beaten this topic to death.  For the numerous reason I've g=
iven, I'm not interested in you proposal.

Yours Irrespectively,

John

> -----Original Message-----
> From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
> Sent: Friday, March 27, 2015 11:12 AM
> To: Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT -
> thomas.morin@orange.com; bess@ietf.org
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>=20
> Hi Jorge,
> Let give you another indefinitely forwarding loop case, the picture is as
> follows:
>=20
> CE3------------CE4
>   |                        |
> PE3                   PE4
>=20
>=20
>=20
> PE1                   PE2
>  |                         |
> CE1------------CE2
>=20
> If PE1 and PE2 are dual DF PE in transiet time, assuing PE3 and PE4 are i=
n
> stable state ,PE3 is DF PE, PE4 is non-DF PE, the BUM traffic from CE3 to=
 CE1,
> the forwarding loop path is as follows;
> CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->PE4---=
-
> >forever......
>=20
> Do you think the harm is enough?
> Thanks,
> weiguo
> ________________________________________
> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
> Sent: Friday, March 27, 2015 22:59
> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
> bess@ietf.org
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>=20
> Hi Weiguo,
>=20
> OK, thanks for clarifying, that is the part I did not understand in your =
loop :-)
>=20
> To me a loop means you send a frame and that frame keeps being forwarded
> by the same nodes indefinitely till you take a corrective action.
> In this case, CE3 might get a few packets during the transient state, but=
 those
> packets will have its own MAC SA so CE3 will discard them. I don't think =
that is
> harmful for CE3 in most of the cases. And if it is, you just need to make=
 sure
> the timers are long enough and the DF election properly implemented to
> avoid the transient DF-DF state.
>=20
> The handshaking is definitely overkilling IMHO.
>=20
> Thanks.
> Jorge
>=20
> -----Original Message-----
> From: Haoweiguo <haoweiguo@huawei.com>
> Date: Friday, March 27, 2015 at 7:29 AM
> To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
> (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
> <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>=20
> >Hi Jorge,
> >Sorry, maybe i have not explain clearly, pls allow me to clarify it
> >more clearly.
> >In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2
> >is single homed to remote PE3. CE1 and CE2 forms one layer 2 network,
> >so it multi-homed network scenario.
> >           CE3
> >             |
> >            PE3
> >
> >
> >PE1                  PE2
> >  |                       |
> >CE1-----------CE2
> >
> >In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends
> >BUM traffic to CE1, the traffic will reach PE1 and PE2, CE1 will
> >receive two copy of the traffic from CE3, also loop is formed, the form =
path:
> >
> >CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
> >
> >I think the traffic loop is absolutely not allowed. In MHN case, the
> >loop will formed due to dual-DF PE in transiet period.
> >
> >Thanks,
> >weiguo
> >
> >
> >
> >________________________________________
> >From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
> >Sent: Friday, March 27, 2015 20:26
> >To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
> >bess@ietf.org
> >Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> >algorithm
> >
> >Hi Weiguo,
> >
> >About this:
> >
> >[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
> >case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
> >and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
> >loop back to CE1 through PE2. If CE1 can filter and drop the traffic ,
> >it is fortunate, maybe no loop.
> >For multi-homed network scenario, say local network1(CE1 and CE2) is
> >multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
> >back through PE1-->PE2-->CE2, it is very serious. It is absolutely not a=
llowed.
> >
> >
> >[JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2
> >will not send the traffic back to CE1 due to the ESI-label and
> >split-horizon procedures.
> >In MH network case PE2 will NOT send the traffic to CE2 for the same
> >reason.
> >
> >Let me know if you disagree please.
> >
> >Thanks.
> >Jorge
> >
> >-----Original Message-----
> >From: Haoweiguo <haoweiguo@huawei.com>
> >Date: Thursday, March 26, 2015 at 9:55 PM
> >To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
> >"thomas.morin@orange.com" <thomas.morin@orange.com>,
> "bess@ietf.org"
> ><bess@ietf.org>
> >Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> >algorithm
> >
> >>Hi Satya,
> >>Pls see inline.
> >>
> >>Thanks,
> >>weiguo
> >>
> >>________________________________________
> >>From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
> >>Sent: Friday, March 27, 2015 12:11
> >>To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
> >>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> >>algorithm
> >>
> >>Thomas,
> >>
> >>Thanks for your summary. It captures the essence of the discussion.
> >>[weiguo]: Yes, i agree.
> >>
> >>One feature of the HRW scheme is that it computes the Active and the
> >>Backup DF upfront.
> >>Whenever the Active PE, say PE1, dies; the Backup PE, PE2, can become
> >>the Active immediately upon receiving the withdraw of the ES route
> >>from PE1.
> >>It need not even have to wait for the 3 seconds time.
> >>[weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
> >>message:) But when PE1 leaves ESI, it can withdraw the ES route
> >>immediately.
> >>When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2
> >>will both act as PE for some transiet time.
> >>
> >>On the other hand, the 3 second timer will serve as a bulk and batch
> >>mechanism for the case when new PE(s) is(are) coming up, so we don=B9t
> >>have to do a DF election ever so often.
> >>
> >>Haoweiguo,  I think the "loop" scenario is speculative.
> >>I think everybody understands that, yes, there may be some transient
> >>duplication of bum traffic; however, there is no "loop" problem to be
> >>solved.
> >>[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
> >>case. In multi-homed device case, say CE1 is multi-homed to PE1 and
> >>PE2 and both PE are DF PE. When CE1 sends traffic to PE1, the traffic
> >>will loop back to CE1 through PE2. If CE1 can filter and drop the
> >>traffic , it is fortunate, maybe no loop.
> >>For multi-homed network scenario, say local network1(CE1 and CE2) is
> >>multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
> >>back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
> >>allowed.
> >>
> >>Thanks,
> >>--Satya
> >>
> >>
> >>
> >>On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
> >>
> >>>Hi Thomas,
> >>>I know you are neutral, sorry for my reply vulnerable to be
> >>>miunderstood.
> >>>
> >>>Thanks,
> >>>weiguo
> >>>
> >>>________________________________________
> >>>From: thomas.morin@orange.com [thomas.morin@orange.com]
> >>>Sent: Friday, March 27, 2015 11:59
> >>>To: Haoweiguo; bess@ietf.org
> >>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on
> >>>Paxos algorithm
> >>>
> >>>Hi Weiguo,
> >>>
> >>>2015-03-26, Haoweiguo:
> >>> > Thomas:
> >>>> > - RFC7432 may have transient periods where the DF election state
> >>>> > is not yet synchronized between the two peers:
> >>> >
> >>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic
> >>>> loop
> >>>
> >>>Note well that I didn't write that there can be transient _loops_ .
> >>>I tried to capture the exchange you had all, and what I gather is
> >>>that the split-horizon procedure _prevents_loops_, independently of
> >>>any transient period where DF state is not synchronized and which may
> >>>lead to transient _duplicates_.
> >>>
> >>>-Thomas
> >>>
> >>>>
> >>>> 26/03/2015 20:05, John E Drake :
> >>>>>
> >>>>> Weiguo,
> >>>>>
> >>>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I hav=
e
> >>>>> detailed, is a non-solution to a non-problem with tremendous
> >>>>> control plane cost.
> >>>>>
> >>>>> Yours Irrespectively,
> >>>>>
> >>>>> John
> >>>>>
> >>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
> >>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> Pls see below.
> >>>>>
> >>>>> Thanks,
> >>>>>
> >>>>> weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>> *From:*John E Drake [jdrake@juniper.net]
> >>>>> *Sent:* Friday, March 27, 2015 6:00
> >>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> To recap,
> >>>>>
> >>>>> We have established that your proposal is untenable because of its
> >>>>> control plane load.
> >>>>>
> >>>>> We have established that your proposal is based upon a flawed
> >>>>> understanding of the DF election in RFC 7432.
> >>>>>
> >>>>> [weiguo]: In ethernet world, traffic loop is serious than short
> >>>>> timer traffic disruption. If you want to implement  transiet
> >>>>> traffic loop process, i will modify my draft to solve your issue.
> >>>>>
> >>>>> If i am the developer, i will prefer short timer traffic
> >>>>> disruption based on current EVPN protocol.
> >>>>>
> >>>>> What you are now arguing is that your draft prevents two or more
> >>>>> PEs from being DF simultaneously. This is clearly nonsense.
> >>>>>
> >>>>> [weiguo]: I will modify the draft problem statements, and use the
> >>>>>same  handshaking solution to solve it.
> >>>>>
> >>>>> Furthermore, we have established that having two or more DFs for
> >>>>> what even you admit is a brief transient leads to duplicate
> >>>>> traffic, which is acceptable, but not loops, your assertion to the
> contrary.
> >>>>>
> >>>>> [weiguo]: It is transient loop and traffic duplication issue.
> >>>>>
> >>>>> Yours Irrespectively,
> >>>>>
> >>>>> John
> >>>>>
> >>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
> >>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> John,
> >>>>>
> >>>>> As your understanding of the EVPN draft, the DF election mechanism
> >>>>>has  more serious side effect, it will have short time traffic
> >>>>>loop,i.e.,  dual DF PEs will exist for a short time. I think dual
> >>>>>DF PEs is  absolutely not tolerated, because native ethernet header
> >>>>>has no TTL,  up to several hundred ms traffic loop normally not
> >>>>>tolerated in  commertial networks.
> >>>>>
> >>>>> As your understanding, the PEs should do as following:
> >>>>>
> >>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but
> >>>>> have rarely deployment.
> >>>>>
> >>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3
> >>>>> joins ESI and trigger DF re-election. When reception timer expires:
> >>>>>
> >>>>> PE1 upgrades to DF PE.
> >>>>>
> >>>>> After reception timer+ ES route transmission timer:
> >>>>>
> >>>>> PE2 downloads to non-DF PE.
> >>>>>
> >>>>> So in timer clock sync case, dual DF PEs will exist at least
> >>>>> transmission timer.
> >>>>>
> >>>>> If NTP is used for timer sync, because it has bad accuracy, dual
> >>>>> DF PEs will exist more longer timer.
> >>>>>
> >>>>> So as your understanding for DF election, the drawback is more clea=
r.
> >>>>>
> >>>>> Thanks,
> >>>>>
> >>>>> weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>> *From:*John E Drake [jdrake@juniper.net]
> >>>>> *Sent:* Friday, March 27, 2015 4:41
> >>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> Weiguo,
> >>>>>
> >>>>> We have already established that your proposal is untenable
> >>>>>because of  its control plane load.
> >>>>>
> >>>>> What we are now discussing is that your proposal is based upon a
> >>>>>misunderstanding of the algorithm in RFC 7432.  You are assuming
> >>>>>that
> >>>>> PE1 will advertise an ES route and then wait for the configured
> >>>>>interval before performing the DF election while PE2 and PE3 will
> >>>>>perform the DF election as soon as they receive the ES route from
> >>>>>PE1.
> >>>>> This is not what RFC 7432 says.
> >>>>>
> >>>>> Rather, what is says is that the advertisement of the ES route by
> >>>>> PE1 and its receipt by PE2 and PE3 causes all three PEs to start
> >>>>> the configured interval timer  - =B33. When the timer expires, each
> >>>>> PE builds an ordered list of the IP addresses of all the PE nodes
> >>>>> connected to the Ethernet segment (including itself), in
> >>>>> increasing numeric value.=B2
> >>>>>
> >>>>> Yours Irrespectively,
> >>>>>
> >>>>> John
> >>>>>
> >>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
> >>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> Pls read my detail replies to Satya. If you still can't catch it,
> >>>>> pls read my draft and EVPN base protocol,  thanks
> >>>>>
> >>>>> weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>> *From:*John E Drake [jdrake@juniper.net]
> >>>>> *Sent:* Friday, March 27, 2015 1:28
> >>>>> *To:* Patrice Brissette (pbrisset)
> >>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
> >>>>> <mailto:bess@ietf.org>
> >>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> I think Patrice is correct.  Your proposal doesn't solve the
> >>>>> problem and it does so at huge cost.
> >>>>>
> >>>>> Sent from my iPhone
> >>>>>
> >>>>>
> >>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
> >>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
> >>>>>
> >>>>>      Weiguo,
> >>>>>
> >>>>>      I=B9m not sure I=B9m following here.
> >>>>>
> >>>>>      Don=B9t you have the same issue with your handshaking mechanis=
m?
> >>>>>
> >>>>>      If you don=B9t know your peer, how can you handshake?
> >>>>>
> >>>>>      Regards,
> >>>>>
> >>>>>      Patrice
> >>>>>
> >>>>>      Image removed by sender.
> >>>>>
> >>>>>      *Patrice Brissette*
> >>>>>      TECHNICAL LEADER.ENGINEERING
> >>>>>
> >>>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
> >>>>>      Phone: *+1 613 254 3336*
> >>>>>
> >>>>>
> >>>>>
> >>>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
> >>>>>      Canada
> >>>>>      Cisco.com <http://www.cisco.com/global/CA/>
> >>>>>
> >>>>>
> >>>>>
> >>>>>      Image removed by sender.Think before you print.
> >>>>>
> >>>>>      This email may contain confidential and privileged material fo=
r
> >>>>>      the sole use of the intended recipient. Any review, use,
> >>>>>      distribution or disclosure by others is strictly prohibited. I=
f
> >>>>>      you are not the intended recipient (or authorized to receive f=
or
> >>>>>      the recipient), please contact the sender by reply email and
> >>>>>      delete all copies of this message.
> >>>>>
> >>>>>      Please click here
> >>>>>
> >>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.h
> tml>
> >>>>>      for Company Registration Information.
> >>>>>
> >>>>>
> >>>>>      *From: *Haoweiguo <haoweiguo@huawei.com
> >>>>><mailto:haoweiguo@huawei.com>>
> >>>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
> >>>>>      *To: *Patrice Brissette <pbrisset@cisco.com
> >>>>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
> >>>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
> >>>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
> >>>>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
> >>>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES
> >>>>>based on
> >>>>>      Paxos algorithm
> >>>>>
> >>>>>          Hi Patrice,
> >>>>>
> >>>>>          Up to reception timer traffic disruption in transient
> >>>>>phase is
> >>>>>          one of the issues.
> >>>>>
> >>>>>          Thanks,
> >>>>>
> >>>>>          weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
> >>>>>          <mailto:pbrisset@cisco.com>]
> >>>>>          *Sent:* Thursday, March 26, 2015 20:54
> >>>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
> >>>>>          bess@ietf.org <mailto:bess@ietf.org>
> >>>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
> >>>>>          based on Paxos algorithm
> >>>>>
> >>>>>          Weiguo,
> >>>>>
> >>>>>          You mention "But if your draft have not solved all
> >>>>> issues=B2,
> >>>>>
> >>>>>          Can you explain what Satya=B9s draft is not solving?
> >>>>>
> >>>>>          Regards,
> >>>>>
> >>>>>          Patrice
> >>>>>
> >>>>>          Image removed by sender.
> >>>>>
> >>>>>          *Patrice Brissette*
> >>>>>          TECHNICAL LEADER.ENGINEERING
> >>>>>
> >>>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
> >>>>>          Phone: *+1 613 254 3336*
> >>>>>
> >>>>>
> >>>>>
> >>>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
> >>>>>          Canada
> >>>>>          Cisco.com <http://www.cisco.com/global/CA/>
> >>>>>
> >>>>>
> >>>>>
> >>>>>          Image removed by sender.Think before you print.
> >>>>>
> >>>>>          This email may contain confidential and privileged materia=
l
> >>>>>          for the sole use of the intended recipient. Any review, us=
e,
> >>>>>          distribution or disclosure by others is strictly prohibite=
d.
> >>>>>          If you are not the intended recipient (or authorized to
> >>>>>          receive for the recipient), please contact the sender by
> >>>>>reply
> >>>>>          email and delete all copies of this message.
> >>>>>
> >>>>>          Please click here
> >>>>>
> >>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.h
> tml>
> >>>>>          for Company Registration Information.
> >>>>>
> >>>>>
> >>>>>          *From: *Haoweiguo <haoweiguo@huawei.com
> >>>>>          <mailto:haoweiguo@huawei.com>>
> >>>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
> >>>>>          *To: *John E Drake <jdrake@juniper.net
> >>>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.c=
om
> >>>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
> >>>>>          <mailto:bess@ietf.org>" <bess@ietf.org
> >>>>><mailto:bess@ietf.org>>
> >>>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
> >>>>>          based on Paxos algorithm
> >>>>>
> >>>>>              Hi John,
> >>>>>
> >>>>>               Firstly i think EVPN community should reach
> >>>>>consensus on
> >>>>>              the issues of current DF election mechanism. All these
> >>>>>              issues should be resolved in a single new DF election
> >>>>>              draft,rather than in multiple separate drafts. If your
> >>>>>              draft can solve all these issues and stable, i have no
> >>>>>              question for its progressing. But if your draft have n=
ot
> >>>>>              solved all issues, i think it had better combine with
> >>>>>              other drafts to provide a comprehensive solution. I
> >>>>>think
> >>>>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01
> >>>>>and
> >>>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it
> >>>>>should
> >>>>>              be resolved. So i think although your new Hash algorit=
hm
> >>>>>              for DF election is good, it only includes partial
> >>>>>              enhancements, maybe it still needs some time for
> >>>>>consensus.
> >>>>>
> >>>>>              Thanks,
> >>>>>
> >>>>>              weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>>              *From:*John E Drake [jdrake@juniper.net
> >>>>>              <mailto:jdrake@juniper.net>]
> >>>>>              *Sent:* Thursday, March 26, 2015 7:16
> >>>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
> >>>>>              <mailto:bess@ietf.org>
> >>>>>              *Subject:* RE: [bess] Handshaking among PEs in an
> >>>>>EVPN ES
> >>>>>              based on Paxos algorithm
> >>>>>
> >>>>>              Weiguo,
> >>>>>
> >>>>>              Your proposal introduces a control plane processing lo=
ad
> >>>>>              that is O(#EVIs * PEs) per DF election and given that
> >>>>>              there can be 4K EVIs per ES, this looks like a
> >>>>>              **substantial** load. Furthermore,
> >>>>>
> >>>>>              you can=B9t  use the ES route to co-ordinate DF electi=
on
> >>>>>              because you would need to carry your new extended
> >>>>>              community for each EVI and they would not all fit.  Yo=
u
> >>>>>              also can=B9t use the Per EVI Ethernet AD route because
> >>>>>that
> >>>>>              is processed by all PEs in the EVI.
> >>>>>
> >>>>>              I think that from a practical perspective the new DF
> >>>>>              election proposed in Satya=B9s draft is sufficiently
> >>>>>stable
> >>>>>              that it renders your draft moot, even if it could be
> >>>>>made
> >>>>>              to work.
> >>>>>
> >>>>>              Yours Irrespectively,
> >>>>>
> >>>>>              John
> >>>>>
> >>>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf =
Of
> >>>>>              *Haoweiguo
> >>>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
> >>>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
> >>>>>              <mailto:bess@ietf.org>
> >>>>>              *Subject:* Re: [bess] Handshaking among PEs in an
> >>>>>EVPN ES
> >>>>>              based on Paxos algorithm
> >>>>>
> >>>>>              Hi Ali,
> >>>>>
> >>>>>              Thanks for your information. I scanned through this
> >>>>>draft,
> >>>>>              it really introduces inter-chassis message for DF
> >>>>>election
> >>>>>              handshaking, the requirements in this draft is to
> >>>>>              eliminate transiet Loop and traffic duplication. Curre=
nt
> >>>>>              EVPN DF election mechanism already eliminated loop and
> >>>>>              traffic duplication by configuring long reception
> >>>>>timer on
> >>>>>              each multi-homed PE, but up to reception timer traffic
> >>>>>              disruption issue still exist. EVPN for DCI is an
> >>>>>important
> >>>>>              use case for EVPN, up to reception timer traffic
> >>>>>              disruption can't be tolerated for service providers, i=
t
> >>>>>              should be improved.
> >>>>>
> >>>>>              Also for accuracy, i think handshaking state machine o=
n
> >>>>>              each multi-homed PE is also needed. From solution
> >>>>>              perspective, in my draft, no inter-chassis message is
> >>>>>              introduced, only one new extended community is
> >>>>>introduced,
> >>>>>              i think the process is comparatively simple than your
> >>>>>              following draft.
> >>>>>
> >>>>>              Current EVPN DF election has some drawbacks, so there
> >>>>>are
> >>>>>              three new drafts about DF election emerged. I think BE=
SS
> >>>>>              WG can consider these three drafts in global view, a
> >>>>>              single,comprehensive new DF election draft is hoped.
> >>>>>
> >>>>>              thanks.
> >>>>>
> >>>>>              Thanks,
> >>>>>
> >>>>>              weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>>              *From:*BESS [bess-bounces@ietf.org
> >>>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajas=
si
> >>>>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com=
>]
> >>>>>              *Sent:* Thursday, March 26, 2015 0:21
> >>>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
> >>>>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
> >>>>>              based on Paxos algorithm
> >>>>>
> >>>>>              FYI- First published July 4, 2011
> >>>>>
> >>>>>
> >>>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-r
> >>>>>out
> >>>>>e
> >>>>>/
> >>>>>
> >>>>>              -Ali
> >>>>>
> >>>>>
> >>>>>
> >>>>> _______________________________________________
> >>>>> BESS mailing list
> >>>>> BESS@ietf.org
> >>>>> https://www.ietf.org/mailman/listinfo/bess
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>______________________________________________________
> ______________
> >>>>___
> >>>>_
> >>>>_
> >>>>________________________________________________
> >>>>
> >>>> Ce message et ses pieces jointes peuvent contenir des informations
> >>>>confidentielles ou privilegiees et ne doivent donc  pas etre
> >>>>diffuses, exploites ou copies sans autorisation. Si vous avez recu
> >>>>ce message par erreur, veuillez le signaler  a l'expediteur et le
> >>>>detruire ainsi que les pieces jointes. Les messages electroniques
> >>>>etant susceptibles d'alteration,  Orange decline toute
> >>>>responsabilite si ce message a ete altere, deforme ou falsifie.
> >>>>Merci.
> >>>>
> >>>> This message and its attachments may contain confidential or
> >>>>privileged information that may be protected by law;  they should
> >>>>not be distributed, used or copied without authorisation.
> >>>> If you have received this email in error, please notify the sender
> >>>>and delete this message and its attachments.
> >>>> As emails may be altered, Orange is not liable for messages that
> >>>>have been modified, changed or falsified.
> >>>> Thank you.
> >>>>
> >>>> _______________________________________________
> >>>> BESS mailing list
> >>>> BESS@ietf.org
> >>>> https://www.ietf.org/mailman/listinfo/bess
> >>>>
> >>>
> >>>
> >>>_______________________________________________________
> ______________
> >>>___
> >>>_
> >>>_
> >>>_______________________________________________
> >>>
> >>>Ce message et ses pieces jointes peuvent contenir des informations
> >>>confidentielles ou privilegiees et ne doivent donc pas etre diffuses,
> >>>exploites ou copies sans autorisation. Si vous avez recu ce message
> >>>par erreur, veuillez le signaler a l'expediteur et le detruire ainsi
> >>>que les pieces jointes. Les messages electroniques etant susceptibles
> >>>d'alteration, Orange decline toute responsabilite si ce message a ete
> >>>altere, deforme ou falsifie. Merci.
> >>>
> >>>This message and its attachments may contain confidential or
> >>>privileged information that may be protected by law; they should not
> >>>be distributed, used or copied without authorisation.
> >>>If you have received this email in error, please notify the sender
> >>>and delete this message and its attachments.
> >>>As emails may be altered, Orange is not liable for messages that have
> >>>been modified, changed or falsified.
> >>>Thank you.
> >>>_______________________________________________
> >>>BESS mailing list
> >>>BESS@ietf.org
> >>>https://www.ietf.org/mailman/listinfo/bess
> >>_______________________________________________
> >>BESS mailing list
> >>BESS@ietf.org
> >>https://www.ietf.org/mailman/listinfo/bess
> >_______________________________________________
> >BESS mailing list
> >BESS@ietf.org
> >https://www.ietf.org/mailman/listinfo/bess
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess


From nobody Fri Mar 27 08:27:57 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A863A1ACEFA for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:27:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.611
X-Spam-Level: 
X-Spam-Status: No, score=-3.611 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1UuXcsERWawI for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:27:50 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CC0111A0040 for <bess@ietf.org>; Fri, 27 Mar 2015 08:27:49 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml402-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BUF18508; Fri, 27 Mar 2015 15:27:48 +0000 (GMT)
Received: from nkgeml409-hub.china.huawei.com (10.98.56.40) by lhreml402-hub.china.huawei.com (10.201.5.241) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 15:27:47 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml409-hub.china.huawei.com ([10.98.56.40]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 23:27:42 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: John E Drake <jdrake@juniper.net>, "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "EXT - thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HYAABS9Z9AAD2dSgAAN/i8VACwBuAAARy7QAw==
Date: Fri, 27 Mar 2015 15:27:41 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F8577BC@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>, <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562504AAF732AF476E5C218C7090@BLUPR05MB562.namprd05.prod.outlook.com>
In-Reply-To: <BLUPR05MB562504AAF732AF476E5C218C7090@BLUPR05MB562.namprd05.prod.outlook.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.144.149]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/FLxDXVbf9WC6Sqeq6TB7yzPf8cE>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 15:27:55 -0000

Hi John,
If you can tolerate indefinite forwarding loop during transiet time, i have=
 no problem. If the community think it is a serious problem, i think we mus=
t propose an applicable solution for it, maybe there are other solutions be=
tter than handshaking mechanism.

Thanks,
weiguo

________________________________________
From: John E Drake [jdrake@juniper.net]
Sent: Friday, March 27, 2015 23:22
To: Haoweiguo; Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT - thom=
as.morin@orange.com; bess@ietf.org
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

I think we have beaten this topic to death.  For the numerous reason I've g=
iven, I'm not interested in you proposal.

Yours Irrespectively,

John

> -----Original Message-----
> From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
> Sent: Friday, March 27, 2015 11:12 AM
> To: Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT -
> thomas.morin@orange.com; bess@ietf.org
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>
> Hi Jorge,
> Let give you another indefinitely forwarding loop case, the picture is as
> follows:
>
> CE3------------CE4
>   |                        |
> PE3                   PE4
>
>
>
> PE1                   PE2
>  |                         |
> CE1------------CE2
>
> If PE1 and PE2 are dual DF PE in transiet time, assuing PE3 and PE4 are i=
n
> stable state ,PE3 is DF PE, PE4 is non-DF PE, the BUM traffic from CE3 to=
 CE1,
> the forwarding loop path is as follows;
> CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->PE4---=
-
> >forever......
>
> Do you think the harm is enough?
> Thanks,
> weiguo
> ________________________________________
> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
> Sent: Friday, March 27, 2015 22:59
> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
> bess@ietf.org
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>
> Hi Weiguo,
>
> OK, thanks for clarifying, that is the part I did not understand in your =
loop :-)
>
> To me a loop means you send a frame and that frame keeps being forwarded
> by the same nodes indefinitely till you take a corrective action.
> In this case, CE3 might get a few packets during the transient state, but=
 those
> packets will have its own MAC SA so CE3 will discard them. I don't think =
that is
> harmful for CE3 in most of the cases. And if it is, you just need to make=
 sure
> the timers are long enough and the DF election properly implemented to
> avoid the transient DF-DF state.
>
> The handshaking is definitely overkilling IMHO.
>
> Thanks.
> Jorge
>
> -----Original Message-----
> From: Haoweiguo <haoweiguo@huawei.com>
> Date: Friday, March 27, 2015 at 7:29 AM
> To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
> (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
> <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>
> >Hi Jorge,
> >Sorry, maybe i have not explain clearly, pls allow me to clarify it
> >more clearly.
> >In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2
> >is single homed to remote PE3. CE1 and CE2 forms one layer 2 network,
> >so it multi-homed network scenario.
> >           CE3
> >             |
> >            PE3
> >
> >
> >PE1                  PE2
> >  |                       |
> >CE1-----------CE2
> >
> >In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends
> >BUM traffic to CE1, the traffic will reach PE1 and PE2, CE1 will
> >receive two copy of the traffic from CE3, also loop is formed, the form =
path:
> >
> >CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
> >
> >I think the traffic loop is absolutely not allowed. In MHN case, the
> >loop will formed due to dual-DF PE in transiet period.
> >
> >Thanks,
> >weiguo
> >
> >
> >
> >________________________________________
> >From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
> >Sent: Friday, March 27, 2015 20:26
> >To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
> >bess@ietf.org
> >Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> >algorithm
> >
> >Hi Weiguo,
> >
> >About this:
> >
> >[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
> >case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
> >and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
> >loop back to CE1 through PE2. If CE1 can filter and drop the traffic ,
> >it is fortunate, maybe no loop.
> >For multi-homed network scenario, say local network1(CE1 and CE2) is
> >multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
> >back through PE1-->PE2-->CE2, it is very serious. It is absolutely not a=
llowed.
> >
> >
> >[JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2
> >will not send the traffic back to CE1 due to the ESI-label and
> >split-horizon procedures.
> >In MH network case PE2 will NOT send the traffic to CE2 for the same
> >reason.
> >
> >Let me know if you disagree please.
> >
> >Thanks.
> >Jorge
> >
> >-----Original Message-----
> >From: Haoweiguo <haoweiguo@huawei.com>
> >Date: Thursday, March 26, 2015 at 9:55 PM
> >To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
> >"thomas.morin@orange.com" <thomas.morin@orange.com>,
> "bess@ietf.org"
> ><bess@ietf.org>
> >Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> >algorithm
> >
> >>Hi Satya,
> >>Pls see inline.
> >>
> >>Thanks,
> >>weiguo
> >>
> >>________________________________________
> >>From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
> >>Sent: Friday, March 27, 2015 12:11
> >>To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
> >>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> >>algorithm
> >>
> >>Thomas,
> >>
> >>Thanks for your summary. It captures the essence of the discussion.
> >>[weiguo]: Yes, i agree.
> >>
> >>One feature of the HRW scheme is that it computes the Active and the
> >>Backup DF upfront.
> >>Whenever the Active PE, say PE1, dies; the Backup PE, PE2, can become
> >>the Active immediately upon receiving the withdraw of the ES route
> >>from PE1.
> >>It need not even have to wait for the 3 seconds time.
> >>[weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
> >>message:) But when PE1 leaves ESI, it can withdraw the ES route
> >>immediately.
> >>When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2
> >>will both act as PE for some transiet time.
> >>
> >>On the other hand, the 3 second timer will serve as a bulk and batch
> >>mechanism for the case when new PE(s) is(are) coming up, so we don=B9t
> >>have to do a DF election ever so often.
> >>
> >>Haoweiguo,  I think the "loop" scenario is speculative.
> >>I think everybody understands that, yes, there may be some transient
> >>duplication of bum traffic; however, there is no "loop" problem to be
> >>solved.
> >>[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
> >>case. In multi-homed device case, say CE1 is multi-homed to PE1 and
> >>PE2 and both PE are DF PE. When CE1 sends traffic to PE1, the traffic
> >>will loop back to CE1 through PE2. If CE1 can filter and drop the
> >>traffic , it is fortunate, maybe no loop.
> >>For multi-homed network scenario, say local network1(CE1 and CE2) is
> >>multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
> >>back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
> >>allowed.
> >>
> >>Thanks,
> >>--Satya
> >>
> >>
> >>
> >>On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
> >>
> >>>Hi Thomas,
> >>>I know you are neutral, sorry for my reply vulnerable to be
> >>>miunderstood.
> >>>
> >>>Thanks,
> >>>weiguo
> >>>
> >>>________________________________________
> >>>From: thomas.morin@orange.com [thomas.morin@orange.com]
> >>>Sent: Friday, March 27, 2015 11:59
> >>>To: Haoweiguo; bess@ietf.org
> >>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on
> >>>Paxos algorithm
> >>>
> >>>Hi Weiguo,
> >>>
> >>>2015-03-26, Haoweiguo:
> >>> > Thomas:
> >>>> > - RFC7432 may have transient periods where the DF election state
> >>>> > is not yet synchronized between the two peers:
> >>> >
> >>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic
> >>>> loop
> >>>
> >>>Note well that I didn't write that there can be transient _loops_ .
> >>>I tried to capture the exchange you had all, and what I gather is
> >>>that the split-horizon procedure _prevents_loops_, independently of
> >>>any transient period where DF state is not synchronized and which may
> >>>lead to transient _duplicates_.
> >>>
> >>>-Thomas
> >>>
> >>>>
> >>>> 26/03/2015 20:05, John E Drake :
> >>>>>
> >>>>> Weiguo,
> >>>>>
> >>>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I hav=
e
> >>>>> detailed, is a non-solution to a non-problem with tremendous
> >>>>> control plane cost.
> >>>>>
> >>>>> Yours Irrespectively,
> >>>>>
> >>>>> John
> >>>>>
> >>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
> >>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> Pls see below.
> >>>>>
> >>>>> Thanks,
> >>>>>
> >>>>> weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>> *From:*John E Drake [jdrake@juniper.net]
> >>>>> *Sent:* Friday, March 27, 2015 6:00
> >>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> To recap,
> >>>>>
> >>>>> We have established that your proposal is untenable because of its
> >>>>> control plane load.
> >>>>>
> >>>>> We have established that your proposal is based upon a flawed
> >>>>> understanding of the DF election in RFC 7432.
> >>>>>
> >>>>> [weiguo]: In ethernet world, traffic loop is serious than short
> >>>>> timer traffic disruption. If you want to implement  transiet
> >>>>> traffic loop process, i will modify my draft to solve your issue.
> >>>>>
> >>>>> If i am the developer, i will prefer short timer traffic
> >>>>> disruption based on current EVPN protocol.
> >>>>>
> >>>>> What you are now arguing is that your draft prevents two or more
> >>>>> PEs from being DF simultaneously. This is clearly nonsense.
> >>>>>
> >>>>> [weiguo]: I will modify the draft problem statements, and use the
> >>>>>same  handshaking solution to solve it.
> >>>>>
> >>>>> Furthermore, we have established that having two or more DFs for
> >>>>> what even you admit is a brief transient leads to duplicate
> >>>>> traffic, which is acceptable, but not loops, your assertion to the
> contrary.
> >>>>>
> >>>>> [weiguo]: It is transient loop and traffic duplication issue.
> >>>>>
> >>>>> Yours Irrespectively,
> >>>>>
> >>>>> John
> >>>>>
> >>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
> >>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> John,
> >>>>>
> >>>>> As your understanding of the EVPN draft, the DF election mechanism
> >>>>>has  more serious side effect, it will have short time traffic
> >>>>>loop,i.e.,  dual DF PEs will exist for a short time. I think dual
> >>>>>DF PEs is  absolutely not tolerated, because native ethernet header
> >>>>>has no TTL,  up to several hundred ms traffic loop normally not
> >>>>>tolerated in  commertial networks.
> >>>>>
> >>>>> As your understanding, the PEs should do as following:
> >>>>>
> >>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but
> >>>>> have rarely deployment.
> >>>>>
> >>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3
> >>>>> joins ESI and trigger DF re-election. When reception timer expires:
> >>>>>
> >>>>> PE1 upgrades to DF PE.
> >>>>>
> >>>>> After reception timer+ ES route transmission timer:
> >>>>>
> >>>>> PE2 downloads to non-DF PE.
> >>>>>
> >>>>> So in timer clock sync case, dual DF PEs will exist at least
> >>>>> transmission timer.
> >>>>>
> >>>>> If NTP is used for timer sync, because it has bad accuracy, dual
> >>>>> DF PEs will exist more longer timer.
> >>>>>
> >>>>> So as your understanding for DF election, the drawback is more clea=
r.
> >>>>>
> >>>>> Thanks,
> >>>>>
> >>>>> weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>> *From:*John E Drake [jdrake@juniper.net]
> >>>>> *Sent:* Friday, March 27, 2015 4:41
> >>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> Weiguo,
> >>>>>
> >>>>> We have already established that your proposal is untenable
> >>>>>because of  its control plane load.
> >>>>>
> >>>>> What we are now discussing is that your proposal is based upon a
> >>>>>misunderstanding of the algorithm in RFC 7432.  You are assuming
> >>>>>that
> >>>>> PE1 will advertise an ES route and then wait for the configured
> >>>>>interval before performing the DF election while PE2 and PE3 will
> >>>>>perform the DF election as soon as they receive the ES route from
> >>>>>PE1.
> >>>>> This is not what RFC 7432 says.
> >>>>>
> >>>>> Rather, what is says is that the advertisement of the ES route by
> >>>>> PE1 and its receipt by PE2 and PE3 causes all three PEs to start
> >>>>> the configured interval timer  - =B33. When the timer expires, each
> >>>>> PE builds an ordered list of the IP addresses of all the PE nodes
> >>>>> connected to the Ethernet segment (including itself), in
> >>>>> increasing numeric value.=B2
> >>>>>
> >>>>> Yours Irrespectively,
> >>>>>
> >>>>> John
> >>>>>
> >>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> >>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
> >>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
> >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> Pls read my detail replies to Satya. If you still can't catch it,
> >>>>> pls read my draft and EVPN base protocol,  thanks
> >>>>>
> >>>>> weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>> *From:*John E Drake [jdrake@juniper.net]
> >>>>> *Sent:* Friday, March 27, 2015 1:28
> >>>>> *To:* Patrice Brissette (pbrisset)
> >>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
> >>>>> <mailto:bess@ietf.org>
> >>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
> >>>>> Paxos algorithm
> >>>>>
> >>>>> I think Patrice is correct.  Your proposal doesn't solve the
> >>>>> problem and it does so at huge cost.
> >>>>>
> >>>>> Sent from my iPhone
> >>>>>
> >>>>>
> >>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
> >>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
> >>>>>
> >>>>>      Weiguo,
> >>>>>
> >>>>>      I=B9m not sure I=B9m following here.
> >>>>>
> >>>>>      Don=B9t you have the same issue with your handshaking mechanis=
m?
> >>>>>
> >>>>>      If you don=B9t know your peer, how can you handshake?
> >>>>>
> >>>>>      Regards,
> >>>>>
> >>>>>      Patrice
> >>>>>
> >>>>>      Image removed by sender.
> >>>>>
> >>>>>      *Patrice Brissette*
> >>>>>      TECHNICAL LEADER.ENGINEERING
> >>>>>
> >>>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
> >>>>>      Phone: *+1 613 254 3336*
> >>>>>
> >>>>>
> >>>>>
> >>>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
> >>>>>      Canada
> >>>>>      Cisco.com <http://www.cisco.com/global/CA/>
> >>>>>
> >>>>>
> >>>>>
> >>>>>      Image removed by sender.Think before you print.
> >>>>>
> >>>>>      This email may contain confidential and privileged material fo=
r
> >>>>>      the sole use of the intended recipient. Any review, use,
> >>>>>      distribution or disclosure by others is strictly prohibited. I=
f
> >>>>>      you are not the intended recipient (or authorized to receive f=
or
> >>>>>      the recipient), please contact the sender by reply email and
> >>>>>      delete all copies of this message.
> >>>>>
> >>>>>      Please click here
> >>>>>
> >>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.h
> tml>
> >>>>>      for Company Registration Information.
> >>>>>
> >>>>>
> >>>>>      *From: *Haoweiguo <haoweiguo@huawei.com
> >>>>><mailto:haoweiguo@huawei.com>>
> >>>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
> >>>>>      *To: *Patrice Brissette <pbrisset@cisco.com
> >>>>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
> >>>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
> >>>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
> >>>>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
> >>>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES
> >>>>>based on
> >>>>>      Paxos algorithm
> >>>>>
> >>>>>          Hi Patrice,
> >>>>>
> >>>>>          Up to reception timer traffic disruption in transient
> >>>>>phase is
> >>>>>          one of the issues.
> >>>>>
> >>>>>          Thanks,
> >>>>>
> >>>>>          weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
> >>>>>          <mailto:pbrisset@cisco.com>]
> >>>>>          *Sent:* Thursday, March 26, 2015 20:54
> >>>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
> >>>>>          bess@ietf.org <mailto:bess@ietf.org>
> >>>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
> >>>>>          based on Paxos algorithm
> >>>>>
> >>>>>          Weiguo,
> >>>>>
> >>>>>          You mention "But if your draft have not solved all
> >>>>> issues=B2,
> >>>>>
> >>>>>          Can you explain what Satya=B9s draft is not solving?
> >>>>>
> >>>>>          Regards,
> >>>>>
> >>>>>          Patrice
> >>>>>
> >>>>>          Image removed by sender.
> >>>>>
> >>>>>          *Patrice Brissette*
> >>>>>          TECHNICAL LEADER.ENGINEERING
> >>>>>
> >>>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
> >>>>>          Phone: *+1 613 254 3336*
> >>>>>
> >>>>>
> >>>>>
> >>>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
> >>>>>          Canada
> >>>>>          Cisco.com <http://www.cisco.com/global/CA/>
> >>>>>
> >>>>>
> >>>>>
> >>>>>          Image removed by sender.Think before you print.
> >>>>>
> >>>>>          This email may contain confidential and privileged materia=
l
> >>>>>          for the sole use of the intended recipient. Any review, us=
e,
> >>>>>          distribution or disclosure by others is strictly prohibite=
d.
> >>>>>          If you are not the intended recipient (or authorized to
> >>>>>          receive for the recipient), please contact the sender by
> >>>>>reply
> >>>>>          email and delete all copies of this message.
> >>>>>
> >>>>>          Please click here
> >>>>>
> >>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.h
> tml>
> >>>>>          for Company Registration Information.
> >>>>>
> >>>>>
> >>>>>          *From: *Haoweiguo <haoweiguo@huawei.com
> >>>>>          <mailto:haoweiguo@huawei.com>>
> >>>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
> >>>>>          *To: *John E Drake <jdrake@juniper.net
> >>>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.c=
om
> >>>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
> >>>>>          <mailto:bess@ietf.org>" <bess@ietf.org
> >>>>><mailto:bess@ietf.org>>
> >>>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
> >>>>>          based on Paxos algorithm
> >>>>>
> >>>>>              Hi John,
> >>>>>
> >>>>>               Firstly i think EVPN community should reach
> >>>>>consensus on
> >>>>>              the issues of current DF election mechanism. All these
> >>>>>              issues should be resolved in a single new DF election
> >>>>>              draft,rather than in multiple separate drafts. If your
> >>>>>              draft can solve all these issues and stable, i have no
> >>>>>              question for its progressing. But if your draft have n=
ot
> >>>>>              solved all issues, i think it had better combine with
> >>>>>              other drafts to provide a comprehensive solution. I
> >>>>>think
> >>>>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01
> >>>>>and
> >>>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it
> >>>>>should
> >>>>>              be resolved. So i think although your new Hash algorit=
hm
> >>>>>              for DF election is good, it only includes partial
> >>>>>              enhancements, maybe it still needs some time for
> >>>>>consensus.
> >>>>>
> >>>>>              Thanks,
> >>>>>
> >>>>>              weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>>              *From:*John E Drake [jdrake@juniper.net
> >>>>>              <mailto:jdrake@juniper.net>]
> >>>>>              *Sent:* Thursday, March 26, 2015 7:16
> >>>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
> >>>>>              <mailto:bess@ietf.org>
> >>>>>              *Subject:* RE: [bess] Handshaking among PEs in an
> >>>>>EVPN ES
> >>>>>              based on Paxos algorithm
> >>>>>
> >>>>>              Weiguo,
> >>>>>
> >>>>>              Your proposal introduces a control plane processing lo=
ad
> >>>>>              that is O(#EVIs * PEs) per DF election and given that
> >>>>>              there can be 4K EVIs per ES, this looks like a
> >>>>>              **substantial** load. Furthermore,
> >>>>>
> >>>>>              you can=B9t  use the ES route to co-ordinate DF electi=
on
> >>>>>              because you would need to carry your new extended
> >>>>>              community for each EVI and they would not all fit.  Yo=
u
> >>>>>              also can=B9t use the Per EVI Ethernet AD route because
> >>>>>that
> >>>>>              is processed by all PEs in the EVI.
> >>>>>
> >>>>>              I think that from a practical perspective the new DF
> >>>>>              election proposed in Satya=B9s draft is sufficiently
> >>>>>stable
> >>>>>              that it renders your draft moot, even if it could be
> >>>>>made
> >>>>>              to work.
> >>>>>
> >>>>>              Yours Irrespectively,
> >>>>>
> >>>>>              John
> >>>>>
> >>>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf =
Of
> >>>>>              *Haoweiguo
> >>>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
> >>>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
> >>>>>              <mailto:bess@ietf.org>
> >>>>>              *Subject:* Re: [bess] Handshaking among PEs in an
> >>>>>EVPN ES
> >>>>>              based on Paxos algorithm
> >>>>>
> >>>>>              Hi Ali,
> >>>>>
> >>>>>              Thanks for your information. I scanned through this
> >>>>>draft,
> >>>>>              it really introduces inter-chassis message for DF
> >>>>>election
> >>>>>              handshaking, the requirements in this draft is to
> >>>>>              eliminate transiet Loop and traffic duplication. Curre=
nt
> >>>>>              EVPN DF election mechanism already eliminated loop and
> >>>>>              traffic duplication by configuring long reception
> >>>>>timer on
> >>>>>              each multi-homed PE, but up to reception timer traffic
> >>>>>              disruption issue still exist. EVPN for DCI is an
> >>>>>important
> >>>>>              use case for EVPN, up to reception timer traffic
> >>>>>              disruption can't be tolerated for service providers, i=
t
> >>>>>              should be improved.
> >>>>>
> >>>>>              Also for accuracy, i think handshaking state machine o=
n
> >>>>>              each multi-homed PE is also needed. From solution
> >>>>>              perspective, in my draft, no inter-chassis message is
> >>>>>              introduced, only one new extended community is
> >>>>>introduced,
> >>>>>              i think the process is comparatively simple than your
> >>>>>              following draft.
> >>>>>
> >>>>>              Current EVPN DF election has some drawbacks, so there
> >>>>>are
> >>>>>              three new drafts about DF election emerged. I think BE=
SS
> >>>>>              WG can consider these three drafts in global view, a
> >>>>>              single,comprehensive new DF election draft is hoped.
> >>>>>
> >>>>>              thanks.
> >>>>>
> >>>>>              Thanks,
> >>>>>
> >>>>>              weiguo
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------------------
> >>>>>---
> >>>>>-
> >>>>>-
> >>>>>
> >>>>>              *From:*BESS [bess-bounces@ietf.org
> >>>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajas=
si
> >>>>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com=
>]
> >>>>>              *Sent:* Thursday, March 26, 2015 0:21
> >>>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
> >>>>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
> >>>>>              based on Paxos algorithm
> >>>>>
> >>>>>              FYI- First published July 4, 2011
> >>>>>
> >>>>>
> >>>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-r
> >>>>>out
> >>>>>e
> >>>>>/
> >>>>>
> >>>>>              -Ali
> >>>>>
> >>>>>
> >>>>>
> >>>>> _______________________________________________
> >>>>> BESS mailing list
> >>>>> BESS@ietf.org
> >>>>> https://www.ietf.org/mailman/listinfo/bess
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>______________________________________________________
> ______________
> >>>>___
> >>>>_
> >>>>_
> >>>>________________________________________________
> >>>>
> >>>> Ce message et ses pieces jointes peuvent contenir des informations
> >>>>confidentielles ou privilegiees et ne doivent donc  pas etre
> >>>>diffuses, exploites ou copies sans autorisation. Si vous avez recu
> >>>>ce message par erreur, veuillez le signaler  a l'expediteur et le
> >>>>detruire ainsi que les pieces jointes. Les messages electroniques
> >>>>etant susceptibles d'alteration,  Orange decline toute
> >>>>responsabilite si ce message a ete altere, deforme ou falsifie.
> >>>>Merci.
> >>>>
> >>>> This message and its attachments may contain confidential or
> >>>>privileged information that may be protected by law;  they should
> >>>>not be distributed, used or copied without authorisation.
> >>>> If you have received this email in error, please notify the sender
> >>>>and delete this message and its attachments.
> >>>> As emails may be altered, Orange is not liable for messages that
> >>>>have been modified, changed or falsified.
> >>>> Thank you.
> >>>>
> >>>> _______________________________________________
> >>>> BESS mailing list
> >>>> BESS@ietf.org
> >>>> https://www.ietf.org/mailman/listinfo/bess
> >>>>
> >>>
> >>>
> >>>_______________________________________________________
> ______________
> >>>___
> >>>_
> >>>_
> >>>_______________________________________________
> >>>
> >>>Ce message et ses pieces jointes peuvent contenir des informations
> >>>confidentielles ou privilegiees et ne doivent donc pas etre diffuses,
> >>>exploites ou copies sans autorisation. Si vous avez recu ce message
> >>>par erreur, veuillez le signaler a l'expediteur et le detruire ainsi
> >>>que les pieces jointes. Les messages electroniques etant susceptibles
> >>>d'alteration, Orange decline toute responsabilite si ce message a ete
> >>>altere, deforme ou falsifie. Merci.
> >>>
> >>>This message and its attachments may contain confidential or
> >>>privileged information that may be protected by law; they should not
> >>>be distributed, used or copied without authorisation.
> >>>If you have received this email in error, please notify the sender
> >>>and delete this message and its attachments.
> >>>As emails may be altered, Orange is not liable for messages that have
> >>>been modified, changed or falsified.
> >>>Thank you.
> >>>_______________________________________________
> >>>BESS mailing list
> >>>BESS@ietf.org
> >>>https://www.ietf.org/mailman/listinfo/bess
> >>_______________________________________________
> >>BESS mailing list
> >>BESS@ietf.org
> >>https://www.ietf.org/mailman/listinfo/bess
> >_______________________________________________
> >BESS mailing list
> >BESS@ietf.org
> >https://www.ietf.org/mailman/listinfo/bess
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess=


From nobody Fri Mar 27 08:37:39 2015
Return-Path: <lucy.yong@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 348981A0040 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:37:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.21
X-Spam-Level: 
X-Spam-Status: No, score=-4.21 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id SykPWFWdsex9 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:37:37 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 097401A0282 for <bess@ietf.org>; Fri, 27 Mar 2015 08:37:23 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT95596; Fri, 27 Mar 2015 15:37:22 +0000 (GMT)
Received: from DFWEML702-CHM.china.huawei.com (10.193.5.72) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 15:37:21 +0000
Received: from DFWEML701-CHM.china.huawei.com ([10.193.5.50]) by dfweml702-chm ([10.193.5.72]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 08:37:17 -0700
From: Lucy yong <lucy.yong@huawei.com>
To: "thomas.morin@orange.com" <thomas.morin@orange.com>, "martin.vigoureux@alcatel-lucent.com" <martin.vigoureux@alcatel-lucent.com>
Thread-Topic: WG adoption request for draft-hao-bess-inter-nvo3-vpn
Thread-Index: AdBopBPY9q/FfYjWSw2YrfJi6HtEaw==
Date: Fri, 27 Mar 2015 15:37:17 +0000
Message-ID: <2691CE0099834E4A9C5044EEC662BB9D57154528@dfweml701-chm>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.158.94]
Content-Type: multipart/alternative; boundary="_000_2691CE0099834E4A9C5044EEC662BB9D57154528dfweml701chm_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/0KPFbpVLpWCKQIdORsT8Ht6qypk>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: [bess] WG adoption request for draft-hao-bess-inter-nvo3-vpn
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 15:37:38 -0000

--_000_2691CE0099834E4A9C5044EEC662BB9D57154528dfweml701chm_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Dear Thomas and Martin,

I presented the draft-hao-bess-inter-nvo3-vpn in BESS meeting. Due to the c=
utting time, the meeting did not discuss our request: adopting this draft a=
s WG draft.

BESS already adopted some nvo3-vpn interconnection drafts that are based on=
 non-inter-as solutions. We think that the WG should also adopt inter-as ba=
sed solutions that are useful for some scenarios.

Some operator also expressed the use of inter-as option C in this space, wh=
ich is described in draft-hao-inter-nvo3-vpn-optionc.   We ask to adopt thi=
s draft as well.

Look forward to hearing from your action.

Regards,
Lucy

--_000_2691CE0099834E4A9C5044EEC662BB9D57154528dfweml701chm_
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=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: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;}
@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=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div class=3D"WordSection1">
<p class=3D"MsoNormal">Dear Thomas and Martin,<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">I presented the draft-hao-bess-inter-nvo3-vpn in BES=
S meeting. Due to the cutting time, the meeting did not discuss our request=
: adopting this draft as WG draft.<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">BESS already adopted some nvo3-vpn interconnection d=
rafts that are based on non-inter-as solutions. We think that the WG should=
 also adopt inter-as based solutions that are useful for some scenarios.<o:=
p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Some operator also expressed the use of inter-as opt=
ion C in this space, which is described in draft-hao-inter-nvo3-vpn-optionc=
. &nbsp;&nbsp;We ask to adopt this draft as well.<o:p></o:p></p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Look forward to hearing from your action.<o:p></o:p>=
</p>
<p class=3D"MsoNormal"><o:p>&nbsp;</o:p></p>
<p class=3D"MsoNormal">Regards,<o:p></o:p></p>
<p class=3D"MsoNormal">Lucy<o:p></o:p></p>
</div>
</body>
</html>

--_000_2691CE0099834E4A9C5044EEC662BB9D57154528dfweml701chm_--


From nobody Fri Mar 27 08:37:56 2015
Return-Path: <jdrake@juniper.net>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CA6251A0231 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:37:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.302
X-Spam-Level: 
X-Spam-Status: No, score=-1.302 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id C6KpeLlcCcy4 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:37:38 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1on0711.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::711]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5CB101A0178 for <bess@ietf.org>; Fri, 27 Mar 2015 08:37:33 -0700 (PDT)
Received: from BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) by BLUPR05MB562.namprd05.prod.outlook.com (10.141.202.141) with Microsoft SMTP Server (TLS) id 15.1.118.21; Fri, 27 Mar 2015 15:37:14 +0000
Received: from BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) by BLUPR05MB562.namprd05.prod.outlook.com ([10.141.202.141]) with mapi id 15.01.0118.022; Fri, 27 Mar 2015 15:37:14 +0000
From: John E Drake <jdrake@juniper.net>
To: Haoweiguo <haoweiguo@huawei.com>, "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "EXT - thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8P/o146fJTTukSNC2GolBinQZ0u4lsAgAACyoCAAB/gjYAAILMAgAAMEtCAABidAIAAA48AgAAHqACAAC468IAAHjiAgAAP+ACAAAM7AIAAApwAgAAA7YCAAAw6gIAAfgsAgAAiUICAAAhggIAAA1IAgAACs5CAAAHSgIAAAU7Q
Date: Fri, 27 Mar 2015 15:37:13 +0000
Message-ID: <BLUPR05MB562BC04BEB54A67589A796DC7090@BLUPR05MB562.namprd05.prod.outlook.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>, <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562504AAF732AF476E5C218C7090@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8577BC@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F8577BC@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [66.129.239.10]
authentication-results: huawei.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BLUPR05MB562;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(377454003)(164054003)(52604005)(479174004)(51704005)(36304003)(69234005)(13464003)(377424004)(24454002)(2656002)(15975445007)(77096005)(102836002)(62966003)(77156002)(76576001)(99286002)(87936001)(66066001)(92566002)(74316001)(19580395003)(19580405001)(106116001)(2501003)(33656002)(5890100001)(561944003)(46102003)(76176999)(54356999)(50986999)(86362001)(2900100001)(122556002)(93886004)(107886001)(2950100001)(579004)(559001); DIR:OUT; SFP:1102; SCL:1; SRVR:BLUPR05MB562; H:BLUPR05MB562.namprd05.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BLUPR05MB56212AA040B5DA93A3B8749C7090@BLUPR05MB562.namprd05.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BLUPR05MB562; BCL:0; PCL:0; RULEID:; SRVR:BLUPR05MB562; 
x-forefront-prvs: 0528942FD8
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: juniper.net
X-MS-Exchange-CrossTenant-originalarrivaltime: 27 Mar 2015 15:37:13.7255 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: bea78b3c-4cdb-4130-854a-1d193232e5f4
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BLUPR05MB562
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/AIC_Kl0Lf57yuJoIJ3mB3a99N_o>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 15:37:44 -0000

It goes without saying that there are better solutions.  As Ali pointed out=
, he proposed one years ago and decided to drop it because he decided the p=
roblem wasn't worth solving.  This is a sentiment I share.=20

Yours Irrespectively,

John

> -----Original Message-----
> From: Haoweiguo [mailto:haoweiguo@huawei.com]
> Sent: Friday, March 27, 2015 11:28 AM
> To: John E Drake; Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT -
> thomas.morin@orange.com; bess@ietf.org
> Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>=20
> Hi John,
> If you can tolerate indefinite forwarding loop during transiet time, i ha=
ve no
> problem. If the community think it is a serious problem, i think we must
> propose an applicable solution for it, maybe there are other solutions be=
tter
> than handshaking mechanism.
>=20
> Thanks,
> weiguo
>=20
> ________________________________________
> From: John E Drake [jdrake@juniper.net]
> Sent: Friday, March 27, 2015 23:22
> To: Haoweiguo; Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT -
> thomas.morin@orange.com; bess@ietf.org
> Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>=20
> I think we have beaten this topic to death.  For the numerous reason I've
> given, I'm not interested in you proposal.
>=20
> Yours Irrespectively,
>=20
> John
>=20
> > -----Original Message-----
> > From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
> > Sent: Friday, March 27, 2015 11:12 AM
> > To: Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT -
> > thomas.morin@orange.com; bess@ietf.org
> > Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> > algorithm
> >
> > Hi Jorge,
> > Let give you another indefinitely forwarding loop case, the picture is
> > as
> > follows:
> >
> > CE3------------CE4
> >   |                        |
> > PE3                   PE4
> >
> >
> >
> > PE1                   PE2
> >  |                         |
> > CE1------------CE2
> >
> > If PE1 and PE2 are dual DF PE in transiet time, assuing PE3 and PE4
> > are in stable state ,PE3 is DF PE, PE4 is non-DF PE, the BUM traffic
> > from CE3 to CE1, the forwarding loop path is as follows;
> > CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->PE4
> > CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->---
> > CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->-
> > >forever......
> >
> > Do you think the harm is enough?
> > Thanks,
> > weiguo
> > ________________________________________
> > From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
> > Sent: Friday, March 27, 2015 22:59
> > To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
> > bess@ietf.org
> > Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> > algorithm
> >
> > Hi Weiguo,
> >
> > OK, thanks for clarifying, that is the part I did not understand in
> > your loop :-)
> >
> > To me a loop means you send a frame and that frame keeps being
> > forwarded by the same nodes indefinitely till you take a corrective act=
ion.
> > In this case, CE3 might get a few packets during the transient state,
> > but those packets will have its own MAC SA so CE3 will discard them. I
> > don't think that is harmful for CE3 in most of the cases. And if it
> > is, you just need to make sure the timers are long enough and the DF
> > election properly implemented to avoid the transient DF-DF state.
> >
> > The handshaking is definitely overkilling IMHO.
> >
> > Thanks.
> > Jorge
> >
> > -----Original Message-----
> > From: Haoweiguo <haoweiguo@huawei.com>
> > Date: Friday, March 27, 2015 at 7:29 AM
> > To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
> > (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
> > <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
> > Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> > algorithm
> >
> > >Hi Jorge,
> > >Sorry, maybe i have not explain clearly, pls allow me to clarify it
> > >more clearly.
> > >In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2,
> > >CE2 is single homed to remote PE3. CE1 and CE2 forms one layer 2
> > >network, so it multi-homed network scenario.
> > >           CE3
> > >             |
> > >            PE3
> > >
> > >
> > >PE1                  PE2
> > >  |                       |
> > >CE1-----------CE2
> > >
> > >In transiet period, PE1 and PE2 are both DF PE. At this time, CE3
> > >sends BUM traffic to CE1, the traffic will reach PE1 and PE2, CE1
> > >will receive two copy of the traffic from CE3, also loop is formed, th=
e form
> path:
> > >
> > >CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
> > >
> > >I think the traffic loop is absolutely not allowed. In MHN case, the
> > >loop will formed due to dual-DF PE in transiet period.
> > >
> > >Thanks,
> > >weiguo
> > >
> > >
> > >
> > >________________________________________
> > >From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
> > >Sent: Friday, March 27, 2015 20:26
> > >To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
> > >bess@ietf.org
> > >Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on
> > >Paxos algorithm
> > >
> > >Hi Weiguo,
> > >
> > >About this:
> > >
> > >[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
> > >case. In multi-homed device case, say CE1 is multi-homed to PE1 and
> > >PE2 and both PE are DF PE. When CE1 sends traffic to PE1, the traffic
> > >will loop back to CE1 through PE2. If CE1 can filter and drop the
> > >traffic , it is fortunate, maybe no loop.
> > >For multi-homed network scenario, say local network1(CE1 and CE2) is
> > >multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
> > >back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
> allowed.
> > >
> > >
> > >[JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2.
> PE2
> > >will not send the traffic back to CE1 due to the ESI-label and
> > >split-horizon procedures.
> > >In MH network case PE2 will NOT send the traffic to CE2 for the same
> > >reason.
> > >
> > >Let me know if you disagree please.
> > >
> > >Thanks.
> > >Jorge
> > >
> > >-----Original Message-----
> > >From: Haoweiguo <haoweiguo@huawei.com>
> > >Date: Thursday, March 26, 2015 at 9:55 PM
> > >To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
> > >"thomas.morin@orange.com" <thomas.morin@orange.com>,
> > "bess@ietf.org"
> > ><bess@ietf.org>
> > >Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on
> > >Paxos algorithm
> > >
> > >>Hi Satya,
> > >>Pls see inline.
> > >>
> > >>Thanks,
> > >>weiguo
> > >>
> > >>________________________________________
> > >>From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
> > >>Sent: Friday, March 27, 2015 12:11
> > >>To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
> > >>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on
> > >>Paxos algorithm
> > >>
> > >>Thomas,
> > >>
> > >>Thanks for your summary. It captures the essence of the discussion.
> > >>[weiguo]: Yes, i agree.
> > >>
> > >>One feature of the HRW scheme is that it computes the Active and the
> > >>Backup DF upfront.
> > >>Whenever the Active PE, say PE1, dies; the Backup PE, PE2, can
> > >>become the Active immediately upon receiving the withdraw of the ES
> > >>route from PE1.
> > >>It need not even have to wait for the 3 seconds time.
> > >>[weiguo]: I think maybe your negligence, dies PE1 can't send
> > >>withdraw
> > >>message:) But when PE1 leaves ESI, it can withdraw the ES route
> > >>immediately.
> > >>When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2
> > >>will both act as PE for some transiet time.
> > >>
> > >>On the other hand, the 3 second timer will serve as a bulk and batch
> > >>mechanism for the case when new PE(s) is(are) coming up, so we don=B9=
t
> > >>have to do a DF election ever so often.
> > >>
> > >>Haoweiguo,  I think the "loop" scenario is speculative.
> > >>I think everybody understands that, yes, there may be some transient
> > >>duplication of bum traffic; however, there is no "loop" problem to
> > >>be solved.
> > >>[weiguo]: Traffic duplication and loop issue both exist in dual DF
> > >>PE case. In multi-homed device case, say CE1 is multi-homed to PE1
> > >>and
> > >>PE2 and both PE are DF PE. When CE1 sends traffic to PE1, the
> > >>traffic will loop back to CE1 through PE2. If CE1 can filter and
> > >>drop the traffic , it is fortunate, maybe no loop.
> > >>For multi-homed network scenario, say local network1(CE1 and CE2) is
> > >>multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will
> > >>loop back through PE1-->PE2-->CE2, it is very serious. It is
> > >>absolutely not allowed.
> > >>
> > >>Thanks,
> > >>--Satya
> > >>
> > >>
> > >>
> > >>On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
> > >>
> > >>>Hi Thomas,
> > >>>I know you are neutral, sorry for my reply vulnerable to be
> > >>>miunderstood.
> > >>>
> > >>>Thanks,
> > >>>weiguo
> > >>>
> > >>>________________________________________
> > >>>From: thomas.morin@orange.com [thomas.morin@orange.com]
> > >>>Sent: Friday, March 27, 2015 11:59
> > >>>To: Haoweiguo; bess@ietf.org
> > >>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on
> > >>>Paxos algorithm
> > >>>
> > >>>Hi Weiguo,
> > >>>
> > >>>2015-03-26, Haoweiguo:
> > >>> > Thomas:
> > >>>> > - RFC7432 may have transient periods where the DF election
> > >>>> > state is not yet synchronized between the two peers:
> > >>> >
> > >>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic
> > >>>> loop
> > >>>
> > >>>Note well that I didn't write that there can be transient _loops_ .
> > >>>I tried to capture the exchange you had all, and what I gather is
> > >>>that the split-horizon procedure _prevents_loops_, independently of
> > >>>any transient period where DF state is not synchronized and which
> > >>>may lead to transient _duplicates_.
> > >>>
> > >>>-Thomas
> > >>>
> > >>>>
> > >>>> 26/03/2015 20:05, John E Drake :
> > >>>>>
> > >>>>> Weiguo,
> > >>>>>
> > >>>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I
> > >>>>> have detailed, is a non-solution to a non-problem with
> > >>>>> tremendous control plane cost.
> > >>>>>
> > >>>>> Yours Irrespectively,
> > >>>>>
> > >>>>> John
> > >>>>>
> > >>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> > >>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
> > >>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> > >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> > >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based
> > >>>>> on Paxos algorithm
> > >>>>>
> > >>>>> Pls see below.
> > >>>>>
> > >>>>> Thanks,
> > >>>>>
> > >>>>> weiguo
> > >>>>>
> > >>>>>
> > >>>>>-----------------------------------------------------------------
> > >>>>>--
> > >>>>>---
> > >>>>>-
> > >>>>>-
> > >>>>>
> > >>>>> *From:*John E Drake [jdrake@juniper.net]
> > >>>>> *Sent:* Friday, March 27, 2015 6:00
> > >>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> > >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> > >>>>> <mailto:bess@ietf.org>
> > >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based
> > >>>>> on Paxos algorithm
> > >>>>>
> > >>>>> To recap,
> > >>>>>
> > >>>>> We have established that your proposal is untenable because of
> > >>>>> its control plane load.
> > >>>>>
> > >>>>> We have established that your proposal is based upon a flawed
> > >>>>> understanding of the DF election in RFC 7432.
> > >>>>>
> > >>>>> [weiguo]: In ethernet world, traffic loop is serious than short
> > >>>>> timer traffic disruption. If you want to implement  transiet
> > >>>>> traffic loop process, i will modify my draft to solve your issue.
> > >>>>>
> > >>>>> If i am the developer, i will prefer short timer traffic
> > >>>>> disruption based on current EVPN protocol.
> > >>>>>
> > >>>>> What you are now arguing is that your draft prevents two or more
> > >>>>> PEs from being DF simultaneously. This is clearly nonsense.
> > >>>>>
> > >>>>> [weiguo]: I will modify the draft problem statements, and use
> > >>>>>the same  handshaking solution to solve it.
> > >>>>>
> > >>>>> Furthermore, we have established that having two or more DFs for
> > >>>>> what even you admit is a brief transient leads to duplicate
> > >>>>> traffic, which is acceptable, but not loops, your assertion to
> > >>>>> the
> > contrary.
> > >>>>>
> > >>>>> [weiguo]: It is transient loop and traffic duplication issue.
> > >>>>>
> > >>>>> Yours Irrespectively,
> > >>>>>
> > >>>>> John
> > >>>>>
> > >>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> > >>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
> > >>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> > >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> > >>>>> <mailto:bess@ietf.org>
> > >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based
> > >>>>> on Paxos algorithm
> > >>>>>
> > >>>>> John,
> > >>>>>
> > >>>>> As your understanding of the EVPN draft, the DF election
> > >>>>>mechanism has  more serious side effect, it will have short time
> > >>>>>traffic loop,i.e.,  dual DF PEs will exist for a short time. I
> > >>>>>think dual DF PEs is  absolutely not tolerated, because native
> > >>>>>ethernet header has no TTL,  up to several hundred ms traffic
> > >>>>>loop normally not tolerated in  commertial networks.
> > >>>>>
> > >>>>> As your understanding, the PEs should do as following:
> > >>>>>
> > >>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but
> > >>>>> have rarely deployment.
> > >>>>>
> > >>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3
> > >>>>> joins ESI and trigger DF re-election. When reception timer expire=
s:
> > >>>>>
> > >>>>> PE1 upgrades to DF PE.
> > >>>>>
> > >>>>> After reception timer+ ES route transmission timer:
> > >>>>>
> > >>>>> PE2 downloads to non-DF PE.
> > >>>>>
> > >>>>> So in timer clock sync case, dual DF PEs will exist at least
> > >>>>> transmission timer.
> > >>>>>
> > >>>>> If NTP is used for timer sync, because it has bad accuracy, dual
> > >>>>> DF PEs will exist more longer timer.
> > >>>>>
> > >>>>> So as your understanding for DF election, the drawback is more
> clear.
> > >>>>>
> > >>>>> Thanks,
> > >>>>>
> > >>>>> weiguo
> > >>>>>
> > >>>>>
> > >>>>>-----------------------------------------------------------------
> > >>>>>--
> > >>>>>---
> > >>>>>-
> > >>>>>-
> > >>>>>
> > >>>>> *From:*John E Drake [jdrake@juniper.net]
> > >>>>> *Sent:* Friday, March 27, 2015 4:41
> > >>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
> > >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> > >>>>> <mailto:bess@ietf.org>
> > >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based
> > >>>>> on Paxos algorithm
> > >>>>>
> > >>>>> Weiguo,
> > >>>>>
> > >>>>> We have already established that your proposal is untenable
> > >>>>>because of  its control plane load.
> > >>>>>
> > >>>>> What we are now discussing is that your proposal is based upon a
> > >>>>>misunderstanding of the algorithm in RFC 7432.  You are assuming
> > >>>>>that
> > >>>>> PE1 will advertise an ES route and then wait for the configured
> > >>>>>interval before performing the DF election while PE2 and PE3 will
> > >>>>>perform the DF election as soon as they receive the ES route from
> > >>>>>PE1.
> > >>>>> This is not what RFC 7432 says.
> > >>>>>
> > >>>>> Rather, what is says is that the advertisement of the ES route
> > >>>>> by
> > >>>>> PE1 and its receipt by PE2 and PE3 causes all three PEs to start
> > >>>>> the configured interval timer  - =B33. When the timer expires,
> > >>>>> each PE builds an ordered list of the IP addresses of all the PE
> > >>>>> nodes connected to the Ethernet segment (including itself), in
> > >>>>> increasing numeric value.=B2
> > >>>>>
> > >>>>> Yours Irrespectively,
> > >>>>>
> > >>>>> John
> > >>>>>
> > >>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
> > >>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
> > >>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
> > >>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
> > >>>>> <mailto:bess@ietf.org>
> > >>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based
> > >>>>> on Paxos algorithm
> > >>>>>
> > >>>>> Pls read my detail replies to Satya. If you still can't catch
> > >>>>> it, pls read my draft and EVPN base protocol,  thanks
> > >>>>>
> > >>>>> weiguo
> > >>>>>
> > >>>>>
> > >>>>>-----------------------------------------------------------------
> > >>>>>--
> > >>>>>---
> > >>>>>-
> > >>>>>-
> > >>>>>
> > >>>>> *From:*John E Drake [jdrake@juniper.net]
> > >>>>> *Sent:* Friday, March 27, 2015 1:28
> > >>>>> *To:* Patrice Brissette (pbrisset)
> > >>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
> > >>>>> <mailto:bess@ietf.org>
> > >>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based
> > >>>>> on Paxos algorithm
> > >>>>>
> > >>>>> I think Patrice is correct.  Your proposal doesn't solve the
> > >>>>> problem and it does so at huge cost.
> > >>>>>
> > >>>>> Sent from my iPhone
> > >>>>>
> > >>>>>
> > >>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
> > >>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
> > >>>>>
> > >>>>>      Weiguo,
> > >>>>>
> > >>>>>      I=B9m not sure I=B9m following here.
> > >>>>>
> > >>>>>      Don=B9t you have the same issue with your handshaking
> mechanism?
> > >>>>>
> > >>>>>      If you don=B9t know your peer, how can you handshake?
> > >>>>>
> > >>>>>      Regards,
> > >>>>>
> > >>>>>      Patrice
> > >>>>>
> > >>>>>      Image removed by sender.
> > >>>>>
> > >>>>>      *Patrice Brissette*
> > >>>>>      TECHNICAL LEADER.ENGINEERING
> > >>>>>
> > >>>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
> > >>>>>      Phone: *+1 613 254 3336*
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
> > >>>>>      Canada
> > >>>>>      Cisco.com <http://www.cisco.com/global/CA/>
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>>      Image removed by sender.Think before you print.
> > >>>>>
> > >>>>>      This email may contain confidential and privileged material =
for
> > >>>>>      the sole use of the intended recipient. Any review, use,
> > >>>>>      distribution or disclosure by others is strictly prohibited.=
 If
> > >>>>>      you are not the intended recipient (or authorized to receive=
 for
> > >>>>>      the recipient), please contact the sender by reply email and
> > >>>>>      delete all copies of this message.
> > >>>>>
> > >>>>>      Please click here
> > >>>>>
> >
> >>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.h
> > tml>
> > >>>>>      for Company Registration Information.
> > >>>>>
> > >>>>>
> > >>>>>      *From: *Haoweiguo <haoweiguo@huawei.com
> > >>>>><mailto:haoweiguo@huawei.com>>
> > >>>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
> > >>>>>      *To: *Patrice Brissette <pbrisset@cisco.com
> > >>>>>      <mailto:pbrisset@cisco.com>>, John E Drake
> <jdrake@juniper.net
> > >>>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
> > >>>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
> > >>>>>      <mailto:bess@ietf.org>" <bess@ietf.org
> <mailto:bess@ietf.org>>
> > >>>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES
> > >>>>>based on
> > >>>>>      Paxos algorithm
> > >>>>>
> > >>>>>          Hi Patrice,
> > >>>>>
> > >>>>>          Up to reception timer traffic disruption in transient
> > >>>>>phase is
> > >>>>>          one of the issues.
> > >>>>>
> > >>>>>          Thanks,
> > >>>>>
> > >>>>>          weiguo
> > >>>>>
> > >>>>>
> > >>>>>-----------------------------------------------------------------
> > >>>>>--
> > >>>>>---
> > >>>>>-
> > >>>>>-
> > >>>>>
> > >>>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
> > >>>>>          <mailto:pbrisset@cisco.com>]
> > >>>>>          *Sent:* Thursday, March 26, 2015 20:54
> > >>>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
> > >>>>>          bess@ietf.org <mailto:bess@ietf.org>
> > >>>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN E=
S
> > >>>>>          based on Paxos algorithm
> > >>>>>
> > >>>>>          Weiguo,
> > >>>>>
> > >>>>>          You mention "But if your draft have not solved all
> > >>>>> issues=B2,
> > >>>>>
> > >>>>>          Can you explain what Satya=B9s draft is not solving?
> > >>>>>
> > >>>>>          Regards,
> > >>>>>
> > >>>>>          Patrice
> > >>>>>
> > >>>>>          Image removed by sender.
> > >>>>>
> > >>>>>          *Patrice Brissette*
> > >>>>>          TECHNICAL LEADER.ENGINEERING
> > >>>>>
> > >>>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
> > >>>>>          Phone: *+1 613 254 3336*
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CI=
E*
> > >>>>>          Canada
> > >>>>>          Cisco.com <http://www.cisco.com/global/CA/>
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>>          Image removed by sender.Think before you print.
> > >>>>>
> > >>>>>          This email may contain confidential and privileged mater=
ial
> > >>>>>          for the sole use of the intended recipient. Any review, =
use,
> > >>>>>          distribution or disclosure by others is strictly prohibi=
ted.
> > >>>>>          If you are not the intended recipient (or authorized to
> > >>>>>          receive for the recipient), please contact the sender
> > >>>>>by reply
> > >>>>>          email and delete all copies of this message.
> > >>>>>
> > >>>>>          Please click here
> > >>>>>
> >
> >>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.h
> > tml>
> > >>>>>          for Company Registration Information.
> > >>>>>
> > >>>>>
> > >>>>>          *From: *Haoweiguo <haoweiguo@huawei.com
> > >>>>>          <mailto:haoweiguo@huawei.com>>
> > >>>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
> > >>>>>          *To: *John E Drake <jdrake@juniper.net
> > >>>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco=
.com
> > >>>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
> > >>>>>          <mailto:bess@ietf.org>" <bess@ietf.org
> > >>>>><mailto:bess@ietf.org>>
> > >>>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN E=
S
> > >>>>>          based on Paxos algorithm
> > >>>>>
> > >>>>>              Hi John,
> > >>>>>
> > >>>>>               Firstly i think EVPN community should reach
> > >>>>>consensus on
> > >>>>>              the issues of current DF election mechanism. All the=
se
> > >>>>>              issues should be resolved in a single new DF electio=
n
> > >>>>>              draft,rather than in multiple separate drafts. If yo=
ur
> > >>>>>              draft can solve all these issues and stable, i have =
no
> > >>>>>              question for its progressing. But if your draft have=
 not
> > >>>>>              solved all issues, i think it had better combine wit=
h
> > >>>>>              other drafts to provide a comprehensive solution. I
> > >>>>>think
> > >>>>>              the issues listed in
> > >>>>>draft-rabadan-bess-evpn-ac-df-01 and
> > >>>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it
> > >>>>>should
> > >>>>>              be resolved. So i think although your new Hash algor=
ithm
> > >>>>>              for DF election is good, it only includes partial
> > >>>>>              enhancements, maybe it still needs some time for
> > >>>>>consensus.
> > >>>>>
> > >>>>>              Thanks,
> > >>>>>
> > >>>>>              weiguo
> > >>>>>
> > >>>>>
> > >>>>>-----------------------------------------------------------------
> > >>>>>--
> > >>>>>---
> > >>>>>-
> > >>>>>-
> > >>>>>
> > >>>>>              *From:*John E Drake [jdrake@juniper.net
> > >>>>>              <mailto:jdrake@juniper.net>]
> > >>>>>              *Sent:* Thursday, March 26, 2015 7:16
> > >>>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.or=
g
> > >>>>>              <mailto:bess@ietf.org>
> > >>>>>              *Subject:* RE: [bess] Handshaking among PEs in an
> > >>>>>EVPN ES
> > >>>>>              based on Paxos algorithm
> > >>>>>
> > >>>>>              Weiguo,
> > >>>>>
> > >>>>>              Your proposal introduces a control plane processing =
load
> > >>>>>              that is O(#EVIs * PEs) per DF election and given tha=
t
> > >>>>>              there can be 4K EVIs per ES, this looks like a
> > >>>>>              **substantial** load. Furthermore,
> > >>>>>
> > >>>>>              you can=B9t  use the ES route to co-ordinate DF elec=
tion
> > >>>>>              because you would need to carry your new extended
> > >>>>>              community for each EVI and they would not all fit.  =
You
> > >>>>>              also can=B9t use the Per EVI Ethernet AD route
> > >>>>>because that
> > >>>>>              is processed by all PEs in the EVI.
> > >>>>>
> > >>>>>              I think that from a practical perspective the new DF
> > >>>>>              election proposed in Satya=B9s draft is sufficiently
> > >>>>>stable
> > >>>>>              that it renders your draft moot, even if it could
> > >>>>>be made
> > >>>>>              to work.
> > >>>>>
> > >>>>>              Yours Irrespectively,
> > >>>>>
> > >>>>>              John
> > >>>>>
> > >>>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behal=
f Of
> > >>>>>              *Haoweiguo
> > >>>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
> > >>>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
> > >>>>>              <mailto:bess@ietf.org>
> > >>>>>              *Subject:* Re: [bess] Handshaking among PEs in an
> > >>>>>EVPN ES
> > >>>>>              based on Paxos algorithm
> > >>>>>
> > >>>>>              Hi Ali,
> > >>>>>
> > >>>>>              Thanks for your information. I scanned through this
> > >>>>>draft,
> > >>>>>              it really introduces inter-chassis message for DF
> > >>>>>election
> > >>>>>              handshaking, the requirements in this draft is to
> > >>>>>              eliminate transiet Loop and traffic duplication. Cur=
rent
> > >>>>>              EVPN DF election mechanism already eliminated loop a=
nd
> > >>>>>              traffic duplication by configuring long reception
> > >>>>>timer on
> > >>>>>              each multi-homed PE, but up to reception timer traff=
ic
> > >>>>>              disruption issue still exist. EVPN for DCI is an
> > >>>>>important
> > >>>>>              use case for EVPN, up to reception timer traffic
> > >>>>>              disruption can't be tolerated for service providers,=
 it
> > >>>>>              should be improved.
> > >>>>>
> > >>>>>              Also for accuracy, i think handshaking state machine=
 on
> > >>>>>              each multi-homed PE is also needed. From solution
> > >>>>>              perspective, in my draft, no inter-chassis message i=
s
> > >>>>>              introduced, only one new extended community is
> > >>>>>introduced,
> > >>>>>              i think the process is comparatively simple than you=
r
> > >>>>>              following draft.
> > >>>>>
> > >>>>>              Current EVPN DF election has some drawbacks, so
> > >>>>>there are
> > >>>>>              three new drafts about DF election emerged. I think =
BESS
> > >>>>>              WG can consider these three drafts in global view, a
> > >>>>>              single,comprehensive new DF election draft is hoped.
> > >>>>>
> > >>>>>              thanks.
> > >>>>>
> > >>>>>              Thanks,
> > >>>>>
> > >>>>>              weiguo
> > >>>>>
> > >>>>>
> > >>>>>-----------------------------------------------------------------
> > >>>>>--
> > >>>>>---
> > >>>>>-
> > >>>>>-
> > >>>>>
> > >>>>>              *From:*BESS [bess-bounces@ietf.org
> > >>>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali Saj=
assi
> > >>>>>              (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.c=
om>]
> > >>>>>              *Sent:* Thursday, March 26, 2015 0:21
> > >>>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
> > >>>>>              *Subject:* [bess] Handshaking among PEs in an EVPN E=
S
> > >>>>>              based on Paxos algorithm
> > >>>>>
> > >>>>>              FYI- First published July 4, 2011
> > >>>>>
> > >>>>>
> > >>>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment
> > >>>>>-r
> > >>>>>out
> > >>>>>e
> > >>>>>/
> > >>>>>
> > >>>>>              -Ali
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>> _______________________________________________
> > >>>>> BESS mailing list
> > >>>>> BESS@ietf.org
> > >>>>> https://www.ietf.org/mailman/listinfo/bess
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> >
> >>>>______________________________________________________
> > ______________
> > >>>>___
> > >>>>_
> > >>>>_
> > >>>>________________________________________________
> > >>>>
> > >>>> Ce message et ses pieces jointes peuvent contenir des
> > >>>>informations confidentielles ou privilegiees et ne doivent donc
> > >>>>pas etre diffuses, exploites ou copies sans autorisation. Si vous
> > >>>>avez recu ce message par erreur, veuillez le signaler  a
> > >>>>l'expediteur et le detruire ainsi que les pieces jointes. Les
> > >>>>messages electroniques etant susceptibles d'alteration,  Orange
> > >>>>decline toute responsabilite si ce message a ete altere, deforme ou
> falsifie.
> > >>>>Merci.
> > >>>>
> > >>>> This message and its attachments may contain confidential or
> > >>>>privileged information that may be protected by law;  they should
> > >>>>not be distributed, used or copied without authorisation.
> > >>>> If you have received this email in error, please notify the
> > >>>>sender and delete this message and its attachments.
> > >>>> As emails may be altered, Orange is not liable for messages that
> > >>>>have been modified, changed or falsified.
> > >>>> Thank you.
> > >>>>
> > >>>> _______________________________________________
> > >>>> BESS mailing list
> > >>>> BESS@ietf.org
> > >>>> https://www.ietf.org/mailman/listinfo/bess
> > >>>>
> > >>>
> > >>>
> >
> >>>_______________________________________________________
> > ______________
> > >>>___
> > >>>_
> > >>>_
> > >>>_______________________________________________
> > >>>
> > >>>Ce message et ses pieces jointes peuvent contenir des informations
> > >>>confidentielles ou privilegiees et ne doivent donc pas etre
> > >>>diffuses, exploites ou copies sans autorisation. Si vous avez recu
> > >>>ce message par erreur, veuillez le signaler a l'expediteur et le
> > >>>detruire ainsi que les pieces jointes. Les messages electroniques
> > >>>etant susceptibles d'alteration, Orange decline toute
> > >>>responsabilite si ce message a ete altere, deforme ou falsifie. Merc=
i.
> > >>>
> > >>>This message and its attachments may contain confidential or
> > >>>privileged information that may be protected by law; they should
> > >>>not be distributed, used or copied without authorisation.
> > >>>If you have received this email in error, please notify the sender
> > >>>and delete this message and its attachments.
> > >>>As emails may be altered, Orange is not liable for messages that
> > >>>have been modified, changed or falsified.
> > >>>Thank you.
> > >>>_______________________________________________
> > >>>BESS mailing list
> > >>>BESS@ietf.org
> > >>>https://www.ietf.org/mailman/listinfo/bess
> > >>_______________________________________________
> > >>BESS mailing list
> > >>BESS@ietf.org
> > >>https://www.ietf.org/mailman/listinfo/bess
> > >_______________________________________________
> > >BESS mailing list
> > >BESS@ietf.org
> > >https://www.ietf.org/mailman/listinfo/bess
> > _______________________________________________
> > BESS mailing list
> > BESS@ietf.org
> > https://www.ietf.org/mailman/listinfo/bess


From nobody Fri Mar 27 08:39:27 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F2E431A0282 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:39:25 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id T1A-eeww_nnB for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:39:21 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4B1151A0231 for <bess@ietf.org>; Fri, 27 Mar 2015 08:39:20 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id E6BD0184D8069; Fri, 27 Mar 2015 15:39:15 +0000 (GMT)
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2RFdIHJ012133 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 16:39:18 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR712WXCHHUB03.zeu.alcatel-lucent.com ([135.239.2.74]) with mapi id 14.03.0195.001; Fri, 27 Mar 2015 16:39:18 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Haoweiguo <haoweiguo@huawei.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8QfEaq8IGmBqka8wmUCt0qdaZ0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HOAABL1oIAADZ/DgAAMKZcAACYGUQA=
Date: Fri, 27 Mar 2015 15:39:17 +0000
Message-ID: <D13AC565.6C964%jorge.rabadan@alcatel-lucent.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com> <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.38]
Content-Type: text/plain; charset="utf-8"
Content-ID: <F2BB6265AE8576499EDB81453A9B820A@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/ZNV9r0FHnHpm7__UT6dZMcCjv8g>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 15:39:26 -0000
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From nobody Fri Mar 27 08:40:27 2015
Return-Path: <thomas.morin@orange.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1618F1AD0C5 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:40:25 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FREEMAIL_FROM=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001, UNPARSEABLE_RELAY=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6tEan3qNpIwG for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:40:18 -0700 (PDT)
Received: from relais-inet.francetelecom.com (relais-ias245.francetelecom.com [80.12.204.245]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E88271A0231 for <bess@ietf.org>; Fri, 27 Mar 2015 08:40:16 -0700 (PDT)
Received: from omfeda08.si.francetelecom.fr (unknown [xx.xx.xx.201]) by omfeda10.si.francetelecom.fr (ESMTP service) with ESMTP id AEB5C374322 for <bess@ietf.org>; Fri, 27 Mar 2015 16:40:13 +0100 (CET)
Received: from Exchangemail-eme1.itn.ftgroup (unknown [10.114.1.183]) by omfeda08.si.francetelecom.fr (ESMTP service) with ESMTP id 8FD8038404E for <bess@ietf.org>; Fri, 27 Mar 2015 16:40:13 +0100 (CET)
Received: from [10.193.116.37] (10.197.38.6) by PEXCVZYH02.corporate.adroot.infra.ftgroup (10.114.1.183) with Microsoft SMTP Server (TLS) id 14.3.224.2; Fri, 27 Mar 2015 16:40:12 +0100
Message-ID: <21658_1427470813_551579DD_21658_1989_1_551579DA.7060102@orange.com>
Date: Fri, 27 Mar 2015 10:40:10 -0500
From: <thomas.morin@orange.com>
Organization: Orange
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: <bess@ietf.org>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>, <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562504AAF732AF476E5C218C7090@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8577BC@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F8577BC@nkgeml501-mbs.china.huawei.com>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit
X-Originating-IP: [10.197.38.6]
X-PMX-Version: 6.2.1.2478543, Antispam-Engine: 2.7.2.2107409, Antispam-Data: 2015.2.12.3031
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/TzH3kNJiczR4j_IX03uYYiIaQyU>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 15:40:25 -0000

HI Weigo, all,

Let me try to help again...

RFC7432, section 8.5 says:

    If a bridged network is multihomed to more than one PE in an EVPN
    network via switches, then the support of All-Active redundancy mode
    requires the bridged network to be connected to two or more PEs using
    a LAG.

    If a bridged network does not connect to the PEs using a LAG, then
    only one of the links between the bridged network and the PEs must be
    the active link for a given <ES, VLAN> or <ES, VLAN bundle>.  In this
    case, the set of Ethernet A-D per ES routes advertised by each PE
    MUST have the "Single-Active" bit in the flags of the ESI Label
    extended community set to 1.

My understanding was that the above excludes the kind of topology in 
which you exclude a possible loop.

-Thomas



Weiguo :
> Hi John,
> If you can tolerate indefinite forwarding loop during transiet time, i have no problem. If the community think it is a serious problem, i think we must propose an applicable solution for it, maybe there are other solutions better than handshaking mechanism.
>
> Thanks,
> weiguo
>
> ________________________________________
> From: John E Drake [jdrake@juniper.net]
> Sent: Friday, March 27, 2015 23:22
> To: Haoweiguo; Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT - thomas.morin@orange.com; bess@ietf.org
> Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
>
> I think we have beaten this topic to death.  For the numerous reason I've given, I'm not interested in you proposal.
>
> Yours Irrespectively,
>
> John
>
>> -----Original Message-----
>> From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
>> Sent: Friday, March 27, 2015 11:12 AM
>> To: Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT -
>> thomas.morin@orange.com; bess@ietf.org
>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>> algorithm
>>
>> Hi Jorge,
>> Let give you another indefinitely forwarding loop case, the picture is as
>> follows:
>>
>> CE3------------CE4
>>    |                        |
>> PE3                   PE4
>>
>>
>>
>> PE1                   PE2
>>   |                         |
>> CE1------------CE2
>>
>> If PE1 and PE2 are dual DF PE in transiet time, assuing PE3 and PE4 are in
>> stable state ,PE3 is DF PE, PE4 is non-DF PE, the BUM traffic from CE3 to CE1,
>> the forwarding loop path is as follows;
>> CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->PE4----
>>> forever......
>> Do you think the harm is enough?
>> Thanks,
>> weiguo
>> ________________________________________
>> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>> Sent: Friday, March 27, 2015 22:59
>> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>> bess@ietf.org
>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>> algorithm
>>
>> Hi Weiguo,
>>
>> OK, thanks for clarifying, that is the part I did not understand in your loop :-)
>>
>> To me a loop means you send a frame and that frame keeps being forwarded
>> by the same nodes indefinitely till you take a corrective action.
>> In this case, CE3 might get a few packets during the transient state, but those
>> packets will have its own MAC SA so CE3 will discard them. I don't think that is
>> harmful for CE3 in most of the cases. And if it is, you just need to make sure
>> the timers are long enough and the DF election properly implemented to
>> avoid the transient DF-DF state.
>>
>> The handshaking is definitely overkilling IMHO.
>>
>> Thanks.
>> Jorge
>>
>> -----Original Message-----
>> From: Haoweiguo <haoweiguo@huawei.com>
>> Date: Friday, March 27, 2015 at 7:29 AM
>> To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
>> (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
>> <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>> algorithm
>>
>>> Hi Jorge,
>>> Sorry, maybe i have not explain clearly, pls allow me to clarify it
>>> more clearly.
>>> In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2
>>> is single homed to remote PE3. CE1 and CE2 forms one layer 2 network,
>>> so it multi-homed network scenario.
>>>            CE3
>>>              |
>>>             PE3
>>>
>>>
>>> PE1                  PE2
>>>   |                       |
>>> CE1-----------CE2
>>>
>>> In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends
>>> BUM traffic to CE1, the traffic will reach PE1 and PE2, CE1 will
>>> receive two copy of the traffic from CE3, also loop is formed, the form path:
>>>
>>> CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
>>>
>>> I think the traffic loop is absolutely not allowed. In MHN case, the
>>> loop will formed due to dual-DF PE in transiet period.
>>>
>>> Thanks,
>>> weiguo
>>>
>>>
>>>
>>> ________________________________________
>>> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>>> Sent: Friday, March 27, 2015 20:26
>>> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>>> bess@ietf.org
>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>> algorithm
>>>
>>> Hi Weiguo,
>>>
>>> About this:
>>>
>>> [weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>> case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
>>> and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>>> loop back to CE1 through PE2. If CE1 can filter and drop the traffic ,
>>> it is fortunate, maybe no loop.
>>> For multi-homed network scenario, say local network1(CE1 and CE2) is
>>> multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>> back through PE1-->PE2-->CE2, it is very serious. It is absolutely not allowed.
>>>
>>>
>>> [JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2
>>> will not send the traffic back to CE1 due to the ESI-label and
>>> split-horizon procedures.
>>> In MH network case PE2 will NOT send the traffic to CE2 for the same
>>> reason.
>>>
>>> Let me know if you disagree please.
>>>
>>> Thanks.
>>> Jorge
>>>
>>> -----Original Message-----
>>> From: Haoweiguo <haoweiguo@huawei.com>
>>> Date: Thursday, March 26, 2015 at 9:55 PM
>>> To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
>>> "thomas.morin@orange.com" <thomas.morin@orange.com>,
>> "bess@ietf.org"
>>> <bess@ietf.org>
>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>> algorithm
>>>
>>>> Hi Satya,
>>>> Pls see inline.
>>>>
>>>> Thanks,
>>>> weiguo
>>>>
>>>> ________________________________________
>>>> From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
>>>> Sent: Friday, March 27, 2015 12:11
>>>> To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>> algorithm
>>>>
>>>> Thomas,
>>>>
>>>> Thanks for your summary. It captures the essence of the discussion.
>>>> [weiguo]: Yes, i agree.
>>>>
>>>> One feature of the HRW scheme is that it computes the Active and the
>>>> Backup DF upfront.
>>>> Whenever the Active PE, say PE1, dies; the Backup PE, PE2, can become
>>>> the Active immediately upon receiving the withdraw of the ES route
>>> >from PE1.
>>>> It need not even have to wait for the 3 seconds time.
>>>> [weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
>>>> message:) But when PE1 leaves ESI, it can withdraw the ES route
>>>> immediately.
>>>> When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2
>>>> will both act as PE for some transiet time.
>>>>
>>>> On the other hand, the 3 second timer will serve as a bulk and batch
>>>> mechanism for the case when new PE(s) is(are) coming up, so we don¹t
>>>> have to do a DF election ever so often.
>>>>
>>>> Haoweiguo,  I think the "loop" scenario is speculative.
>>>> I think everybody understands that, yes, there may be some transient
>>>> duplication of bum traffic; however, there is no "loop" problem to be
>>>> solved.
>>>> [weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>>> case. In multi-homed device case, say CE1 is multi-homed to PE1 and
>>>> PE2 and both PE are DF PE. When CE1 sends traffic to PE1, the traffic
>>>> will loop back to CE1 through PE2. If CE1 can filter and drop the
>>>> traffic , it is fortunate, maybe no loop.
>>>> For multi-homed network scenario, say local network1(CE1 and CE2) is
>>>> multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>>> back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
>>>> allowed.
>>>>
>>>> Thanks,
>>>> --Satya
>>>>
>>>>
>>>>
>>>> On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
>>>>
>>>>> Hi Thomas,
>>>>> I know you are neutral, sorry for my reply vulnerable to be
>>>>> miunderstood.
>>>>>
>>>>> Thanks,
>>>>> weiguo
>>>>>
>>>>> ________________________________________
>>>>> From: thomas.morin@orange.com [thomas.morin@orange.com]
>>>>> Sent: Friday, March 27, 2015 11:59
>>>>> To: Haoweiguo; bess@ietf.org
>>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on
>>>>> Paxos algorithm
>>>>>
>>>>> Hi Weiguo,
>>>>>
>>>>> 2015-03-26, Haoweiguo:
>>>>>> Thomas:
>>>>>>> - RFC7432 may have transient periods where the DF election state
>>>>>>> is not yet synchronized between the two peers:
>>>>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic
>>>>>> loop
>>>>> Note well that I didn't write that there can be transient _loops_ .
>>>>> I tried to capture the exchange you had all, and what I gather is
>>>>> that the split-horizon procedure _prevents_loops_, independently of
>>>>> any transient period where DF state is not synchronized and which may
>>>>> lead to transient _duplicates_.
>>>>>
>>>>> -Thomas
>>>>>
>>>>>> 26/03/2015 20:05, John E Drake :
>>>>>>> Weiguo,
>>>>>>>
>>>>>>> I guess I wasn¹t clear.  I think you draft, for the reasons I have
>>>>>>> detailed, is a non-solution to a non-problem with tremendous
>>>>>>> control plane cost.
>>>>>>>
>>>>>>> Yours Irrespectively,
>>>>>>>
>>>>>>> John
>>>>>>>
>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> Pls see below.
>>>>>>>
>>>>>>> Thanks,
>>>>>>>
>>>>>>> weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>> *Sent:* Friday, March 27, 2015 6:00
>>>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> To recap,
>>>>>>>
>>>>>>> We have established that your proposal is untenable because of its
>>>>>>> control plane load.
>>>>>>>
>>>>>>> We have established that your proposal is based upon a flawed
>>>>>>> understanding of the DF election in RFC 7432.
>>>>>>>
>>>>>>> [weiguo]: In ethernet world, traffic loop is serious than short
>>>>>>> timer traffic disruption. If you want to implement  transiet
>>>>>>> traffic loop process, i will modify my draft to solve your issue.
>>>>>>>
>>>>>>> If i am the developer, i will prefer short timer traffic
>>>>>>> disruption based on current EVPN protocol.
>>>>>>>
>>>>>>> What you are now arguing is that your draft prevents two or more
>>>>>>> PEs from being DF simultaneously. This is clearly nonsense.
>>>>>>>
>>>>>>> [weiguo]: I will modify the draft problem statements, and use the
>>>>>>> same  handshaking solution to solve it.
>>>>>>>
>>>>>>> Furthermore, we have established that having two or more DFs for
>>>>>>> what even you admit is a brief transient leads to duplicate
>>>>>>> traffic, which is acceptable, but not loops, your assertion to the
>> contrary.
>>>>>>> [weiguo]: It is transient loop and traffic duplication issue.
>>>>>>>
>>>>>>> Yours Irrespectively,
>>>>>>>
>>>>>>> John
>>>>>>>
>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> John,
>>>>>>>
>>>>>>> As your understanding of the EVPN draft, the DF election mechanism
>>>>>>> has  more serious side effect, it will have short time traffic
>>>>>>> loop,i.e.,  dual DF PEs will exist for a short time. I think dual
>>>>>>> DF PEs is  absolutely not tolerated, because native ethernet header
>>>>>>> has no TTL,  up to several hundred ms traffic loop normally not
>>>>>>> tolerated in  commertial networks.
>>>>>>>
>>>>>>> As your understanding, the PEs should do as following:
>>>>>>>
>>>>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but
>>>>>>> have rarely deployment.
>>>>>>>
>>>>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3
>>>>>>> joins ESI and trigger DF re-election. When reception timer expires:
>>>>>>>
>>>>>>> PE1 upgrades to DF PE.
>>>>>>>
>>>>>>> After reception timer+ ES route transmission timer:
>>>>>>>
>>>>>>> PE2 downloads to non-DF PE.
>>>>>>>
>>>>>>> So in timer clock sync case, dual DF PEs will exist at least
>>>>>>> transmission timer.
>>>>>>>
>>>>>>> If NTP is used for timer sync, because it has bad accuracy, dual
>>>>>>> DF PEs will exist more longer timer.
>>>>>>>
>>>>>>> So as your understanding for DF election, the drawback is more clear.
>>>>>>>
>>>>>>> Thanks,
>>>>>>>
>>>>>>> weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>> *Sent:* Friday, March 27, 2015 4:41
>>>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> Weiguo,
>>>>>>>
>>>>>>> We have already established that your proposal is untenable
>>>>>>> because of  its control plane load.
>>>>>>>
>>>>>>> What we are now discussing is that your proposal is based upon a
>>>>>>> misunderstanding of the algorithm in RFC 7432.  You are assuming
>>>>>>> that
>>>>>>> PE1 will advertise an ES route and then wait for the configured
>>>>>>> interval before performing the DF election while PE2 and PE3 will
>>>>>>> perform the DF election as soon as they receive the ES route from
>>>>>>> PE1.
>>>>>>> This is not what RFC 7432 says.
>>>>>>>
>>>>>>> Rather, what is says is that the advertisement of the ES route by
>>>>>>> PE1 and its receipt by PE2 and PE3 causes all three PEs to start
>>>>>>> the configured interval timer  - ³3. When the timer expires, each
>>>>>>> PE builds an ordered list of the IP addresses of all the PE nodes
>>>>>>> connected to the Ethernet segment (including itself), in
>>>>>>> increasing numeric value.²
>>>>>>>
>>>>>>> Yours Irrespectively,
>>>>>>>
>>>>>>> John
>>>>>>>
>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> Pls read my detail replies to Satya. If you still can't catch it,
>>>>>>> pls read my draft and EVPN base protocol,  thanks
>>>>>>>
>>>>>>> weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>> *Sent:* Friday, March 27, 2015 1:28
>>>>>>> *To:* Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>>> <mailto:bess@ietf.org>
>>>>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> I think Patrice is correct.  Your proposal doesn't solve the
>>>>>>> problem and it does so at huge cost.
>>>>>>>
>>>>>>> Sent from my iPhone
>>>>>>>
>>>>>>>
>>>>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>>>>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>>>>>>
>>>>>>>       Weiguo,
>>>>>>>
>>>>>>>       I¹m not sure I¹m following here.
>>>>>>>
>>>>>>>       Don¹t you have the same issue with your handshaking mechanism?
>>>>>>>
>>>>>>>       If you don¹t know your peer, how can you handshake?
>>>>>>>
>>>>>>>       Regards,
>>>>>>>
>>>>>>>       Patrice
>>>>>>>
>>>>>>>       Image removed by sender.
>>>>>>>
>>>>>>>       *Patrice Brissette*
>>>>>>>       TECHNICAL LEADER.ENGINEERING
>>>>>>>
>>>>>>>       pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>>>       Phone: *+1 613 254 3336*
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>       *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>>>       Canada
>>>>>>>       Cisco.com <http://www.cisco.com/global/CA/>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>       Image removed by sender.Think before you print.
>>>>>>>
>>>>>>>       This email may contain confidential and privileged material for
>>>>>>>       the sole use of the intended recipient. Any review, use,
>>>>>>>       distribution or disclosure by others is strictly prohibited. If
>>>>>>>       you are not the intended recipient (or authorized to receive for
>>>>>>>       the recipient), please contact the sender by reply email and
>>>>>>>       delete all copies of this message.
>>>>>>>
>>>>>>>       Please click here
>>>>>>>
>>>>>>> <http://www.cisco.com/web/about/doing_business/legal/cri/index.h
>> tml>
>>>>>>>       for Company Registration Information.
>>>>>>>
>>>>>>>
>>>>>>>       *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>>> <mailto:haoweiguo@huawei.com>>
>>>>>>>       *Date: *Thursday, March 26, 2015 at 10:24 AM
>>>>>>>       *To: *Patrice Brissette <pbrisset@cisco.com
>>>>>>>       <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>>>>>>       <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>>>>       <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>>>       <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>>>>>       *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES
>>>>>>> based on
>>>>>>>       Paxos algorithm
>>>>>>>
>>>>>>>           Hi Patrice,
>>>>>>>
>>>>>>>           Up to reception timer traffic disruption in transient
>>>>>>> phase is
>>>>>>>           one of the issues.
>>>>>>>
>>>>>>>           Thanks,
>>>>>>>
>>>>>>>           weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>>           *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>>>>>>           <mailto:pbrisset@cisco.com>]
>>>>>>>           *Sent:* Thursday, March 26, 2015 20:54
>>>>>>>           *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>>>>>>           bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>           *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>>>           based on Paxos algorithm
>>>>>>>
>>>>>>>           Weiguo,
>>>>>>>
>>>>>>>           You mention "But if your draft have not solved all
>>>>>>> issues²,
>>>>>>>
>>>>>>>           Can you explain what Satya¹s draft is not solving?
>>>>>>>
>>>>>>>           Regards,
>>>>>>>
>>>>>>>           Patrice
>>>>>>>
>>>>>>>           Image removed by sender.
>>>>>>>
>>>>>>>           *Patrice Brissette*
>>>>>>>           TECHNICAL LEADER.ENGINEERING
>>>>>>>
>>>>>>>           pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>>>           Phone: *+1 613 254 3336*
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>           *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>>>           Canada
>>>>>>>           Cisco.com <http://www.cisco.com/global/CA/>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>           Image removed by sender.Think before you print.
>>>>>>>
>>>>>>>           This email may contain confidential and privileged material
>>>>>>>           for the sole use of the intended recipient. Any review, use,
>>>>>>>           distribution or disclosure by others is strictly prohibited.
>>>>>>>           If you are not the intended recipient (or authorized to
>>>>>>>           receive for the recipient), please contact the sender by
>>>>>>> reply
>>>>>>>           email and delete all copies of this message.
>>>>>>>
>>>>>>>           Please click here
>>>>>>>
>>>>>>> <http://www.cisco.com/web/about/doing_business/legal/cri/index.h
>> tml>
>>>>>>>           for Company Registration Information.
>>>>>>>
>>>>>>>
>>>>>>>           *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>>>           <mailto:haoweiguo@huawei.com>>
>>>>>>>           *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>>>>>>           *To: *John E Drake <jdrake@juniper.net
>>>>>>>           <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>>>>           <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>>>           <mailto:bess@ietf.org>" <bess@ietf.org
>>>>>>> <mailto:bess@ietf.org>>
>>>>>>>           *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>>>           based on Paxos algorithm
>>>>>>>
>>>>>>>               Hi John,
>>>>>>>
>>>>>>>                Firstly i think EVPN community should reach
>>>>>>> consensus on
>>>>>>>               the issues of current DF election mechanism. All these
>>>>>>>               issues should be resolved in a single new DF election
>>>>>>>               draft,rather than in multiple separate drafts. If your
>>>>>>>               draft can solve all these issues and stable, i have no
>>>>>>>               question for its progressing. But if your draft have not
>>>>>>>               solved all issues, i think it had better combine with
>>>>>>>               other drafts to provide a comprehensive solution. I
>>>>>>> think
>>>>>>>               the issues listed in draft-rabadan-bess-evpn-ac-df-01
>>>>>>> and
>>>>>>>               draft-hao-bess-evpn-df-handshaking-00 is valid, it
>>>>>>> should
>>>>>>>               be resolved. So i think although your new Hash algorithm
>>>>>>>               for DF election is good, it only includes partial
>>>>>>>               enhancements, maybe it still needs some time for
>>>>>>> consensus.
>>>>>>>
>>>>>>>               Thanks,
>>>>>>>
>>>>>>>               weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>>               *From:*John E Drake [jdrake@juniper.net
>>>>>>>               <mailto:jdrake@juniper.net>]
>>>>>>>               *Sent:* Thursday, March 26, 2015 7:16
>>>>>>>               *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>               <mailto:bess@ietf.org>
>>>>>>>               *Subject:* RE: [bess] Handshaking among PEs in an
>>>>>>> EVPN ES
>>>>>>>               based on Paxos algorithm
>>>>>>>
>>>>>>>               Weiguo,
>>>>>>>
>>>>>>>               Your proposal introduces a control plane processing load
>>>>>>>               that is O(#EVIs * PEs) per DF election and given that
>>>>>>>               there can be 4K EVIs per ES, this looks like a
>>>>>>>               **substantial** load. Furthermore,
>>>>>>>
>>>>>>>               you can¹t  use the ES route to co-ordinate DF election
>>>>>>>               because you would need to carry your new extended
>>>>>>>               community for each EVI and they would not all fit.  You
>>>>>>>               also can¹t use the Per EVI Ethernet AD route because
>>>>>>> that
>>>>>>>               is processed by all PEs in the EVI.
>>>>>>>
>>>>>>>               I think that from a practical perspective the new DF
>>>>>>>               election proposed in Satya¹s draft is sufficiently
>>>>>>> stable
>>>>>>>               that it renders your draft moot, even if it could be
>>>>>>> made
>>>>>>>               to work.
>>>>>>>
>>>>>>>               Yours Irrespectively,
>>>>>>>
>>>>>>>               John
>>>>>>>
>>>>>>>               *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf Of
>>>>>>>               *Haoweiguo
>>>>>>>               *Sent:* Wednesday, March 25, 2015 5:34 PM
>>>>>>>               *To:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>               <mailto:bess@ietf.org>
>>>>>>>               *Subject:* Re: [bess] Handshaking among PEs in an
>>>>>>> EVPN ES
>>>>>>>               based on Paxos algorithm
>>>>>>>
>>>>>>>               Hi Ali,
>>>>>>>
>>>>>>>               Thanks for your information. I scanned through this
>>>>>>> draft,
>>>>>>>               it really introduces inter-chassis message for DF
>>>>>>> election
>>>>>>>               handshaking, the requirements in this draft is to
>>>>>>>               eliminate transiet Loop and traffic duplication. Current
>>>>>>>               EVPN DF election mechanism already eliminated loop and
>>>>>>>               traffic duplication by configuring long reception
>>>>>>> timer on
>>>>>>>               each multi-homed PE, but up to reception timer traffic
>>>>>>>               disruption issue still exist. EVPN for DCI is an
>>>>>>> important
>>>>>>>               use case for EVPN, up to reception timer traffic
>>>>>>>               disruption can't be tolerated for service providers, it
>>>>>>>               should be improved.
>>>>>>>
>>>>>>>               Also for accuracy, i think handshaking state machine on
>>>>>>>               each multi-homed PE is also needed. From solution
>>>>>>>               perspective, in my draft, no inter-chassis message is
>>>>>>>               introduced, only one new extended community is
>>>>>>> introduced,
>>>>>>>               i think the process is comparatively simple than your
>>>>>>>               following draft.
>>>>>>>
>>>>>>>               Current EVPN DF election has some drawbacks, so there
>>>>>>> are
>>>>>>>               three new drafts about DF election emerged. I think BESS
>>>>>>>               WG can consider these three drafts in global view, a
>>>>>>>               single,comprehensive new DF election draft is hoped.
>>>>>>>
>>>>>>>               thanks.
>>>>>>>
>>>>>>>               Thanks,
>>>>>>>
>>>>>>>               weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>>               *From:*BESS [bess-bounces@ietf.org
>>>>>>>               <mailto:bess-bounces@ietf.org>] on behalf of Ali Sajassi
>>>>>>>               (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.com>]
>>>>>>>               *Sent:* Thursday, March 26, 2015 0:21
>>>>>>>               *To:* bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>               *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>>>>>>               based on Paxos algorithm
>>>>>>>
>>>>>>>               FYI- First published July 4, 2011
>>>>>>>
>>>>>>>
>>>>>>> https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-r
>>>>>>> out
>>>>>>> e
>>>>>>> /
>>>>>>>
>>>>>>>               -Ali
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> BESS mailing list
>>>>>>> BESS@ietf.org
>>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> ______________________________________________________
>> ______________
>>>>>> ___
>>>>>> _
>>>>>> _
>>>>>> ________________________________________________
>>>>>>
>>>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>>>> confidentielles ou privilegiees et ne doivent donc  pas etre
>>>>>> diffuses, exploites ou copies sans autorisation. Si vous avez recu
>>>>>> ce message par erreur, veuillez le signaler  a l'expediteur et le
>>>>>> detruire ainsi que les pieces jointes. Les messages electroniques
>>>>>> etant susceptibles d'alteration,  Orange decline toute
>>>>>> responsabilite si ce message a ete altere, deforme ou falsifie.
>>>>>> Merci.
>>>>>>
>>>>>> This message and its attachments may contain confidential or
>>>>>> privileged information that may be protected by law;  they should
>>>>>> not be distributed, used or copied without authorisation.
>>>>>> If you have received this email in error, please notify the sender
>>>>>> and delete this message and its attachments.
>>>>>> As emails may be altered, Orange is not liable for messages that
>>>>>> have been modified, changed or falsified.
>>>>>> Thank you.
>>>>>>
>>>>>> _______________________________________________
>>>>>> BESS mailing list
>>>>>> BESS@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>>>
>>>>>
>>>>> _______________________________________________________
>> ______________
>>>>> ___
>>>>> _
>>>>> _
>>>>> _______________________________________________
>>>>>
>>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>>> confidentielles ou privilegiees et ne doivent donc pas etre diffuses,
>>>>> exploites ou copies sans autorisation. Si vous avez recu ce message
>>>>> par erreur, veuillez le signaler a l'expediteur et le detruire ainsi
>>>>> que les pieces jointes. Les messages electroniques etant susceptibles
>>>>> d'alteration, Orange decline toute responsabilite si ce message a ete
>>>>> altere, deforme ou falsifie. Merci.
>>>>>
>>>>> This message and its attachments may contain confidential or
>>>>> privileged information that may be protected by law; they should not
>>>>> be distributed, used or copied without authorisation.
>>>>> If you have received this email in error, please notify the sender
>>>>> and delete this message and its attachments.
>>>>> As emails may be altered, Orange is not liable for messages that have
>>>>> been modified, changed or falsified.
>>>>> Thank you.


_________________________________________________________________________________________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration,
Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci.

This message and its attachments may contain confidential or privileged information that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and delete this message and its attachments.
As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified.
Thank you.


From nobody Fri Mar 27 08:45:06 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5DCAC1ACE89 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:45:04 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id lXdNA9PAZv_d for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:44:57 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpgre-esg-01.alcatel-lucent.com [135.245.210.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C99A01ACD9E for <bess@ietf.org>; Fri, 27 Mar 2015 08:44:55 -0700 (PDT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (unknown [135.239.2.122]) by Websense Email Security Gateway with ESMTPS id 77CEF8C6532EC; Fri, 27 Mar 2015 15:44:51 +0000 (GMT)
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id t2RFirfV016413 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 16:44:54 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR712WXCHHUB03.zeu.alcatel-lucent.com ([135.239.2.74]) with mapi id 14.03.0195.001; Fri, 27 Mar 2015 16:44:54 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8QfEaq8IGmBqka8wmUCt0qdaZ0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HOAABL1oIAADZ/DgAAMKZcAABfyUwAAL7TQgABe+gQAAMx1jYA=
Date: Fri, 27 Mar 2015 15:44:53 +0000
Message-ID: <D13AC86D.6C986%jorge.rabadan@alcatel-lucent.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com> <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562504AAF732AF476E5C218C7090@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8577BC@nkgeml501-mbs.china.huawei.com> <21658_1427470813_551579DD_21658_1989_1_551579DA.7060102@orange.com>
In-Reply-To: <21658_1427470813_551579DD_21658_1989_1_551579DA.7060102@orange.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.40]
Content-Type: text/plain; charset="utf-8"
Content-ID: <7D3EA419A4D53B4EA879C89C377556DC@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/VwO0XwNuVN4PXpauy0XgO_7nHlY>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 15:45:04 -0000

VGhhbmtzIGZvciB0aGUgcmVmZXJlbmNlIFRob21hcy4NCkFzIEkgd2FzIHNheWluZyBpbiBteSBl
bWFpbCwgdGhpcyBwcmV2ZW50cyB0aGUgbG9vcCBpbiBXZWlndW/igJlzIGRpYWdyYW0sDQpzaW5j
ZSBvbmx5IHNpbmdsZS1hY3RpdmUgaXMgc3VwcG9ydGVkIGluIFdlaWd1b+KAmXMgZXhhbXBsZS4N
Cg0KVGhhbmsgeW91Lg0KSm9yZ2UNCg0KDQotLS0tLU9yaWdpbmFsIE1lc3NhZ2UtLS0tLQ0KRnJv
bTogInRob21hcy5tb3JpbkBvcmFuZ2UuY29tIiA8dGhvbWFzLm1vcmluQG9yYW5nZS5jb20+DQpP
cmdhbml6YXRpb246IE9yYW5nZQ0KRGF0ZTogRnJpZGF5LCBNYXJjaCAyNywgMjAxNSBhdCA4OjQw
IEFNDQpUbzogImJlc3NAaWV0Zi5vcmciIDxiZXNzQGlldGYub3JnPg0KU3ViamVjdDogUmU6IFti
ZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQRXMgaW4gYW4gRVZQTiBFUyBiYXNlZCBvbiBQYXhvcw0K
YWxnb3JpdGhtDQoNCj5ISSBXZWlnbywgYWxsLA0KPg0KPkxldCBtZSB0cnkgdG8gaGVscCBhZ2Fp
bi4uLg0KPg0KPlJGQzc0MzIsIHNlY3Rpb24gOC41IHNheXM6DQo+DQo+ICAgIElmIGEgYnJpZGdl
ZCBuZXR3b3JrIGlzIG11bHRpaG9tZWQgdG8gbW9yZSB0aGFuIG9uZSBQRSBpbiBhbiBFVlBODQo+
ICAgIG5ldHdvcmsgdmlhIHN3aXRjaGVzLCB0aGVuIHRoZSBzdXBwb3J0IG9mIEFsbC1BY3RpdmUg
cmVkdW5kYW5jeSBtb2RlDQo+ICAgIHJlcXVpcmVzIHRoZSBicmlkZ2VkIG5ldHdvcmsgdG8gYmUg
Y29ubmVjdGVkIHRvIHR3byBvciBtb3JlIFBFcyB1c2luZw0KPiAgICBhIExBRy4NCj4NCj4gICAg
SWYgYSBicmlkZ2VkIG5ldHdvcmsgZG9lcyBub3QgY29ubmVjdCB0byB0aGUgUEVzIHVzaW5nIGEg
TEFHLCB0aGVuDQo+ICAgIG9ubHkgb25lIG9mIHRoZSBsaW5rcyBiZXR3ZWVuIHRoZSBicmlkZ2Vk
IG5ldHdvcmsgYW5kIHRoZSBQRXMgbXVzdCBiZQ0KPiAgICB0aGUgYWN0aXZlIGxpbmsgZm9yIGEg
Z2l2ZW4gPEVTLCBWTEFOPiBvciA8RVMsIFZMQU4gYnVuZGxlPi4gIEluIHRoaXMNCj4gICAgY2Fz
ZSwgdGhlIHNldCBvZiBFdGhlcm5ldCBBLUQgcGVyIEVTIHJvdXRlcyBhZHZlcnRpc2VkIGJ5IGVh
Y2ggUEUNCj4gICAgTVVTVCBoYXZlIHRoZSAiU2luZ2xlLUFjdGl2ZSIgYml0IGluIHRoZSBmbGFn
cyBvZiB0aGUgRVNJIExhYmVsDQo+ICAgIGV4dGVuZGVkIGNvbW11bml0eSBzZXQgdG8gMS4NCj4N
Cj5NeSB1bmRlcnN0YW5kaW5nIHdhcyB0aGF0IHRoZSBhYm92ZSBleGNsdWRlcyB0aGUga2luZCBv
ZiB0b3BvbG9neSBpbg0KPndoaWNoIHlvdSBleGNsdWRlIGEgcG9zc2libGUgbG9vcC4NCj4NCj4t
VGhvbWFzDQo+DQo+DQo+DQo+V2VpZ3VvIDoNCj4+IEhpIEpvaG4sDQo+PiBJZiB5b3UgY2FuIHRv
bGVyYXRlIGluZGVmaW5pdGUgZm9yd2FyZGluZyBsb29wIGR1cmluZyB0cmFuc2lldCB0aW1lLCBp
DQo+PmhhdmUgbm8gcHJvYmxlbS4gSWYgdGhlIGNvbW11bml0eSB0aGluayBpdCBpcyBhIHNlcmlv
dXMgcHJvYmxlbSwgaSB0aGluaw0KPj53ZSBtdXN0IHByb3Bvc2UgYW4gYXBwbGljYWJsZSBzb2x1
dGlvbiBmb3IgaXQsIG1heWJlIHRoZXJlIGFyZSBvdGhlcg0KPj5zb2x1dGlvbnMgYmV0dGVyIHRo
YW4gaGFuZHNoYWtpbmcgbWVjaGFuaXNtLg0KPj4NCj4+IFRoYW5rcywNCj4+IHdlaWd1bw0KPj4N
Cj4+IF9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18NCj4+IEZyb206IEpv
aG4gRSBEcmFrZSBbamRyYWtlQGp1bmlwZXIubmV0XQ0KPj4gU2VudDogRnJpZGF5LCBNYXJjaCAy
NywgMjAxNSAyMzoyMg0KPj4gVG86IEhhb3dlaWd1bzsgUmFiYWRhbiwgSm9yZ2UgKEpvcmdlKTsg
U2F0eWEgTW9oYW50eSAoc2F0eWFtb2gpOyBFWFQgLQ0KPj50aG9tYXMubW9yaW5Ab3JhbmdlLmNv
bTsgYmVzc0BpZXRmLm9yZw0KPj4gU3ViamVjdDogUkU6IFtiZXNzXSBIYW5kc2hha2luZyBhbW9u
ZyBQRXMgaW4gYW4gRVZQTiBFUyBiYXNlZCBvbiBQYXhvcw0KPj5hbGdvcml0aG0NCj4+DQo+PiBJ
IHRoaW5rIHdlIGhhdmUgYmVhdGVuIHRoaXMgdG9waWMgdG8gZGVhdGguICBGb3IgdGhlIG51bWVy
b3VzIHJlYXNvbg0KPj5JJ3ZlIGdpdmVuLCBJJ20gbm90IGludGVyZXN0ZWQgaW4geW91IHByb3Bv
c2FsLg0KPj4NCj4+IFlvdXJzIElycmVzcGVjdGl2ZWx5LA0KPj4NCj4+IEpvaG4NCj4+DQo+Pj4g
LS0tLS1PcmlnaW5hbCBNZXNzYWdlLS0tLS0NCj4+PiBGcm9tOiBCRVNTIFttYWlsdG86YmVzcy1i
b3VuY2VzQGlldGYub3JnXSBPbiBCZWhhbGYgT2YgSGFvd2VpZ3VvDQo+Pj4gU2VudDogRnJpZGF5
LCBNYXJjaCAyNywgMjAxNSAxMToxMiBBTQ0KPj4+IFRvOiBSYWJhZGFuLCBKb3JnZSAoSm9yZ2Up
OyBTYXR5YSBNb2hhbnR5IChzYXR5YW1vaCk7IEVYVCAtDQo+Pj4gdGhvbWFzLm1vcmluQG9yYW5n
ZS5jb207IGJlc3NAaWV0Zi5vcmcNCj4+PiBTdWJqZWN0OiBSZTogW2Jlc3NdIEhhbmRzaGFraW5n
IGFtb25nIFBFcyBpbiBhbiBFVlBOIEVTIGJhc2VkIG9uIFBheG9zDQo+Pj4gYWxnb3JpdGhtDQo+
Pj4NCj4+PiBIaSBKb3JnZSwNCj4+PiBMZXQgZ2l2ZSB5b3UgYW5vdGhlciBpbmRlZmluaXRlbHkg
Zm9yd2FyZGluZyBsb29wIGNhc2UsIHRoZSBwaWN0dXJlIGlzDQo+Pj5hcw0KPj4+IGZvbGxvd3M6
DQo+Pj4NCj4+PiBDRTMtLS0tLS0tLS0tLS1DRTQNCj4+PiAgICB8ICAgICAgICAgICAgICAgICAg
ICAgICAgfA0KPj4+IFBFMyAgICAgICAgICAgICAgICAgICBQRTQNCj4+Pg0KPj4+DQo+Pj4NCj4+
PiBQRTEgICAgICAgICAgICAgICAgICAgUEUyDQo+Pj4gICB8ICAgICAgICAgICAgICAgICAgICAg
ICAgIHwNCj4+PiBDRTEtLS0tLS0tLS0tLS1DRTINCj4+Pg0KPj4+IElmIFBFMSBhbmQgUEUyIGFy
ZSBkdWFsIERGIFBFIGluIHRyYW5zaWV0IHRpbWUsIGFzc3VpbmcgUEUzIGFuZCBQRTQNCj4+PmFy
ZSBpbg0KPj4+IHN0YWJsZSBzdGF0ZSAsUEUzIGlzIERGIFBFLCBQRTQgaXMgbm9uLURGIFBFLCB0
aGUgQlVNIHRyYWZmaWMgZnJvbSBDRTMNCj4+PnRvIENFMSwNCj4+PiB0aGUgZm9yd2FyZGluZyBs
b29wIHBhdGggaXMgYXMgZm9sbG93czsNCj4+PiANCj4+PkNFMy0tLT5QRTMtLS0tLS0+UEUyLS0t
PkNFMi0tLS0+Q0UxLS0tLT5QRTEtLS0tPlBFMy0tPkNFMy0tPkNFNC0tLT5QRTQtLQ0KPj4+LS0N
Cj4+Pj4gZm9yZXZlci4uLi4uLg0KPj4+IERvIHlvdSB0aGluayB0aGUgaGFybSBpcyBlbm91Z2g/
DQo+Pj4gVGhhbmtzLA0KPj4+IHdlaWd1bw0KPj4+IF9fX19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fX19fX19fX18NCj4+PiBGcm9tOiBSYWJhZGFuLCBKb3JnZSAoSm9yZ2UpIFtqb3JnZS5y
YWJhZGFuQGFsY2F0ZWwtbHVjZW50LmNvbV0NCj4+PiBTZW50OiBGcmlkYXksIE1hcmNoIDI3LCAy
MDE1IDIyOjU5DQo+Pj4gVG86IEhhb3dlaWd1bzsgU2F0eWEgTW9oYW50eSAoc2F0eWFtb2gpOyB0
aG9tYXMubW9yaW5Ab3JhbmdlLmNvbTsNCj4+PiBiZXNzQGlldGYub3JnDQo+Pj4gU3ViamVjdDog
UmU6IFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQRXMgaW4gYW4gRVZQTiBFUyBiYXNlZCBvbiBQ
YXhvcw0KPj4+IGFsZ29yaXRobQ0KPj4+DQo+Pj4gSGkgV2VpZ3VvLA0KPj4+DQo+Pj4gT0ssIHRo
YW5rcyBmb3IgY2xhcmlmeWluZywgdGhhdCBpcyB0aGUgcGFydCBJIGRpZCBub3QgdW5kZXJzdGFu
ZCBpbg0KPj4+eW91ciBsb29wIDotKQ0KPj4+DQo+Pj4gVG8gbWUgYSBsb29wIG1lYW5zIHlvdSBz
ZW5kIGEgZnJhbWUgYW5kIHRoYXQgZnJhbWUga2VlcHMgYmVpbmcNCj4+PmZvcndhcmRlZA0KPj4+
IGJ5IHRoZSBzYW1lIG5vZGVzIGluZGVmaW5pdGVseSB0aWxsIHlvdSB0YWtlIGEgY29ycmVjdGl2
ZSBhY3Rpb24uDQo+Pj4gSW4gdGhpcyBjYXNlLCBDRTMgbWlnaHQgZ2V0IGEgZmV3IHBhY2tldHMg
ZHVyaW5nIHRoZSB0cmFuc2llbnQgc3RhdGUsDQo+Pj5idXQgdGhvc2UNCj4+PiBwYWNrZXRzIHdp
bGwgaGF2ZSBpdHMgb3duIE1BQyBTQSBzbyBDRTMgd2lsbCBkaXNjYXJkIHRoZW0uIEkgZG9uJ3QN
Cj4+PnRoaW5rIHRoYXQgaXMNCj4+PiBoYXJtZnVsIGZvciBDRTMgaW4gbW9zdCBvZiB0aGUgY2Fz
ZXMuIEFuZCBpZiBpdCBpcywgeW91IGp1c3QgbmVlZCB0bw0KPj4+bWFrZSBzdXJlDQo+Pj4gdGhl
IHRpbWVycyBhcmUgbG9uZyBlbm91Z2ggYW5kIHRoZSBERiBlbGVjdGlvbiBwcm9wZXJseSBpbXBs
ZW1lbnRlZCB0bw0KPj4+IGF2b2lkIHRoZSB0cmFuc2llbnQgREYtREYgc3RhdGUuDQo+Pj4NCj4+
PiBUaGUgaGFuZHNoYWtpbmcgaXMgZGVmaW5pdGVseSBvdmVya2lsbGluZyBJTUhPLg0KPj4+DQo+
Pj4gVGhhbmtzLg0KPj4+IEpvcmdlDQo+Pj4NCj4+PiAtLS0tLU9yaWdpbmFsIE1lc3NhZ2UtLS0t
LQ0KPj4+IEZyb206IEhhb3dlaWd1byA8aGFvd2VpZ3VvQGh1YXdlaS5jb20+DQo+Pj4gRGF0ZTog
RnJpZGF5LCBNYXJjaCAyNywgMjAxNSBhdCA3OjI5IEFNDQo+Pj4gVG86IEpvcmdlIFJhYmFkYW4g
PGpvcmdlLnJhYmFkYW5AYWxjYXRlbC1sdWNlbnQuY29tPiwgIlNhdHlhIE1vaGFudHkNCj4+PiAo
c2F0eWFtb2gpIiA8c2F0eWFtb2hAY2lzY28uY29tPiwgInRob21hcy5tb3JpbkBvcmFuZ2UuY29t
Ig0KPj4+IDx0aG9tYXMubW9yaW5Ab3JhbmdlLmNvbT4sICJiZXNzQGlldGYub3JnIiA8YmVzc0Bp
ZXRmLm9yZz4NCj4+PiBTdWJqZWN0OiBSZTogW2Jlc3NdIEhhbmRzaGFraW5nIGFtb25nIFBFcyBp
biBhbiBFVlBOIEVTIGJhc2VkIG9uIFBheG9zDQo+Pj4gYWxnb3JpdGhtDQo+Pj4NCj4+Pj4gSGkg
Sm9yZ2UsDQo+Pj4+IFNvcnJ5LCBtYXliZSBpIGhhdmUgbm90IGV4cGxhaW4gY2xlYXJseSwgcGxz
IGFsbG93IG1lIHRvIGNsYXJpZnkgaXQNCj4+Pj4gbW9yZSBjbGVhcmx5Lg0KPj4+PiBJbiBNSE4g
Y2FzZSwgYXNzdW1pbmcgQ0UxIGFuZCBDRTIgYXJlIG11bHR0aS1ob21lZCB0byBQRTEgYW5kIFBF
MiwgQ0UyDQo+Pj4+IGlzIHNpbmdsZSBob21lZCB0byByZW1vdGUgUEUzLiBDRTEgYW5kIENFMiBm
b3JtcyBvbmUgbGF5ZXIgMiBuZXR3b3JrLA0KPj4+PiBzbyBpdCBtdWx0aS1ob21lZCBuZXR3b3Jr
IHNjZW5hcmlvLg0KPj4+PiAgICAgICAgICAgIENFMw0KPj4+PiAgICAgICAgICAgICAgfA0KPj4+
PiAgICAgICAgICAgICBQRTMNCj4+Pj4NCj4+Pj4NCj4+Pj4gUEUxICAgICAgICAgICAgICAgICAg
UEUyDQo+Pj4+ICAgfCAgICAgICAgICAgICAgICAgICAgICAgfA0KPj4+PiBDRTEtLS0tLS0tLS0t
LUNFMg0KPj4+Pg0KPj4+PiBJbiB0cmFuc2lldCBwZXJpb2QsIFBFMSBhbmQgUEUyIGFyZSBib3Ro
IERGIFBFLiBBdCB0aGlzIHRpbWUsIENFMw0KPj4+PnNlbmRzDQo+Pj4+IEJVTSB0cmFmZmljIHRv
IENFMSwgdGhlIHRyYWZmaWMgd2lsbCByZWFjaCBQRTEgYW5kIFBFMiwgQ0UxIHdpbGwNCj4+Pj4g
cmVjZWl2ZSB0d28gY29weSBvZiB0aGUgdHJhZmZpYyBmcm9tIENFMywgYWxzbyBsb29wIGlzIGZv
cm1lZCwgdGhlDQo+Pj4+Zm9ybSBwYXRoOg0KPj4+Pg0KPj4+PiBDRTMtLS0tPlBFMy0tLS0+UEUy
LS0tLT5DRTItLS0tPkNFMS0tLS0+UEUxLS0tPlBFMy0tLT5DRTMNCj4+Pj4NCj4+Pj4gSSB0aGlu
ayB0aGUgdHJhZmZpYyBsb29wIGlzIGFic29sdXRlbHkgbm90IGFsbG93ZWQuIEluIE1ITiBjYXNl
LCB0aGUNCj4+Pj4gbG9vcCB3aWxsIGZvcm1lZCBkdWUgdG8gZHVhbC1ERiBQRSBpbiB0cmFuc2ll
dCBwZXJpb2QuDQo+Pj4+DQo+Pj4+IFRoYW5rcywNCj4+Pj4gd2VpZ3VvDQo+Pj4+DQo+Pj4+DQo+
Pj4+DQo+Pj4+IF9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18NCj4+Pj4g
RnJvbTogUmFiYWRhbiwgSm9yZ2UgKEpvcmdlKSBbam9yZ2UucmFiYWRhbkBhbGNhdGVsLWx1Y2Vu
dC5jb21dDQo+Pj4+IFNlbnQ6IEZyaWRheSwgTWFyY2ggMjcsIDIwMTUgMjA6MjYNCj4+Pj4gVG86
IEhhb3dlaWd1bzsgU2F0eWEgTW9oYW50eSAoc2F0eWFtb2gpOyB0aG9tYXMubW9yaW5Ab3Jhbmdl
LmNvbTsNCj4+Pj4gYmVzc0BpZXRmLm9yZw0KPj4+PiBTdWJqZWN0OiBSZTogW2Jlc3NdIEhhbmRz
aGFraW5nIGFtb25nIFBFcyBpbiBhbiBFVlBOIEVTIGJhc2VkIG9uIFBheG9zDQo+Pj4+IGFsZ29y
aXRobQ0KPj4+Pg0KPj4+PiBIaSBXZWlndW8sDQo+Pj4+DQo+Pj4+IEFib3V0IHRoaXM6DQo+Pj4+
DQo+Pj4+IFt3ZWlndW9dOiBUcmFmZmljIGR1cGxpY2F0aW9uIGFuZCBsb29wIGlzc3VlIGJvdGgg
ZXhpc3QgaW4gZHVhbCBERiBQRQ0KPj4+PiBjYXNlLiBJbiBtdWx0aS1ob21lZCBkZXZpY2UgY2Fz
ZSwgc2F5IENFMSBpcyBtdWx0aS1ob21lZCB0byBQRTEgYW5kDQo+Pj4+UEUyDQo+Pj4+IGFuZCBi
b3RoIFBFIGFyZSBERiBQRS4gV2hlbiBDRTEgc2VuZHMgdHJhZmZpYyB0byBQRTEsIHRoZSB0cmFm
ZmljIHdpbGwNCj4+Pj4gbG9vcCBiYWNrIHRvIENFMSB0aHJvdWdoIFBFMi4gSWYgQ0UxIGNhbiBm
aWx0ZXIgYW5kIGRyb3AgdGhlIHRyYWZmaWMgLA0KPj4+PiBpdCBpcyBmb3J0dW5hdGUsIG1heWJl
IG5vIGxvb3AuDQo+Pj4+IEZvciBtdWx0aS1ob21lZCBuZXR3b3JrIHNjZW5hcmlvLCBzYXkgbG9j
YWwgbmV0d29yazEoQ0UxIGFuZCBDRTIpIGlzDQo+Pj4+IG11bHRpLWhvbWVkIHRvIFBFMSBhbmQg
UEUyLiBTaW1pbGFybHksIHRoZSB0cmFmZmljIGZyb20gQ0UxIHdpbGwgbG9vcA0KPj4+PiBiYWNr
IHRocm91Z2ggUEUxLS0+UEUyLS0+Q0UyLCBpdCBpcyB2ZXJ5IHNlcmlvdXMuIEl0IGlzIGFic29s
dXRlbHkNCj4+Pj5ub3QgYWxsb3dlZC4NCj4+Pj4NCj4+Pj4NCj4+Pj4gW0pPUkdFXSBJbiBhIE1I
IGRldmljZSBjYXNlOiBDRTEtPkJVTS0+UEUxLT4oRVNJLWxhYmVsK0JVTSktPlBFMi4gUEUyDQo+
Pj4+IHdpbGwgbm90IHNlbmQgdGhlIHRyYWZmaWMgYmFjayB0byBDRTEgZHVlIHRvIHRoZSBFU0kt
bGFiZWwgYW5kDQo+Pj4+IHNwbGl0LWhvcml6b24gcHJvY2VkdXJlcy4NCj4+Pj4gSW4gTUggbmV0
d29yayBjYXNlIFBFMiB3aWxsIE5PVCBzZW5kIHRoZSB0cmFmZmljIHRvIENFMiBmb3IgdGhlIHNh
bWUNCj4+Pj4gcmVhc29uLg0KPj4+Pg0KPj4+PiBMZXQgbWUga25vdyBpZiB5b3UgZGlzYWdyZWUg
cGxlYXNlLg0KPj4+Pg0KPj4+PiBUaGFua3MuDQo+Pj4+IEpvcmdlDQo+Pj4+DQo+Pj4+IC0tLS0t
T3JpZ2luYWwgTWVzc2FnZS0tLS0tDQo+Pj4+IEZyb206IEhhb3dlaWd1byA8aGFvd2VpZ3VvQGh1
YXdlaS5jb20+DQo+Pj4+IERhdGU6IFRodXJzZGF5LCBNYXJjaCAyNiwgMjAxNSBhdCA5OjU1IFBN
DQo+Pj4+IFRvOiAiU2F0eWEgTW9oYW50eSAoc2F0eWFtb2gpIiA8c2F0eWFtb2hAY2lzY28uY29t
PiwNCj4+Pj4gInRob21hcy5tb3JpbkBvcmFuZ2UuY29tIiA8dGhvbWFzLm1vcmluQG9yYW5nZS5j
b20+LA0KPj4+ICJiZXNzQGlldGYub3JnIg0KPj4+PiA8YmVzc0BpZXRmLm9yZz4NCj4+Pj4gU3Vi
amVjdDogUmU6IFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQRXMgaW4gYW4gRVZQTiBFUyBiYXNl
ZCBvbiBQYXhvcw0KPj4+PiBhbGdvcml0aG0NCj4+Pj4NCj4+Pj4+IEhpIFNhdHlhLA0KPj4+Pj4g
UGxzIHNlZSBpbmxpbmUuDQo+Pj4+Pg0KPj4+Pj4gVGhhbmtzLA0KPj4+Pj4gd2VpZ3VvDQo+Pj4+
Pg0KPj4+Pj4gX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXw0KPj4+Pj4g
RnJvbTogU2F0eWEgTW9oYW50eSAoc2F0eWFtb2gpIFtzYXR5YW1vaEBjaXNjby5jb21dDQo+Pj4+
PiBTZW50OiBGcmlkYXksIE1hcmNoIDI3LCAyMDE1IDEyOjExDQo+Pj4+PiBUbzogSGFvd2VpZ3Vv
OyB0aG9tYXMubW9yaW5Ab3JhbmdlLmNvbTsgYmVzc0BpZXRmLm9yZw0KPj4+Pj4gU3ViamVjdDog
UmU6IFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQRXMgaW4gYW4gRVZQTiBFUyBiYXNlZCBvbg0K
Pj4+Pj5QYXhvcw0KPj4+Pj4gYWxnb3JpdGhtDQo+Pj4+Pg0KPj4+Pj4gVGhvbWFzLA0KPj4+Pj4N
Cj4+Pj4+IFRoYW5rcyBmb3IgeW91ciBzdW1tYXJ5LiBJdCBjYXB0dXJlcyB0aGUgZXNzZW5jZSBv
ZiB0aGUgZGlzY3Vzc2lvbi4NCj4+Pj4+IFt3ZWlndW9dOiBZZXMsIGkgYWdyZWUuDQo+Pj4+Pg0K
Pj4+Pj4gT25lIGZlYXR1cmUgb2YgdGhlIEhSVyBzY2hlbWUgaXMgdGhhdCBpdCBjb21wdXRlcyB0
aGUgQWN0aXZlIGFuZCB0aGUNCj4+Pj4+IEJhY2t1cCBERiB1cGZyb250Lg0KPj4+Pj4gV2hlbmV2
ZXIgdGhlIEFjdGl2ZSBQRSwgc2F5IFBFMSwgZGllczsgdGhlIEJhY2t1cCBQRSwgUEUyLCBjYW4g
YmVjb21lDQo+Pj4+PiB0aGUgQWN0aXZlIGltbWVkaWF0ZWx5IHVwb24gcmVjZWl2aW5nIHRoZSB3
aXRoZHJhdyBvZiB0aGUgRVMgcm91dGUNCj4+Pj4gPmZyb20gUEUxLg0KPj4+Pj4gSXQgbmVlZCBu
b3QgZXZlbiBoYXZlIHRvIHdhaXQgZm9yIHRoZSAzIHNlY29uZHMgdGltZS4NCj4+Pj4+IFt3ZWln
dW9dOiBJIHRoaW5rIG1heWJlIHlvdXIgbmVnbGlnZW5jZSwgZGllcyBQRTEgY2FuJ3Qgc2VuZCB3
aXRoZHJhdw0KPj4+Pj4gbWVzc2FnZTopIEJ1dCB3aGVuIFBFMSBsZWF2ZXMgRVNJLCBpdCBjYW4g
d2l0aGRyYXcgdGhlIEVTIHJvdXRlDQo+Pj4+PiBpbW1lZGlhdGVseS4NCj4+Pj4+IFdoZW4gUEUx
IGRpZXMgYW5kIHJlY292ZXJzLCBhc3N1bWluZyBQRTEgaXMgREYgUEUsIHRoZW4gUEUxIGFuZCBQ
RTINCj4+Pj4+IHdpbGwgYm90aCBhY3QgYXMgUEUgZm9yIHNvbWUgdHJhbnNpZXQgdGltZS4NCj4+
Pj4+DQo+Pj4+PiBPbiB0aGUgb3RoZXIgaGFuZCwgdGhlIDMgc2Vjb25kIHRpbWVyIHdpbGwgc2Vy
dmUgYXMgYSBidWxrIGFuZCBiYXRjaA0KPj4+Pj4gbWVjaGFuaXNtIGZvciB0aGUgY2FzZSB3aGVu
IG5ldyBQRShzKSBpcyhhcmUpIGNvbWluZyB1cCwgc28gd2UgZG9uwrl0DQo+Pj4+PiBoYXZlIHRv
IGRvIGEgREYgZWxlY3Rpb24gZXZlciBzbyBvZnRlbi4NCj4+Pj4+DQo+Pj4+PiBIYW93ZWlndW8s
ICBJIHRoaW5rIHRoZSAibG9vcCIgc2NlbmFyaW8gaXMgc3BlY3VsYXRpdmUuDQo+Pj4+PiBJIHRo
aW5rIGV2ZXJ5Ym9keSB1bmRlcnN0YW5kcyB0aGF0LCB5ZXMsIHRoZXJlIG1heSBiZSBzb21lIHRy
YW5zaWVudA0KPj4+Pj4gZHVwbGljYXRpb24gb2YgYnVtIHRyYWZmaWM7IGhvd2V2ZXIsIHRoZXJl
IGlzIG5vICJsb29wIiBwcm9ibGVtIHRvIGJlDQo+Pj4+PiBzb2x2ZWQuDQo+Pj4+PiBbd2VpZ3Vv
XTogVHJhZmZpYyBkdXBsaWNhdGlvbiBhbmQgbG9vcCBpc3N1ZSBib3RoIGV4aXN0IGluIGR1YWwg
REYgUEUNCj4+Pj4+IGNhc2UuIEluIG11bHRpLWhvbWVkIGRldmljZSBjYXNlLCBzYXkgQ0UxIGlz
IG11bHRpLWhvbWVkIHRvIFBFMSBhbmQNCj4+Pj4+IFBFMiBhbmQgYm90aCBQRSBhcmUgREYgUEUu
IFdoZW4gQ0UxIHNlbmRzIHRyYWZmaWMgdG8gUEUxLCB0aGUgdHJhZmZpYw0KPj4+Pj4gd2lsbCBs
b29wIGJhY2sgdG8gQ0UxIHRocm91Z2ggUEUyLiBJZiBDRTEgY2FuIGZpbHRlciBhbmQgZHJvcCB0
aGUNCj4+Pj4+IHRyYWZmaWMgLCBpdCBpcyBmb3J0dW5hdGUsIG1heWJlIG5vIGxvb3AuDQo+Pj4+
PiBGb3IgbXVsdGktaG9tZWQgbmV0d29yayBzY2VuYXJpbywgc2F5IGxvY2FsIG5ldHdvcmsxKENF
MSBhbmQgQ0UyKSBpcw0KPj4+Pj4gbXVsdGktaG9tZWQgdG8gUEUxIGFuZCBQRTIuIFNpbWlsYXJs
eSwgdGhlIHRyYWZmaWMgZnJvbSBDRTEgd2lsbCBsb29wDQo+Pj4+PiBiYWNrIHRocm91Z2ggUEUx
LS0+UEUyLS0+Q0UyLCBpdCBpcyB2ZXJ5IHNlcmlvdXMuIEl0IGlzIGFic29sdXRlbHkNCj4+Pj4+
bm90DQo+Pj4+PiBhbGxvd2VkLg0KPj4+Pj4NCj4+Pj4+IFRoYW5rcywNCj4+Pj4+IC0tU2F0eWEN
Cj4+Pj4+DQo+Pj4+Pg0KPj4+Pj4NCj4+Pj4+IE9uIDMvMjYvMTUgOTowOCBQTSwgIkhhb3dlaWd1
byIgPGhhb3dlaWd1b0BodWF3ZWkuY29tPiB3cm90ZToNCj4+Pj4+DQo+Pj4+Pj4gSGkgVGhvbWFz
LA0KPj4+Pj4+IEkga25vdyB5b3UgYXJlIG5ldXRyYWwsIHNvcnJ5IGZvciBteSByZXBseSB2dWxu
ZXJhYmxlIHRvIGJlDQo+Pj4+Pj4gbWl1bmRlcnN0b29kLg0KPj4+Pj4+DQo+Pj4+Pj4gVGhhbmtz
LA0KPj4+Pj4+IHdlaWd1bw0KPj4+Pj4+DQo+Pj4+Pj4gX19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fX19fX19fX19fXw0KPj4+Pj4+IEZyb206IHRob21hcy5tb3JpbkBvcmFuZ2UuY29tIFt0
aG9tYXMubW9yaW5Ab3JhbmdlLmNvbV0NCj4+Pj4+PiBTZW50OiBGcmlkYXksIE1hcmNoIDI3LCAy
MDE1IDExOjU5DQo+Pj4+Pj4gVG86IEhhb3dlaWd1bzsgYmVzc0BpZXRmLm9yZw0KPj4+Pj4+IFN1
YmplY3Q6IFJlOiBbYmVzc10gSGFuZHNoYWtpbmcgYW1vbmcgUEVzIGluIGFuIEVWUE4gRVMgYmFz
ZWQgb24NCj4+Pj4+PiBQYXhvcyBhbGdvcml0aG0NCj4+Pj4+Pg0KPj4+Pj4+IEhpIFdlaWd1bywN
Cj4+Pj4+Pg0KPj4+Pj4+IDIwMTUtMDMtMjYsIEhhb3dlaWd1bzoNCj4+Pj4+Pj4gVGhvbWFzOg0K
Pj4+Pj4+Pj4gLSBSRkM3NDMyIG1heSBoYXZlIHRyYW5zaWVudCBwZXJpb2RzIHdoZXJlIHRoZSBE
RiBlbGVjdGlvbiBzdGF0ZQ0KPj4+Pj4+Pj4gaXMgbm90IHlldCBzeW5jaHJvbml6ZWQgYmV0d2Vl
biB0aGUgdHdvIHBlZXJzOg0KPj4+Pj4+PiBbd2VpZ3VvXTogWWVzLCBpIHRoaW5rIFJGQyA3NDMy
IGhhcyB0cmFuc2llbnQgcGVyaW9kcyBvZiB0cmFmZmljDQo+Pj4+Pj4+IGxvb3ANCj4+Pj4+PiBO
b3RlIHdlbGwgdGhhdCBJIGRpZG4ndCB3cml0ZSB0aGF0IHRoZXJlIGNhbiBiZSB0cmFuc2llbnQg
X2xvb3BzXyAuDQo+Pj4+Pj4gSSB0cmllZCB0byBjYXB0dXJlIHRoZSBleGNoYW5nZSB5b3UgaGFk
IGFsbCwgYW5kIHdoYXQgSSBnYXRoZXIgaXMNCj4+Pj4+PiB0aGF0IHRoZSBzcGxpdC1ob3Jpem9u
IHByb2NlZHVyZSBfcHJldmVudHNfbG9vcHNfLCBpbmRlcGVuZGVudGx5IG9mDQo+Pj4+Pj4gYW55
IHRyYW5zaWVudCBwZXJpb2Qgd2hlcmUgREYgc3RhdGUgaXMgbm90IHN5bmNocm9uaXplZCBhbmQg
d2hpY2gNCj4+Pj4+Pm1heQ0KPj4+Pj4+IGxlYWQgdG8gdHJhbnNpZW50IF9kdXBsaWNhdGVzXy4N
Cj4+Pj4+Pg0KPj4+Pj4+IC1UaG9tYXMNCj4+Pj4+Pg0KPj4+Pj4+PiAyNi8wMy8yMDE1IDIwOjA1
LCBKb2huIEUgRHJha2UgOg0KPj4+Pj4+Pj4gV2VpZ3VvLA0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IEkg
Z3Vlc3MgSSB3YXNuwrl0IGNsZWFyLiAgSSB0aGluayB5b3UgZHJhZnQsIGZvciB0aGUgcmVhc29u
cyBJIGhhdmUNCj4+Pj4+Pj4+IGRldGFpbGVkLCBpcyBhIG5vbi1zb2x1dGlvbiB0byBhIG5vbi1w
cm9ibGVtIHdpdGggdHJlbWVuZG91cw0KPj4+Pj4+Pj4gY29udHJvbCBwbGFuZSBjb3N0Lg0KPj4+
Pj4+Pj4NCj4+Pj4+Pj4+IFlvdXJzIElycmVzcGVjdGl2ZWx5LA0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+
IEpvaG4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiAqRnJvbToqSGFvd2VpZ3VvIFttYWlsdG86aGFvd2Vp
Z3VvQGh1YXdlaS5jb21dDQo+Pj4+Pj4+PiAqU2VudDoqIFRodXJzZGF5LCBNYXJjaCAyNiwgMjAx
NSA2OjE3IFBNDQo+Pj4+Pj4+PiAqVG86KiBKb2huIEUgRHJha2U7IFBhdHJpY2UgQnJpc3NldHRl
IChwYnJpc3NldCkNCj4+Pj4+Pj4+ICpDYzoqIEFsaSBTYWphc3NpIChzYWphc3NpKTsgYmVzc0Bp
ZXRmLm9yZw0KPj4+Pj4+Pj4gKlN1YmplY3Q6KiBSRTogW2Jlc3NdIEhhbmRzaGFraW5nIGFtb25n
IFBFcyBpbiBhbiBFVlBOIEVTIGJhc2VkIG9uDQo+Pj4+Pj4+PiBQYXhvcyBhbGdvcml0aG0NCj4+
Pj4+Pj4+DQo+Pj4+Pj4+PiBQbHMgc2VlIGJlbG93Lg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IFRoYW5r
cywNCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiB3ZWlndW8NCj4+Pj4+Pj4+DQo+Pj4+Pj4+Pg0KPj4+Pj4+
Pj4gDQo+Pj4+Pj4+Pi0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0NCj4+Pj4+Pj4+IC0tLQ0KPj4+Pj4+Pj4gLQ0KPj4+Pj4+
Pj4gLQ0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICpGcm9tOipKb2huIEUgRHJha2UgW2pkcmFrZUBqdW5p
cGVyLm5ldF0NCj4+Pj4+Pj4+ICpTZW50OiogRnJpZGF5LCBNYXJjaCAyNywgMjAxNSA2OjAwDQo+
Pj4+Pj4+PiAqVG86KiBIYW93ZWlndW87IFBhdHJpY2UgQnJpc3NldHRlIChwYnJpc3NldCkNCj4+
Pj4+Pj4+ICpDYzoqIEFsaSBTYWphc3NpIChzYWphc3NpKTsgYmVzc0BpZXRmLm9yZyA8bWFpbHRv
OmJlc3NAaWV0Zi5vcmc+DQo+Pj4+Pj4+PiAqU3ViamVjdDoqIFJFOiBbYmVzc10gSGFuZHNoYWtp
bmcgYW1vbmcgUEVzIGluIGFuIEVWUE4gRVMgYmFzZWQgb24NCj4+Pj4+Pj4+IFBheG9zIGFsZ29y
aXRobQ0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IFRvIHJlY2FwLA0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IFdl
IGhhdmUgZXN0YWJsaXNoZWQgdGhhdCB5b3VyIHByb3Bvc2FsIGlzIHVudGVuYWJsZSBiZWNhdXNl
IG9mIGl0cw0KPj4+Pj4+Pj4gY29udHJvbCBwbGFuZSBsb2FkLg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+
IFdlIGhhdmUgZXN0YWJsaXNoZWQgdGhhdCB5b3VyIHByb3Bvc2FsIGlzIGJhc2VkIHVwb24gYSBm
bGF3ZWQNCj4+Pj4+Pj4+IHVuZGVyc3RhbmRpbmcgb2YgdGhlIERGIGVsZWN0aW9uIGluIFJGQyA3
NDMyLg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IFt3ZWlndW9dOiBJbiBldGhlcm5ldCB3b3JsZCwgdHJh
ZmZpYyBsb29wIGlzIHNlcmlvdXMgdGhhbiBzaG9ydA0KPj4+Pj4+Pj4gdGltZXIgdHJhZmZpYyBk
aXNydXB0aW9uLiBJZiB5b3Ugd2FudCB0byBpbXBsZW1lbnQgIHRyYW5zaWV0DQo+Pj4+Pj4+PiB0
cmFmZmljIGxvb3AgcHJvY2VzcywgaSB3aWxsIG1vZGlmeSBteSBkcmFmdCB0byBzb2x2ZSB5b3Vy
IGlzc3VlLg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IElmIGkgYW0gdGhlIGRldmVsb3BlciwgaSB3aWxs
IHByZWZlciBzaG9ydCB0aW1lciB0cmFmZmljDQo+Pj4+Pj4+PiBkaXNydXB0aW9uIGJhc2VkIG9u
IGN1cnJlbnQgRVZQTiBwcm90b2NvbC4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiBXaGF0IHlvdSBhcmUg
bm93IGFyZ3VpbmcgaXMgdGhhdCB5b3VyIGRyYWZ0IHByZXZlbnRzIHR3byBvciBtb3JlDQo+Pj4+
Pj4+PiBQRXMgZnJvbSBiZWluZyBERiBzaW11bHRhbmVvdXNseS4gVGhpcyBpcyBjbGVhcmx5IG5v
bnNlbnNlLg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IFt3ZWlndW9dOiBJIHdpbGwgbW9kaWZ5IHRoZSBk
cmFmdCBwcm9ibGVtIHN0YXRlbWVudHMsIGFuZCB1c2UgdGhlDQo+Pj4+Pj4+PiBzYW1lICBoYW5k
c2hha2luZyBzb2x1dGlvbiB0byBzb2x2ZSBpdC4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiBGdXJ0aGVy
bW9yZSwgd2UgaGF2ZSBlc3RhYmxpc2hlZCB0aGF0IGhhdmluZyB0d28gb3IgbW9yZSBERnMgZm9y
DQo+Pj4+Pj4+PiB3aGF0IGV2ZW4geW91IGFkbWl0IGlzIGEgYnJpZWYgdHJhbnNpZW50IGxlYWRz
IHRvIGR1cGxpY2F0ZQ0KPj4+Pj4+Pj4gdHJhZmZpYywgd2hpY2ggaXMgYWNjZXB0YWJsZSwgYnV0
IG5vdCBsb29wcywgeW91ciBhc3NlcnRpb24gdG8gdGhlDQo+Pj4gY29udHJhcnkuDQo+Pj4+Pj4+
PiBbd2VpZ3VvXTogSXQgaXMgdHJhbnNpZW50IGxvb3AgYW5kIHRyYWZmaWMgZHVwbGljYXRpb24g
aXNzdWUuDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gWW91cnMgSXJyZXNwZWN0aXZlbHksDQo+Pj4+Pj4+
Pg0KPj4+Pj4+Pj4gSm9obg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICpGcm9tOipIYW93ZWlndW8gW21h
aWx0bzpoYW93ZWlndW9AaHVhd2VpLmNvbV0NCj4+Pj4+Pj4+ICpTZW50OiogVGh1cnNkYXksIE1h
cmNoIDI2LCAyMDE1IDU6MzcgUE0NCj4+Pj4+Pj4+ICpUbzoqIEpvaG4gRSBEcmFrZTsgUGF0cmlj
ZSBCcmlzc2V0dGUgKHBicmlzc2V0KQ0KPj4+Pj4+Pj4gKkNjOiogQWxpIFNhamFzc2kgKHNhamFz
c2kpOyBiZXNzQGlldGYub3JnIDxtYWlsdG86YmVzc0BpZXRmLm9yZz4NCj4+Pj4+Pj4+ICpTdWJq
ZWN0OiogUkU6IFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQRXMgaW4gYW4gRVZQTiBFUyBiYXNl
ZCBvbg0KPj4+Pj4+Pj4gUGF4b3MgYWxnb3JpdGhtDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gSm9obiwN
Cj4+Pj4+Pj4+DQo+Pj4+Pj4+PiBBcyB5b3VyIHVuZGVyc3RhbmRpbmcgb2YgdGhlIEVWUE4gZHJh
ZnQsIHRoZSBERiBlbGVjdGlvbiBtZWNoYW5pc20NCj4+Pj4+Pj4+IGhhcyAgbW9yZSBzZXJpb3Vz
IHNpZGUgZWZmZWN0LCBpdCB3aWxsIGhhdmUgc2hvcnQgdGltZSB0cmFmZmljDQo+Pj4+Pj4+PiBs
b29wLGkuZS4sICBkdWFsIERGIFBFcyB3aWxsIGV4aXN0IGZvciBhIHNob3J0IHRpbWUuIEkgdGhp
bmsgZHVhbA0KPj4+Pj4+Pj4gREYgUEVzIGlzICBhYnNvbHV0ZWx5IG5vdCB0b2xlcmF0ZWQsIGJl
Y2F1c2UgbmF0aXZlIGV0aGVybmV0DQo+Pj4+Pj4+PmhlYWRlcg0KPj4+Pj4+Pj4gaGFzIG5vIFRU
TCwgIHVwIHRvIHNldmVyYWwgaHVuZHJlZCBtcyB0cmFmZmljIGxvb3Agbm9ybWFsbHkgbm90DQo+
Pj4+Pj4+PiB0b2xlcmF0ZWQgaW4gIGNvbW1lcnRpYWwgbmV0d29ya3MuDQo+Pj4+Pj4+Pg0KPj4+
Pj4+Pj4gQXMgeW91ciB1bmRlcnN0YW5kaW5nLCB0aGUgUEVzIHNob3VsZCBkbyBhcyBmb2xsb3dp
bmc6DQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gMS4gQWNjdXJhdGUgdGltZXIgc3luYy4gTlRQIGFjY3Vy
YWN5IGlzIGJhZCwgMTU4OHYyIGlzIGdvb2QgYnV0DQo+Pj4+Pj4+PiBoYXZlIHJhcmVseSBkZXBs
b3ltZW50Lg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IEFzc3VtaW5nIFBFMSxQRTIgYW5kIFBFMyBoYXZl
IGNvbnNpc3RlbnQgdGltZXIgY2xvY2ssIHdoZW4gUEUzDQo+Pj4+Pj4+PiBqb2lucyBFU0kgYW5k
IHRyaWdnZXIgREYgcmUtZWxlY3Rpb24uIFdoZW4gcmVjZXB0aW9uIHRpbWVyDQo+Pj4+Pj4+PmV4
cGlyZXM6DQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gUEUxIHVwZ3JhZGVzIHRvIERGIFBFLg0KPj4+Pj4+
Pj4NCj4+Pj4+Pj4+IEFmdGVyIHJlY2VwdGlvbiB0aW1lcisgRVMgcm91dGUgdHJhbnNtaXNzaW9u
IHRpbWVyOg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IFBFMiBkb3dubG9hZHMgdG8gbm9uLURGIFBFLg0K
Pj4+Pj4+Pj4NCj4+Pj4+Pj4+IFNvIGluIHRpbWVyIGNsb2NrIHN5bmMgY2FzZSwgZHVhbCBERiBQ
RXMgd2lsbCBleGlzdCBhdCBsZWFzdA0KPj4+Pj4+Pj4gdHJhbnNtaXNzaW9uIHRpbWVyLg0KPj4+
Pj4+Pj4NCj4+Pj4+Pj4+IElmIE5UUCBpcyB1c2VkIGZvciB0aW1lciBzeW5jLCBiZWNhdXNlIGl0
IGhhcyBiYWQgYWNjdXJhY3ksIGR1YWwNCj4+Pj4+Pj4+IERGIFBFcyB3aWxsIGV4aXN0IG1vcmUg
bG9uZ2VyIHRpbWVyLg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IFNvIGFzIHlvdXIgdW5kZXJzdGFuZGlu
ZyBmb3IgREYgZWxlY3Rpb24sIHRoZSBkcmF3YmFjayBpcyBtb3JlDQo+Pj4+Pj4+PmNsZWFyLg0K
Pj4+Pj4+Pj4NCj4+Pj4+Pj4+IFRoYW5rcywNCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiB3ZWlndW8NCj4+
Pj4+Pj4+DQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gDQo+Pj4+Pj4+Pi0tLS0tLS0tLS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0NCj4+Pj4+Pj4+
IC0tLQ0KPj4+Pj4+Pj4gLQ0KPj4+Pj4+Pj4gLQ0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICpGcm9tOipK
b2huIEUgRHJha2UgW2pkcmFrZUBqdW5pcGVyLm5ldF0NCj4+Pj4+Pj4+ICpTZW50OiogRnJpZGF5
LCBNYXJjaCAyNywgMjAxNSA0OjQxDQo+Pj4+Pj4+PiAqVG86KiBIYW93ZWlndW87IFBhdHJpY2Ug
QnJpc3NldHRlIChwYnJpc3NldCkNCj4+Pj4+Pj4+ICpDYzoqIEFsaSBTYWphc3NpIChzYWphc3Np
KTsgYmVzc0BpZXRmLm9yZyA8bWFpbHRvOmJlc3NAaWV0Zi5vcmc+DQo+Pj4+Pj4+PiAqU3ViamVj
dDoqIFJFOiBbYmVzc10gSGFuZHNoYWtpbmcgYW1vbmcgUEVzIGluIGFuIEVWUE4gRVMgYmFzZWQg
b24NCj4+Pj4+Pj4+IFBheG9zIGFsZ29yaXRobQ0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IFdlaWd1bywN
Cj4+Pj4+Pj4+DQo+Pj4+Pj4+PiBXZSBoYXZlIGFscmVhZHkgZXN0YWJsaXNoZWQgdGhhdCB5b3Vy
IHByb3Bvc2FsIGlzIHVudGVuYWJsZQ0KPj4+Pj4+Pj4gYmVjYXVzZSBvZiAgaXRzIGNvbnRyb2wg
cGxhbmUgbG9hZC4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiBXaGF0IHdlIGFyZSBub3cgZGlzY3Vzc2lu
ZyBpcyB0aGF0IHlvdXIgcHJvcG9zYWwgaXMgYmFzZWQgdXBvbiBhDQo+Pj4+Pj4+PiBtaXN1bmRl
cnN0YW5kaW5nIG9mIHRoZSBhbGdvcml0aG0gaW4gUkZDIDc0MzIuICBZb3UgYXJlIGFzc3VtaW5n
DQo+Pj4+Pj4+PiB0aGF0DQo+Pj4+Pj4+PiBQRTEgd2lsbCBhZHZlcnRpc2UgYW4gRVMgcm91dGUg
YW5kIHRoZW4gd2FpdCBmb3IgdGhlIGNvbmZpZ3VyZWQNCj4+Pj4+Pj4+IGludGVydmFsIGJlZm9y
ZSBwZXJmb3JtaW5nIHRoZSBERiBlbGVjdGlvbiB3aGlsZSBQRTIgYW5kIFBFMyB3aWxsDQo+Pj4+
Pj4+PiBwZXJmb3JtIHRoZSBERiBlbGVjdGlvbiBhcyBzb29uIGFzIHRoZXkgcmVjZWl2ZSB0aGUg
RVMgcm91dGUgZnJvbQ0KPj4+Pj4+Pj4gUEUxLg0KPj4+Pj4+Pj4gVGhpcyBpcyBub3Qgd2hhdCBS
RkMgNzQzMiBzYXlzLg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IFJhdGhlciwgd2hhdCBpcyBzYXlzIGlz
IHRoYXQgdGhlIGFkdmVydGlzZW1lbnQgb2YgdGhlIEVTIHJvdXRlIGJ5DQo+Pj4+Pj4+PiBQRTEg
YW5kIGl0cyByZWNlaXB0IGJ5IFBFMiBhbmQgUEUzIGNhdXNlcyBhbGwgdGhyZWUgUEVzIHRvIHN0
YXJ0DQo+Pj4+Pj4+PiB0aGUgY29uZmlndXJlZCBpbnRlcnZhbCB0aW1lciAgLSDCszMuIFdoZW4g
dGhlIHRpbWVyIGV4cGlyZXMsIGVhY2gNCj4+Pj4+Pj4+IFBFIGJ1aWxkcyBhbiBvcmRlcmVkIGxp
c3Qgb2YgdGhlIElQIGFkZHJlc3NlcyBvZiBhbGwgdGhlIFBFIG5vZGVzDQo+Pj4+Pj4+PiBjb25u
ZWN0ZWQgdG8gdGhlIEV0aGVybmV0IHNlZ21lbnQgKGluY2x1ZGluZyBpdHNlbGYpLCBpbg0KPj4+
Pj4+Pj4gaW5jcmVhc2luZyBudW1lcmljIHZhbHVlLsKyDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gWW91
cnMgSXJyZXNwZWN0aXZlbHksDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gSm9obg0KPj4+Pj4+Pj4NCj4+
Pj4+Pj4+ICpGcm9tOipIYW93ZWlndW8gW21haWx0bzpoYW93ZWlndW9AaHVhd2VpLmNvbV0NCj4+
Pj4+Pj4+ICpTZW50OiogVGh1cnNkYXksIE1hcmNoIDI2LCAyMDE1IDM6MjYgUE0NCj4+Pj4+Pj4+
ICpUbzoqIEpvaG4gRSBEcmFrZTsgUGF0cmljZSBCcmlzc2V0dGUgKHBicmlzc2V0KQ0KPj4+Pj4+
Pj4gKkNjOiogQWxpIFNhamFzc2kgKHNhamFzc2kpOyBiZXNzQGlldGYub3JnIDxtYWlsdG86YmVz
c0BpZXRmLm9yZz4NCj4+Pj4+Pj4+ICpTdWJqZWN0OiogUkU6IFtiZXNzXSBIYW5kc2hha2luZyBh
bW9uZyBQRXMgaW4gYW4gRVZQTiBFUyBiYXNlZCBvbg0KPj4+Pj4+Pj4gUGF4b3MgYWxnb3JpdGht
DQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gUGxzIHJlYWQgbXkgZGV0YWlsIHJlcGxpZXMgdG8gU2F0eWEu
IElmIHlvdSBzdGlsbCBjYW4ndCBjYXRjaCBpdCwNCj4+Pj4+Pj4+IHBscyByZWFkIG15IGRyYWZ0
IGFuZCBFVlBOIGJhc2UgcHJvdG9jb2wsICB0aGFua3MNCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiB3ZWln
dW8NCj4+Pj4+Pj4+DQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gDQo+Pj4+Pj4+Pi0tLS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0NCj4+
Pj4+Pj4+IC0tLQ0KPj4+Pj4+Pj4gLQ0KPj4+Pj4+Pj4gLQ0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICpG
cm9tOipKb2huIEUgRHJha2UgW2pkcmFrZUBqdW5pcGVyLm5ldF0NCj4+Pj4+Pj4+ICpTZW50Oiog
RnJpZGF5LCBNYXJjaCAyNywgMjAxNSAxOjI4DQo+Pj4+Pj4+PiAqVG86KiBQYXRyaWNlIEJyaXNz
ZXR0ZSAocGJyaXNzZXQpDQo+Pj4+Pj4+PiAqQ2M6KiBIYW93ZWlndW87IEFsaSBTYWphc3NpIChz
YWphc3NpKTsgYmVzc0BpZXRmLm9yZw0KPj4+Pj4+Pj4gPG1haWx0bzpiZXNzQGlldGYub3JnPg0K
Pj4+Pj4+Pj4gKlN1YmplY3Q6KiBSZTogW2Jlc3NdIEhhbmRzaGFraW5nIGFtb25nIFBFcyBpbiBh
biBFVlBOIEVTIGJhc2VkIG9uDQo+Pj4+Pj4+PiBQYXhvcyBhbGdvcml0aG0NCj4+Pj4+Pj4+DQo+
Pj4+Pj4+PiBJIHRoaW5rIFBhdHJpY2UgaXMgY29ycmVjdC4gIFlvdXIgcHJvcG9zYWwgZG9lc24n
dCBzb2x2ZSB0aGUNCj4+Pj4+Pj4+IHByb2JsZW0gYW5kIGl0IGRvZXMgc28gYXQgaHVnZSBjb3N0
Lg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+IFNlbnQgZnJvbSBteSBpUGhvbmUNCj4+Pj4+Pj4+DQo+Pj4+
Pj4+Pg0KPj4+Pj4+Pj4gT24gTWFyIDI2LCAyMDE1LCBhdCAxMDozNCBBTSwgUGF0cmljZSBCcmlz
c2V0dGUgKHBicmlzc2V0KQ0KPj4+Pj4+Pj4gPHBicmlzc2V0QGNpc2NvLmNvbSA8bWFpbHRvOnBi
cmlzc2V0QGNpc2NvLmNvbT4+IHdyb3RlOg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICAgICAgIFdlaWd1
bywNCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiAgICAgICBJwrltIG5vdCBzdXJlIEnCuW0gZm9sbG93aW5n
IGhlcmUuDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgRG9uwrl0IHlvdSBoYXZlIHRoZSBzYW1l
IGlzc3VlIHdpdGggeW91ciBoYW5kc2hha2luZw0KPj4+Pj4+Pj5tZWNoYW5pc20/DQo+Pj4+Pj4+
Pg0KPj4+Pj4+Pj4gICAgICAgSWYgeW91IGRvbsK5dCBrbm93IHlvdXIgcGVlciwgaG93IGNhbiB5
b3UgaGFuZHNoYWtlPw0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICAgICAgIFJlZ2FyZHMsDQo+Pj4+Pj4+
Pg0KPj4+Pj4+Pj4gICAgICAgUGF0cmljZQ0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICAgICAgIEltYWdl
IHJlbW92ZWQgYnkgc2VuZGVyLg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICAgICAgICpQYXRyaWNlIEJy
aXNzZXR0ZSoNCj4+Pj4+Pj4+ICAgICAgIFRFQ0hOSUNBTCBMRUFERVIuRU5HSU5FRVJJTkcNCj4+
Pj4+Pj4+DQo+Pj4+Pj4+PiAgICAgICBwYnJpc3NldEBjaXNjby5jb20gPG1haWx0bzpwYnJpc3Nl
dEBjaXNjby5jb20+DQo+Pj4+Pj4+PiAgICAgICBQaG9uZTogKisxIDYxMyAyNTQgMzMzNioNCj4+
Pj4+Pj4+DQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICAgICAgICpDaXNjbyBTeXN0ZW1z
IENhbmFkYSBDby4gLyBMZXMgU3lzdGVtZXMgQ2lzY28gQ2FuYWRhIENJRSoNCj4+Pj4+Pj4+ICAg
ICAgIENhbmFkYQ0KPj4+Pj4+Pj4gICAgICAgQ2lzY28uY29tIDxodHRwOi8vd3d3LmNpc2NvLmNv
bS9nbG9iYWwvQ0EvPg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAg
ICAgSW1hZ2UgcmVtb3ZlZCBieSBzZW5kZXIuVGhpbmsgYmVmb3JlIHlvdSBwcmludC4NCj4+Pj4+
Pj4+DQo+Pj4+Pj4+PiAgICAgICBUaGlzIGVtYWlsIG1heSBjb250YWluIGNvbmZpZGVudGlhbCBh
bmQgcHJpdmlsZWdlZCBtYXRlcmlhbA0KPj4+Pj4+Pj5mb3INCj4+Pj4+Pj4+ICAgICAgIHRoZSBz
b2xlIHVzZSBvZiB0aGUgaW50ZW5kZWQgcmVjaXBpZW50LiBBbnkgcmV2aWV3LCB1c2UsDQo+Pj4+
Pj4+PiAgICAgICBkaXN0cmlidXRpb24gb3IgZGlzY2xvc3VyZSBieSBvdGhlcnMgaXMgc3RyaWN0
bHkNCj4+Pj4+Pj4+cHJvaGliaXRlZC4gSWYNCj4+Pj4+Pj4+ICAgICAgIHlvdSBhcmUgbm90IHRo
ZSBpbnRlbmRlZCByZWNpcGllbnQgKG9yIGF1dGhvcml6ZWQgdG8NCj4+Pj4+Pj4+cmVjZWl2ZSBm
b3INCj4+Pj4+Pj4+ICAgICAgIHRoZSByZWNpcGllbnQpLCBwbGVhc2UgY29udGFjdCB0aGUgc2Vu
ZGVyIGJ5IHJlcGx5IGVtYWlsIGFuZA0KPj4+Pj4+Pj4gICAgICAgZGVsZXRlIGFsbCBjb3BpZXMg
b2YgdGhpcyBtZXNzYWdlLg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICAgICAgIFBsZWFzZSBjbGljayBo
ZXJlDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gPGh0dHA6Ly93d3cuY2lzY28uY29tL3dlYi9hYm91dC9k
b2luZ19idXNpbmVzcy9sZWdhbC9jcmkvaW5kZXguaA0KPj4+IHRtbD4NCj4+Pj4+Pj4+ICAgICAg
IGZvciBDb21wYW55IFJlZ2lzdHJhdGlvbiBJbmZvcm1hdGlvbi4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+
Pg0KPj4+Pj4+Pj4gICAgICAgKkZyb206ICpIYW93ZWlndW8gPGhhb3dlaWd1b0BodWF3ZWkuY29t
DQo+Pj4+Pj4+PiA8bWFpbHRvOmhhb3dlaWd1b0BodWF3ZWkuY29tPj4NCj4+Pj4+Pj4+ICAgICAg
ICpEYXRlOiAqVGh1cnNkYXksIE1hcmNoIDI2LCAyMDE1IGF0IDEwOjI0IEFNDQo+Pj4+Pj4+PiAg
ICAgICAqVG86ICpQYXRyaWNlIEJyaXNzZXR0ZSA8cGJyaXNzZXRAY2lzY28uY29tDQo+Pj4+Pj4+
PiAgICAgICA8bWFpbHRvOnBicmlzc2V0QGNpc2NvLmNvbT4+LCBKb2huIEUgRHJha2UNCj4+Pj4+
Pj4+PGpkcmFrZUBqdW5pcGVyLm5ldA0KPj4+Pj4+Pj4gICAgICAgPG1haWx0bzpqZHJha2VAanVu
aXBlci5uZXQ+PiwgQWxpIFNhamFzc2kgPHNhamFzc2lAY2lzY28uY29tDQo+Pj4+Pj4+PiAgICAg
ICA8bWFpbHRvOnNhamFzc2lAY2lzY28uY29tPj4sICJiZXNzQGlldGYub3JnDQo+Pj4+Pj4+PiAg
ICAgICA8bWFpbHRvOmJlc3NAaWV0Zi5vcmc+IiA8YmVzc0BpZXRmLm9yZw0KPj4+Pj4+Pj48bWFp
bHRvOmJlc3NAaWV0Zi5vcmc+Pg0KPj4+Pj4+Pj4gICAgICAgKlN1YmplY3Q6ICpSRTogW2Jlc3Nd
IEhhbmRzaGFraW5nIGFtb25nIFBFcyBpbiBhbiBFVlBOIEVTDQo+Pj4+Pj4+PiBiYXNlZCBvbg0K
Pj4+Pj4+Pj4gICAgICAgUGF4b3MgYWxnb3JpdGhtDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAg
ICAgIEhpIFBhdHJpY2UsDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgIFVwIHRvIHJlY2Vw
dGlvbiB0aW1lciB0cmFmZmljIGRpc3J1cHRpb24gaW4gdHJhbnNpZW50DQo+Pj4+Pj4+PiBwaGFz
ZSBpcw0KPj4+Pj4+Pj4gICAgICAgICAgIG9uZSBvZiB0aGUgaXNzdWVzLg0KPj4+Pj4+Pj4NCj4+
Pj4+Pj4+ICAgICAgICAgICBUaGFua3MsDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgIHdl
aWd1bw0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiANCj4+Pj4+Pj4+LS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLQ0K
Pj4+Pj4+Pj4gLS0tDQo+Pj4+Pj4+PiAtDQo+Pj4+Pj4+PiAtDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4g
ICAgICAgICAgICpGcm9tOipQYXRyaWNlIEJyaXNzZXR0ZSAocGJyaXNzZXQpIFtwYnJpc3NldEBj
aXNjby5jb20NCj4+Pj4+Pj4+ICAgICAgICAgICA8bWFpbHRvOnBicmlzc2V0QGNpc2NvLmNvbT5d
DQo+Pj4+Pj4+PiAgICAgICAgICAgKlNlbnQ6KiBUaHVyc2RheSwgTWFyY2ggMjYsIDIwMTUgMjA6
NTQNCj4+Pj4+Pj4+ICAgICAgICAgICAqVG86KiBIYW93ZWlndW87IEpvaG4gRSBEcmFrZTsgQWxp
IFNhamFzc2kgKHNhamFzc2kpOw0KPj4+Pj4+Pj4gICAgICAgICAgIGJlc3NAaWV0Zi5vcmcgPG1h
aWx0bzpiZXNzQGlldGYub3JnPg0KPj4+Pj4+Pj4gICAgICAgICAgICpTdWJqZWN0OiogUmU6IFti
ZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQRXMgaW4gYW4gRVZQTg0KPj4+Pj4+Pj5FUw0KPj4+Pj4+
Pj4gICAgICAgICAgIGJhc2VkIG9uIFBheG9zIGFsZ29yaXRobQ0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+
ICAgICAgICAgICBXZWlndW8sDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgIFlvdSBtZW50
aW9uICJCdXQgaWYgeW91ciBkcmFmdCBoYXZlIG5vdCBzb2x2ZWQgYWxsDQo+Pj4+Pj4+PiBpc3N1
ZXPCsiwNCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiAgICAgICAgICAgQ2FuIHlvdSBleHBsYWluIHdoYXQg
U2F0eWHCuXMgZHJhZnQgaXMgbm90IHNvbHZpbmc/DQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAg
ICAgIFJlZ2FyZHMsDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgIFBhdHJpY2UNCj4+Pj4+
Pj4+DQo+Pj4+Pj4+PiAgICAgICAgICAgSW1hZ2UgcmVtb3ZlZCBieSBzZW5kZXIuDQo+Pj4+Pj4+
Pg0KPj4+Pj4+Pj4gICAgICAgICAgICpQYXRyaWNlIEJyaXNzZXR0ZSoNCj4+Pj4+Pj4+ICAgICAg
ICAgICBURUNITklDQUwgTEVBREVSLkVOR0lORUVSSU5HDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAg
ICAgICAgIHBicmlzc2V0QGNpc2NvLmNvbSA8bWFpbHRvOnBicmlzc2V0QGNpc2NvLmNvbT4NCj4+
Pj4+Pj4+ICAgICAgICAgICBQaG9uZTogKisxIDYxMyAyNTQgMzMzNioNCj4+Pj4+Pj4+DQo+Pj4+
Pj4+Pg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICAgICAgICAgICAqQ2lzY28gU3lzdGVtcyBDYW5hZGEg
Q28uIC8gTGVzIFN5c3RlbWVzIENpc2NvIENhbmFkYQ0KPj4+Pj4+Pj5DSUUqDQo+Pj4+Pj4+PiAg
ICAgICAgICAgQ2FuYWRhDQo+Pj4+Pj4+PiAgICAgICAgICAgQ2lzY28uY29tIDxodHRwOi8vd3d3
LmNpc2NvLmNvbS9nbG9iYWwvQ0EvPg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+Pg0KPj4+
Pj4+Pj4gICAgICAgICAgIEltYWdlIHJlbW92ZWQgYnkgc2VuZGVyLlRoaW5rIGJlZm9yZSB5b3Ug
cHJpbnQuDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgIFRoaXMgZW1haWwgbWF5IGNvbnRh
aW4gY29uZmlkZW50aWFsIGFuZCBwcml2aWxlZ2VkDQo+Pj4+Pj4+Pm1hdGVyaWFsDQo+Pj4+Pj4+
PiAgICAgICAgICAgZm9yIHRoZSBzb2xlIHVzZSBvZiB0aGUgaW50ZW5kZWQgcmVjaXBpZW50LiBB
bnkgcmV2aWV3LA0KPj4+Pj4+Pj51c2UsDQo+Pj4+Pj4+PiAgICAgICAgICAgZGlzdHJpYnV0aW9u
IG9yIGRpc2Nsb3N1cmUgYnkgb3RoZXJzIGlzIHN0cmljdGx5DQo+Pj4+Pj4+PnByb2hpYml0ZWQu
DQo+Pj4+Pj4+PiAgICAgICAgICAgSWYgeW91IGFyZSBub3QgdGhlIGludGVuZGVkIHJlY2lwaWVu
dCAob3IgYXV0aG9yaXplZCB0bw0KPj4+Pj4+Pj4gICAgICAgICAgIHJlY2VpdmUgZm9yIHRoZSBy
ZWNpcGllbnQpLCBwbGVhc2UgY29udGFjdCB0aGUgc2VuZGVyIGJ5DQo+Pj4+Pj4+PiByZXBseQ0K
Pj4+Pj4+Pj4gICAgICAgICAgIGVtYWlsIGFuZCBkZWxldGUgYWxsIGNvcGllcyBvZiB0aGlzIG1l
c3NhZ2UuDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgIFBsZWFzZSBjbGljayBoZXJlDQo+
Pj4+Pj4+Pg0KPj4+Pj4+Pj4gPGh0dHA6Ly93d3cuY2lzY28uY29tL3dlYi9hYm91dC9kb2luZ19i
dXNpbmVzcy9sZWdhbC9jcmkvaW5kZXguaA0KPj4+IHRtbD4NCj4+Pj4+Pj4+ICAgICAgICAgICBm
b3IgQ29tcGFueSBSZWdpc3RyYXRpb24gSW5mb3JtYXRpb24uDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4N
Cj4+Pj4+Pj4+ICAgICAgICAgICAqRnJvbTogKkhhb3dlaWd1byA8aGFvd2VpZ3VvQGh1YXdlaS5j
b20NCj4+Pj4+Pj4+ICAgICAgICAgICA8bWFpbHRvOmhhb3dlaWd1b0BodWF3ZWkuY29tPj4NCj4+
Pj4+Pj4+ICAgICAgICAgICAqRGF0ZTogKldlZG5lc2RheSwgTWFyY2ggMjUsIDIwMTUgYXQgMTA6
MzggUE0NCj4+Pj4+Pj4+ICAgICAgICAgICAqVG86ICpKb2huIEUgRHJha2UgPGpkcmFrZUBqdW5p
cGVyLm5ldA0KPj4+Pj4+Pj4gICAgICAgICAgIDxtYWlsdG86amRyYWtlQGp1bmlwZXIubmV0Pj4s
IEFsaSBTYWphc3NpDQo+Pj4+Pj4+PjxzYWphc3NpQGNpc2NvLmNvbQ0KPj4+Pj4+Pj4gICAgICAg
ICAgIDxtYWlsdG86c2FqYXNzaUBjaXNjby5jb20+PiwgImJlc3NAaWV0Zi5vcmcNCj4+Pj4+Pj4+
ICAgICAgICAgICA8bWFpbHRvOmJlc3NAaWV0Zi5vcmc+IiA8YmVzc0BpZXRmLm9yZw0KPj4+Pj4+
Pj4gPG1haWx0bzpiZXNzQGlldGYub3JnPj4NCj4+Pj4+Pj4+ICAgICAgICAgICAqU3ViamVjdDog
KlJlOiBbYmVzc10gSGFuZHNoYWtpbmcgYW1vbmcgUEVzIGluIGFuIEVWUE4NCj4+Pj4+Pj4+RVMN
Cj4+Pj4+Pj4+ICAgICAgICAgICBiYXNlZCBvbiBQYXhvcyBhbGdvcml0aG0NCj4+Pj4+Pj4+DQo+
Pj4+Pj4+PiAgICAgICAgICAgICAgIEhpIEpvaG4sDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAg
ICAgICAgICAgRmlyc3RseSBpIHRoaW5rIEVWUE4gY29tbXVuaXR5IHNob3VsZCByZWFjaA0KPj4+
Pj4+Pj4gY29uc2Vuc3VzIG9uDQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIHRoZSBpc3N1ZXMgb2Yg
Y3VycmVudCBERiBlbGVjdGlvbiBtZWNoYW5pc20uIEFsbA0KPj4+Pj4+Pj50aGVzZQ0KPj4+Pj4+
Pj4gICAgICAgICAgICAgICBpc3N1ZXMgc2hvdWxkIGJlIHJlc29sdmVkIGluIGEgc2luZ2xlIG5l
dyBERg0KPj4+Pj4+Pj5lbGVjdGlvbg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBkcmFmdCxyYXRo
ZXIgdGhhbiBpbiBtdWx0aXBsZSBzZXBhcmF0ZSBkcmFmdHMuIElmDQo+Pj4+Pj4+PnlvdXINCj4+
Pj4+Pj4+ICAgICAgICAgICAgICAgZHJhZnQgY2FuIHNvbHZlIGFsbCB0aGVzZSBpc3N1ZXMgYW5k
IHN0YWJsZSwgaSBoYXZlDQo+Pj4+Pj4+Pm5vDQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIHF1ZXN0
aW9uIGZvciBpdHMgcHJvZ3Jlc3NpbmcuIEJ1dCBpZiB5b3VyIGRyYWZ0DQo+Pj4+Pj4+PmhhdmUg
bm90DQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIHNvbHZlZCBhbGwgaXNzdWVzLCBpIHRoaW5rIGl0
IGhhZCBiZXR0ZXIgY29tYmluZQ0KPj4+Pj4+Pj53aXRoDQo+Pj4+Pj4+PiAgICAgICAgICAgICAg
IG90aGVyIGRyYWZ0cyB0byBwcm92aWRlIGEgY29tcHJlaGVuc2l2ZSBzb2x1dGlvbi4gSQ0KPj4+
Pj4+Pj4gdGhpbmsNCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgdGhlIGlzc3VlcyBsaXN0ZWQgaW4N
Cj4+Pj4+Pj4+ZHJhZnQtcmFiYWRhbi1iZXNzLWV2cG4tYWMtZGYtMDENCj4+Pj4+Pj4+IGFuZA0K
Pj4+Pj4+Pj4gICAgICAgICAgICAgICBkcmFmdC1oYW8tYmVzcy1ldnBuLWRmLWhhbmRzaGFraW5n
LTAwIGlzIHZhbGlkLCBpdA0KPj4+Pj4+Pj4gc2hvdWxkDQo+Pj4+Pj4+PiAgICAgICAgICAgICAg
IGJlIHJlc29sdmVkLiBTbyBpIHRoaW5rIGFsdGhvdWdoIHlvdXIgbmV3IEhhc2gNCj4+Pj4+Pj4+
YWxnb3JpdGhtDQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIGZvciBERiBlbGVjdGlvbiBpcyBnb29k
LCBpdCBvbmx5IGluY2x1ZGVzIHBhcnRpYWwNCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgZW5oYW5j
ZW1lbnRzLCBtYXliZSBpdCBzdGlsbCBuZWVkcyBzb21lIHRpbWUgZm9yDQo+Pj4+Pj4+PiBjb25z
ZW5zdXMuDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBUaGFua3MsDQo+Pj4+Pj4+
Pg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICB3ZWlndW8NCj4+Pj4+Pj4+DQo+Pj4+Pj4+Pg0KPj4+
Pj4+Pj4gDQo+Pj4+Pj4+Pi0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0NCj4+Pj4+Pj4+IC0tLQ0KPj4+Pj4+Pj4gLQ0KPj4+
Pj4+Pj4gLQ0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgKkZyb206KkpvaG4gRSBE
cmFrZSBbamRyYWtlQGp1bmlwZXIubmV0DQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIDxtYWlsdG86
amRyYWtlQGp1bmlwZXIubmV0Pl0NCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgKlNlbnQ6KiBUaHVy
c2RheSwgTWFyY2ggMjYsIDIwMTUgNzoxNg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICAqVG86KiBI
YW93ZWlndW87IEFsaSBTYWphc3NpIChzYWphc3NpKTsNCj4+Pj4+Pj4+YmVzc0BpZXRmLm9yZw0K
Pj4+Pj4+Pj4gICAgICAgICAgICAgICA8bWFpbHRvOmJlc3NAaWV0Zi5vcmc+DQo+Pj4+Pj4+PiAg
ICAgICAgICAgICAgICpTdWJqZWN0OiogUkU6IFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQRXMg
aW4gYW4NCj4+Pj4+Pj4+IEVWUE4gRVMNCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgYmFzZWQgb24g
UGF4b3MgYWxnb3JpdGhtDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBXZWlndW8s
DQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBZb3VyIHByb3Bvc2FsIGludHJvZHVj
ZXMgYSBjb250cm9sIHBsYW5lIHByb2Nlc3NpbmcNCj4+Pj4+Pj4+bG9hZA0KPj4+Pj4+Pj4gICAg
ICAgICAgICAgICB0aGF0IGlzIE8oI0VWSXMgKiBQRXMpIHBlciBERiBlbGVjdGlvbiBhbmQgZ2l2
ZW4NCj4+Pj4+Pj4+dGhhdA0KPj4+Pj4+Pj4gICAgICAgICAgICAgICB0aGVyZSBjYW4gYmUgNEsg
RVZJcyBwZXIgRVMsIHRoaXMgbG9va3MgbGlrZSBhDQo+Pj4+Pj4+PiAgICAgICAgICAgICAgICoq
c3Vic3RhbnRpYWwqKiBsb2FkLiBGdXJ0aGVybW9yZSwNCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiAgICAg
ICAgICAgICAgIHlvdSBjYW7CuXQgIHVzZSB0aGUgRVMgcm91dGUgdG8gY28tb3JkaW5hdGUgREYN
Cj4+Pj4+Pj4+ZWxlY3Rpb24NCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgYmVjYXVzZSB5b3Ugd291
bGQgbmVlZCB0byBjYXJyeSB5b3VyIG5ldyBleHRlbmRlZA0KPj4+Pj4+Pj4gICAgICAgICAgICAg
ICBjb21tdW5pdHkgZm9yIGVhY2ggRVZJIGFuZCB0aGV5IHdvdWxkIG5vdCBhbGwgZml0Lg0KPj4+
Pj4+Pj5Zb3UNCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgYWxzbyBjYW7CuXQgdXNlIHRoZSBQZXIg
RVZJIEV0aGVybmV0IEFEIHJvdXRlIGJlY2F1c2UNCj4+Pj4+Pj4+IHRoYXQNCj4+Pj4+Pj4+ICAg
ICAgICAgICAgICAgaXMgcHJvY2Vzc2VkIGJ5IGFsbCBQRXMgaW4gdGhlIEVWSS4NCj4+Pj4+Pj4+
DQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIEkgdGhpbmsgdGhhdCBmcm9tIGEgcHJhY3RpY2FsIHBl
cnNwZWN0aXZlIHRoZSBuZXcgREYNCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgZWxlY3Rpb24gcHJv
cG9zZWQgaW4gU2F0eWHCuXMgZHJhZnQgaXMgc3VmZmljaWVudGx5DQo+Pj4+Pj4+PiBzdGFibGUN
Cj4+Pj4+Pj4+ICAgICAgICAgICAgICAgdGhhdCBpdCByZW5kZXJzIHlvdXIgZHJhZnQgbW9vdCwg
ZXZlbiBpZiBpdCBjb3VsZCBiZQ0KPj4+Pj4+Pj4gbWFkZQ0KPj4+Pj4+Pj4gICAgICAgICAgICAg
ICB0byB3b3JrLg0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgWW91cnMgSXJyZXNw
ZWN0aXZlbHksDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBKb2huDQo+Pj4+Pj4+
Pg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICAqRnJvbToqQkVTUyBbbWFpbHRvOmJlc3MtYm91bmNl
c0BpZXRmLm9yZ10gKk9uDQo+Pj4+Pj4+PkJlaGFsZiBPZg0KPj4+Pj4+Pj4gICAgICAgICAgICAg
ICAqSGFvd2VpZ3VvDQo+Pj4+Pj4+PiAgICAgICAgICAgICAgICpTZW50OiogV2VkbmVzZGF5LCBN
YXJjaCAyNSwgMjAxNSA1OjM0IFBNDQo+Pj4+Pj4+PiAgICAgICAgICAgICAgICpUbzoqIEFsaSBT
YWphc3NpIChzYWphc3NpKTsgYmVzc0BpZXRmLm9yZw0KPj4+Pj4+Pj4gICAgICAgICAgICAgICA8
bWFpbHRvOmJlc3NAaWV0Zi5vcmc+DQo+Pj4+Pj4+PiAgICAgICAgICAgICAgICpTdWJqZWN0Oiog
UmU6IFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQRXMgaW4gYW4NCj4+Pj4+Pj4+IEVWUE4gRVMN
Cj4+Pj4+Pj4+ICAgICAgICAgICAgICAgYmFzZWQgb24gUGF4b3MgYWxnb3JpdGhtDQo+Pj4+Pj4+
Pg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBIaSBBbGksDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAg
ICAgICAgICAgICBUaGFua3MgZm9yIHlvdXIgaW5mb3JtYXRpb24uIEkgc2Nhbm5lZCB0aHJvdWdo
IHRoaXMNCj4+Pj4+Pj4+IGRyYWZ0LA0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBpdCByZWFsbHkg
aW50cm9kdWNlcyBpbnRlci1jaGFzc2lzIG1lc3NhZ2UgZm9yIERGDQo+Pj4+Pj4+PiBlbGVjdGlv
bg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBoYW5kc2hha2luZywgdGhlIHJlcXVpcmVtZW50cyBp
biB0aGlzIGRyYWZ0IGlzIHRvDQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIGVsaW1pbmF0ZSB0cmFu
c2lldCBMb29wIGFuZCB0cmFmZmljIGR1cGxpY2F0aW9uLg0KPj4+Pj4+Pj5DdXJyZW50DQo+Pj4+
Pj4+PiAgICAgICAgICAgICAgIEVWUE4gREYgZWxlY3Rpb24gbWVjaGFuaXNtIGFscmVhZHkgZWxp
bWluYXRlZCBsb29wDQo+Pj4+Pj4+PmFuZA0KPj4+Pj4+Pj4gICAgICAgICAgICAgICB0cmFmZmlj
IGR1cGxpY2F0aW9uIGJ5IGNvbmZpZ3VyaW5nIGxvbmcgcmVjZXB0aW9uDQo+Pj4+Pj4+PiB0aW1l
ciBvbg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBlYWNoIG11bHRpLWhvbWVkIFBFLCBidXQgdXAg
dG8gcmVjZXB0aW9uIHRpbWVyDQo+Pj4+Pj4+PnRyYWZmaWMNCj4+Pj4+Pj4+ICAgICAgICAgICAg
ICAgZGlzcnVwdGlvbiBpc3N1ZSBzdGlsbCBleGlzdC4gRVZQTiBmb3IgRENJIGlzIGFuDQo+Pj4+
Pj4+PiBpbXBvcnRhbnQNCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgdXNlIGNhc2UgZm9yIEVWUE4s
IHVwIHRvIHJlY2VwdGlvbiB0aW1lciB0cmFmZmljDQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIGRp
c3J1cHRpb24gY2FuJ3QgYmUgdG9sZXJhdGVkIGZvciBzZXJ2aWNlDQo+Pj4+Pj4+PnByb3ZpZGVy
cywgaXQNCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgc2hvdWxkIGJlIGltcHJvdmVkLg0KPj4+Pj4+
Pj4NCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgQWxzbyBmb3IgYWNjdXJhY3ksIGkgdGhpbmsgaGFu
ZHNoYWtpbmcgc3RhdGUNCj4+Pj4+Pj4+bWFjaGluZSBvbg0KPj4+Pj4+Pj4gICAgICAgICAgICAg
ICBlYWNoIG11bHRpLWhvbWVkIFBFIGlzIGFsc28gbmVlZGVkLiBGcm9tIHNvbHV0aW9uDQo+Pj4+
Pj4+PiAgICAgICAgICAgICAgIHBlcnNwZWN0aXZlLCBpbiBteSBkcmFmdCwgbm8gaW50ZXItY2hh
c3NpcyBtZXNzYWdlDQo+Pj4+Pj4+PmlzDQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIGludHJvZHVj
ZWQsIG9ubHkgb25lIG5ldyBleHRlbmRlZCBjb21tdW5pdHkgaXMNCj4+Pj4+Pj4+IGludHJvZHVj
ZWQsDQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIGkgdGhpbmsgdGhlIHByb2Nlc3MgaXMgY29tcGFy
YXRpdmVseSBzaW1wbGUgdGhhbg0KPj4+Pj4+Pj55b3VyDQo+Pj4+Pj4+PiAgICAgICAgICAgICAg
IGZvbGxvd2luZyBkcmFmdC4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIEN1cnJl
bnQgRVZQTiBERiBlbGVjdGlvbiBoYXMgc29tZSBkcmF3YmFja3MsIHNvDQo+Pj4+Pj4+PnRoZXJl
DQo+Pj4+Pj4+PiBhcmUNCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgdGhyZWUgbmV3IGRyYWZ0cyBh
Ym91dCBERiBlbGVjdGlvbiBlbWVyZ2VkLiBJIHRoaW5rDQo+Pj4+Pj4+PkJFU1MNCj4+Pj4+Pj4+
ICAgICAgICAgICAgICAgV0cgY2FuIGNvbnNpZGVyIHRoZXNlIHRocmVlIGRyYWZ0cyBpbiBnbG9i
YWwgdmlldywgYQ0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBzaW5nbGUsY29tcHJlaGVuc2l2ZSBu
ZXcgREYgZWxlY3Rpb24gZHJhZnQgaXMgaG9wZWQuDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAg
ICAgICAgICB0aGFua3MuDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBUaGFua3Ms
DQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICB3ZWlndW8NCj4+Pj4+Pj4+DQo+Pj4+
Pj4+Pg0KPj4+Pj4+Pj4gDQo+Pj4+Pj4+Pi0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t
LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0NCj4+Pj4+Pj4+IC0tLQ0KPj4+Pj4+
Pj4gLQ0KPj4+Pj4+Pj4gLQ0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+ICAgICAgICAgICAgICAgKkZyb206
KkJFU1MgW2Jlc3MtYm91bmNlc0BpZXRmLm9yZw0KPj4+Pj4+Pj4gICAgICAgICAgICAgICA8bWFp
bHRvOmJlc3MtYm91bmNlc0BpZXRmLm9yZz5dIG9uIGJlaGFsZiBvZiBBbGkNCj4+Pj4+Pj4+U2Fq
YXNzaQ0KPj4+Pj4+Pj4gICAgICAgICAgICAgICAoc2FqYXNzaSkgW3NhamFzc2lAY2lzY28uY29t
DQo+Pj4+Pj4+PjxtYWlsdG86c2FqYXNzaUBjaXNjby5jb20+XQ0KPj4+Pj4+Pj4gICAgICAgICAg
ICAgICAqU2VudDoqIFRodXJzZGF5LCBNYXJjaCAyNiwgMjAxNSAwOjIxDQo+Pj4+Pj4+PiAgICAg
ICAgICAgICAgICpUbzoqIGJlc3NAaWV0Zi5vcmcgPG1haWx0bzpiZXNzQGlldGYub3JnPg0KPj4+
Pj4+Pj4gICAgICAgICAgICAgICAqU3ViamVjdDoqIFtiZXNzXSBIYW5kc2hha2luZyBhbW9uZyBQ
RXMgaW4gYW4gRVZQTg0KPj4+Pj4+Pj5FUw0KPj4+Pj4+Pj4gICAgICAgICAgICAgICBiYXNlZCBv
biBQYXhvcyBhbGdvcml0aG0NCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiAgICAgICAgICAgICAgIEZZSS0g
Rmlyc3QgcHVibGlzaGVkIEp1bHkgNCwgMjAxMQ0KPj4+Pj4+Pj4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+
PiANCj4+Pj4+Pj4+aHR0cHM6Ly9kYXRhdHJhY2tlci5pZXRmLm9yZy9kb2MvZHJhZnQtc2FqYXNz
aS1sMnZwbi1ldnBuLXNlZ21lbnQtcg0KPj4+Pj4+Pj4gb3V0DQo+Pj4+Pj4+PiBlDQo+Pj4+Pj4+
PiAvDQo+Pj4+Pj4+Pg0KPj4+Pj4+Pj4gICAgICAgICAgICAgICAtQWxpDQo+Pj4+Pj4+Pg0KPj4+
Pj4+Pj4NCj4+Pj4+Pj4+DQo+Pj4+Pj4+PiBfX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fX19fX19fX19fXw0KPj4+Pj4+Pj4gQkVTUyBtYWlsaW5nIGxpc3QNCj4+Pj4+Pj4+IEJF
U1NAaWV0Zi5vcmcNCj4+Pj4+Pj4+IGh0dHBzOi8vd3d3LmlldGYub3JnL21haWxtYW4vbGlzdGlu
Zm8vYmVzcw0KPj4+Pj4+Pg0KPj4+Pj4+Pg0KPj4+Pj4+Pg0KPj4+Pj4+Pg0KPj4+Pj4+Pg0KPj4+
Pj4+Pg0KPj4+Pj4+Pg0KPj4+Pj4+Pg0KPj4+Pj4+Pg0KPj4+Pj4+Pg0KPj4+Pj4+Pg0KPj4+Pj4+
Pg0KPj4+Pj4+Pg0KPj4+Pj4+PiBfX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fX19fX19fX19fX18NCj4+PiBfX19fX19fX19fX19fXw0KPj4+Pj4+PiBfX18NCj4+Pj4+
Pj4gXw0KPj4+Pj4+PiBfDQo+Pj4+Pj4+IF9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fX19fX19fX19fXw0KPj4+Pj4+Pg0KPj4+Pj4+PiBDZSBtZXNzYWdlIGV0IHNlcyBwaWVj
ZXMgam9pbnRlcyBwZXV2ZW50IGNvbnRlbmlyIGRlcyBpbmZvcm1hdGlvbnMNCj4+Pj4+Pj4gY29u
ZmlkZW50aWVsbGVzIG91IHByaXZpbGVnaWVlcyBldCBuZSBkb2l2ZW50IGRvbmMgIHBhcyBldHJl
DQo+Pj4+Pj4+IGRpZmZ1c2VzLCBleHBsb2l0ZXMgb3UgY29waWVzIHNhbnMgYXV0b3Jpc2F0aW9u
LiBTaSB2b3VzIGF2ZXogcmVjdQ0KPj4+Pj4+PiBjZSBtZXNzYWdlIHBhciBlcnJldXIsIHZldWls
bGV6IGxlIHNpZ25hbGVyICBhIGwnZXhwZWRpdGV1ciBldCBsZQ0KPj4+Pj4+PiBkZXRydWlyZSBh
aW5zaSBxdWUgbGVzIHBpZWNlcyBqb2ludGVzLiBMZXMgbWVzc2FnZXMgZWxlY3Ryb25pcXVlcw0K
Pj4+Pj4+PiBldGFudCBzdXNjZXB0aWJsZXMgZCdhbHRlcmF0aW9uLCAgT3JhbmdlIGRlY2xpbmUg
dG91dGUNCj4+Pj4+Pj4gcmVzcG9uc2FiaWxpdGUgc2kgY2UgbWVzc2FnZSBhIGV0ZSBhbHRlcmUs
IGRlZm9ybWUgb3UgZmFsc2lmaWUuDQo+Pj4+Pj4+IE1lcmNpLg0KPj4+Pj4+Pg0KPj4+Pj4+PiBU
aGlzIG1lc3NhZ2UgYW5kIGl0cyBhdHRhY2htZW50cyBtYXkgY29udGFpbiBjb25maWRlbnRpYWwg
b3INCj4+Pj4+Pj4gcHJpdmlsZWdlZCBpbmZvcm1hdGlvbiB0aGF0IG1heSBiZSBwcm90ZWN0ZWQg
YnkgbGF3OyAgdGhleSBzaG91bGQNCj4+Pj4+Pj4gbm90IGJlIGRpc3RyaWJ1dGVkLCB1c2VkIG9y
IGNvcGllZCB3aXRob3V0IGF1dGhvcmlzYXRpb24uDQo+Pj4+Pj4+IElmIHlvdSBoYXZlIHJlY2Vp
dmVkIHRoaXMgZW1haWwgaW4gZXJyb3IsIHBsZWFzZSBub3RpZnkgdGhlIHNlbmRlcg0KPj4+Pj4+
PiBhbmQgZGVsZXRlIHRoaXMgbWVzc2FnZSBhbmQgaXRzIGF0dGFjaG1lbnRzLg0KPj4+Pj4+PiBB
cyBlbWFpbHMgbWF5IGJlIGFsdGVyZWQsIE9yYW5nZSBpcyBub3QgbGlhYmxlIGZvciBtZXNzYWdl
cyB0aGF0DQo+Pj4+Pj4+IGhhdmUgYmVlbiBtb2RpZmllZCwgY2hhbmdlZCBvciBmYWxzaWZpZWQu
DQo+Pj4+Pj4+IFRoYW5rIHlvdS4NCj4+Pj4+Pj4NCj4+Pj4+Pj4gX19fX19fX19fX19fX19fX19f
X19fX19fX19fX19fX19fX19fX19fX19fX19fX18NCj4+Pj4+Pj4gQkVTUyBtYWlsaW5nIGxpc3QN
Cj4+Pj4+Pj4gQkVTU0BpZXRmLm9yZw0KPj4+Pj4+PiBodHRwczovL3d3dy5pZXRmLm9yZy9tYWls
bWFuL2xpc3RpbmZvL2Jlc3MNCj4+Pj4+Pj4NCj4+Pj4+Pg0KPj4+Pj4+IF9fX19fX19fX19fX19f
X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18NCj4+PiBfX19fX19fX19f
X19fXw0KPj4+Pj4+IF9fXw0KPj4+Pj4+IF8NCj4+Pj4+PiBfDQo+Pj4+Pj4gX19fX19fX19fX19f
X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18NCj4+Pj4+Pg0KPj4+Pj4+IENlIG1l
c3NhZ2UgZXQgc2VzIHBpZWNlcyBqb2ludGVzIHBldXZlbnQgY29udGVuaXIgZGVzIGluZm9ybWF0
aW9ucw0KPj4+Pj4+IGNvbmZpZGVudGllbGxlcyBvdSBwcml2aWxlZ2llZXMgZXQgbmUgZG9pdmVu
dCBkb25jIHBhcyBldHJlDQo+Pj4+Pj5kaWZmdXNlcywNCj4+Pj4+PiBleHBsb2l0ZXMgb3UgY29w
aWVzIHNhbnMgYXV0b3Jpc2F0aW9uLiBTaSB2b3VzIGF2ZXogcmVjdSBjZSBtZXNzYWdlDQo+Pj4+
Pj4gcGFyIGVycmV1ciwgdmV1aWxsZXogbGUgc2lnbmFsZXIgYSBsJ2V4cGVkaXRldXIgZXQgbGUg
ZGV0cnVpcmUgYWluc2kNCj4+Pj4+PiBxdWUgbGVzIHBpZWNlcyBqb2ludGVzLiBMZXMgbWVzc2Fn
ZXMgZWxlY3Ryb25pcXVlcyBldGFudA0KPj4+Pj4+c3VzY2VwdGlibGVzDQo+Pj4+Pj4gZCdhbHRl
cmF0aW9uLCBPcmFuZ2UgZGVjbGluZSB0b3V0ZSByZXNwb25zYWJpbGl0ZSBzaSBjZSBtZXNzYWdl
IGENCj4+Pj4+PmV0ZQ0KPj4+Pj4+IGFsdGVyZSwgZGVmb3JtZSBvdSBmYWxzaWZpZS4gTWVyY2ku
DQo+Pj4+Pj4NCj4+Pj4+PiBUaGlzIG1lc3NhZ2UgYW5kIGl0cyBhdHRhY2htZW50cyBtYXkgY29u
dGFpbiBjb25maWRlbnRpYWwgb3INCj4+Pj4+PiBwcml2aWxlZ2VkIGluZm9ybWF0aW9uIHRoYXQg
bWF5IGJlIHByb3RlY3RlZCBieSBsYXc7IHRoZXkgc2hvdWxkIG5vdA0KPj4+Pj4+IGJlIGRpc3Ry
aWJ1dGVkLCB1c2VkIG9yIGNvcGllZCB3aXRob3V0IGF1dGhvcmlzYXRpb24uDQo+Pj4+Pj4gSWYg
eW91IGhhdmUgcmVjZWl2ZWQgdGhpcyBlbWFpbCBpbiBlcnJvciwgcGxlYXNlIG5vdGlmeSB0aGUg
c2VuZGVyDQo+Pj4+Pj4gYW5kIGRlbGV0ZSB0aGlzIG1lc3NhZ2UgYW5kIGl0cyBhdHRhY2htZW50
cy4NCj4+Pj4+PiBBcyBlbWFpbHMgbWF5IGJlIGFsdGVyZWQsIE9yYW5nZSBpcyBub3QgbGlhYmxl
IGZvciBtZXNzYWdlcyB0aGF0DQo+Pj4+Pj5oYXZlDQo+Pj4+Pj4gYmVlbiBtb2RpZmllZCwgY2hh
bmdlZCBvciBmYWxzaWZpZWQuDQo+Pj4+Pj4gVGhhbmsgeW91Lg0KPg0KPg0KPl9fX19fX19fX19f
X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fDQo+X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18N
Cj4NCj5DZSBtZXNzYWdlIGV0IHNlcyBwaWVjZXMgam9pbnRlcyBwZXV2ZW50IGNvbnRlbmlyIGRl
cyBpbmZvcm1hdGlvbnMNCj5jb25maWRlbnRpZWxsZXMgb3UgcHJpdmlsZWdpZWVzIGV0IG5lIGRv
aXZlbnQgZG9uYw0KPnBhcyBldHJlIGRpZmZ1c2VzLCBleHBsb2l0ZXMgb3UgY29waWVzIHNhbnMg
YXV0b3Jpc2F0aW9uLiBTaSB2b3VzIGF2ZXoNCj5yZWN1IGNlIG1lc3NhZ2UgcGFyIGVycmV1ciwg
dmV1aWxsZXogbGUgc2lnbmFsZXINCj5hIGwnZXhwZWRpdGV1ciBldCBsZSBkZXRydWlyZSBhaW5z
aSBxdWUgbGVzIHBpZWNlcyBqb2ludGVzLiBMZXMgbWVzc2FnZXMNCj5lbGVjdHJvbmlxdWVzIGV0
YW50IHN1c2NlcHRpYmxlcyBkJ2FsdGVyYXRpb24sDQo+T3JhbmdlIGRlY2xpbmUgdG91dGUgcmVz
cG9uc2FiaWxpdGUgc2kgY2UgbWVzc2FnZSBhIGV0ZSBhbHRlcmUsIGRlZm9ybWUNCj5vdSBmYWxz
aWZpZS4gTWVyY2kuDQo+DQo+VGhpcyBtZXNzYWdlIGFuZCBpdHMgYXR0YWNobWVudHMgbWF5IGNv
bnRhaW4gY29uZmlkZW50aWFsIG9yIHByaXZpbGVnZWQNCj5pbmZvcm1hdGlvbiB0aGF0IG1heSBi
ZSBwcm90ZWN0ZWQgYnkgbGF3Ow0KPnRoZXkgc2hvdWxkIG5vdCBiZSBkaXN0cmlidXRlZCwgdXNl
ZCBvciBjb3BpZWQgd2l0aG91dCBhdXRob3Jpc2F0aW9uLg0KPklmIHlvdSBoYXZlIHJlY2VpdmVk
IHRoaXMgZW1haWwgaW4gZXJyb3IsIHBsZWFzZSBub3RpZnkgdGhlIHNlbmRlciBhbmQNCj5kZWxl
dGUgdGhpcyBtZXNzYWdlIGFuZCBpdHMgYXR0YWNobWVudHMuDQo+QXMgZW1haWxzIG1heSBiZSBh
bHRlcmVkLCBPcmFuZ2UgaXMgbm90IGxpYWJsZSBmb3IgbWVzc2FnZXMgdGhhdCBoYXZlIA0KPmJl
ZW4gbW9kaWZpZWQsIGNoYW5nZWQgb3IgZmFsc2lmaWVkLg0KPlRoYW5rIHlvdS4NCj4NCj5fX19f
X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXw0KPkJFU1MgbWFpbGlu
ZyBsaXN0DQo+QkVTU0BpZXRmLm9yZw0KPmh0dHBzOi8vd3d3LmlldGYub3JnL21haWxtYW4vbGlz
dGluZm8vYmVzcw0KDQo=


From nobody Fri Mar 27 08:54:54 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 91A0B1A066B for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:54:53 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.011
X-Spam-Level: 
X-Spam-Status: No, score=-3.011 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_32=0.6, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id sEbt1yW2xHt9 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 08:54:48 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9A35C1A88BF for <bess@ietf.org>; Fri, 27 Mar 2015 08:54:47 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BUF20646; Fri, 27 Mar 2015 15:54:45 +0000 (GMT)
Received: from nkgeml409-hub.china.huawei.com (10.98.56.40) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 15:54:44 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml409-hub.china.huawei.com ([10.98.56.40]) with mapi id 14.03.0158.001; Fri, 27 Mar 2015 23:54:37 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HYAABS9Z9AAD2dSgAAN/i8VACwBuAAARy7QAwCeks8AASwBiMg=
Date: Fri, 27 Mar 2015 15:54:36 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F8577F2@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com>, <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com>, <BLUPR05MB562504AAF732AF476E5C218C7090@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8577BC@nkgeml501-mbs.china.huawei.com>, <21658_1427470813_551579DD_21658_1989_1_551579DA.7060102@orange.com>
In-Reply-To: <21658_1427470813_551579DD_21658_1989_1_551579DA.7060102@orange.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.144.149]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/-lnNRmafwKzy3EShPycNmUORZSI>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 15:54:53 -0000

Hi Thomas,
Thanks for your reference.
Yes,if a bridged network does not connect to the PEs using a LAG, in stable=
 state, only one link connect to one PE is active. But in transiet state, b=
oth PE1 and PE2 can forward traffic, so loop will occur.

We are talking the process in transiet period, i think in this period traff=
ic loop is not allowed.
Thanks,
weiguo

________________________________________
From: BESS [bess-bounces@ietf.org] on behalf of thomas.morin@orange.com [th=
omas.morin@orange.com]
Sent: Friday, March 27, 2015 23:40
To: bess@ietf.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

HI Weigo, all,

Let me try to help again...

RFC7432, section 8.5 says:

    If a bridged network is multihomed to more than one PE in an EVPN
    network via switches, then the support of All-Active redundancy mode
    requires the bridged network to be connected to two or more PEs using
    a LAG.

    If a bridged network does not connect to the PEs using a LAG, then
    only one of the links between the bridged network and the PEs must be
    the active link for a given <ES, VLAN> or <ES, VLAN bundle>.  In this
    case, the set of Ethernet A-D per ES routes advertised by each PE
    MUST have the "Single-Active" bit in the flags of the ESI Label
    extended community set to 1.

My understanding was that the above excludes the kind of topology in
which you exclude a possible loop.

-Thomas



Weiguo :
> Hi John,
> If you can tolerate indefinite forwarding loop during transiet time, i ha=
ve no problem. If the community think it is a serious problem, i think we m=
ust propose an applicable solution for it, maybe there are other solutions =
better than handshaking mechanism.
>
> Thanks,
> weiguo
>
> ________________________________________
> From: John E Drake [jdrake@juniper.net]
> Sent: Friday, March 27, 2015 23:22
> To: Haoweiguo; Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT - th=
omas.morin@orange.com; bess@ietf.org
> Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos al=
gorithm
>
> I think we have beaten this topic to death.  For the numerous reason I've=
 given, I'm not interested in you proposal.
>
> Yours Irrespectively,
>
> John
>
>> -----Original Message-----
>> From: BESS [mailto:bess-bounces@ietf.org] On Behalf Of Haoweiguo
>> Sent: Friday, March 27, 2015 11:12 AM
>> To: Rabadan, Jorge (Jorge); Satya Mohanty (satyamoh); EXT -
>> thomas.morin@orange.com; bess@ietf.org
>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>> algorithm
>>
>> Hi Jorge,
>> Let give you another indefinitely forwarding loop case, the picture is a=
s
>> follows:
>>
>> CE3------------CE4
>>    |                        |
>> PE3                   PE4
>>
>>
>>
>> PE1                   PE2
>>   |                         |
>> CE1------------CE2
>>
>> If PE1 and PE2 are dual DF PE in transiet time, assuing PE3 and PE4 are =
in
>> stable state ,PE3 is DF PE, PE4 is non-DF PE, the BUM traffic from CE3 t=
o CE1,
>> the forwarding loop path is as follows;
>> CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->PE4--=
--
>>> forever......
>> Do you think the harm is enough?
>> Thanks,
>> weiguo
>> ________________________________________
>> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>> Sent: Friday, March 27, 2015 22:59
>> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>> bess@ietf.org
>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>> algorithm
>>
>> Hi Weiguo,
>>
>> OK, thanks for clarifying, that is the part I did not understand in your=
 loop :-)
>>
>> To me a loop means you send a frame and that frame keeps being forwarded
>> by the same nodes indefinitely till you take a corrective action.
>> In this case, CE3 might get a few packets during the transient state, bu=
t those
>> packets will have its own MAC SA so CE3 will discard them. I don't think=
 that is
>> harmful for CE3 in most of the cases. And if it is, you just need to mak=
e sure
>> the timers are long enough and the DF election properly implemented to
>> avoid the transient DF-DF state.
>>
>> The handshaking is definitely overkilling IMHO.
>>
>> Thanks.
>> Jorge
>>
>> -----Original Message-----
>> From: Haoweiguo <haoweiguo@huawei.com>
>> Date: Friday, March 27, 2015 at 7:29 AM
>> To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
>> (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
>> <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>> algorithm
>>
>>> Hi Jorge,
>>> Sorry, maybe i have not explain clearly, pls allow me to clarify it
>>> more clearly.
>>> In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2
>>> is single homed to remote PE3. CE1 and CE2 forms one layer 2 network,
>>> so it multi-homed network scenario.
>>>            CE3
>>>              |
>>>             PE3
>>>
>>>
>>> PE1                  PE2
>>>   |                       |
>>> CE1-----------CE2
>>>
>>> In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends
>>> BUM traffic to CE1, the traffic will reach PE1 and PE2, CE1 will
>>> receive two copy of the traffic from CE3, also loop is formed, the form=
 path:
>>>
>>> CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
>>>
>>> I think the traffic loop is absolutely not allowed. In MHN case, the
>>> loop will formed due to dual-DF PE in transiet period.
>>>
>>> Thanks,
>>> weiguo
>>>
>>>
>>>
>>> ________________________________________
>>> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>>> Sent: Friday, March 27, 2015 20:26
>>> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>>> bess@ietf.org
>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>> algorithm
>>>
>>> Hi Weiguo,
>>>
>>> About this:
>>>
>>> [weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>> case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
>>> and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>>> loop back to CE1 through PE2. If CE1 can filter and drop the traffic ,
>>> it is fortunate, maybe no loop.
>>> For multi-homed network scenario, say local network1(CE1 and CE2) is
>>> multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>> back through PE1-->PE2-->CE2, it is very serious. It is absolutely not =
allowed.
>>>
>>>
>>> [JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2
>>> will not send the traffic back to CE1 due to the ESI-label and
>>> split-horizon procedures.
>>> In MH network case PE2 will NOT send the traffic to CE2 for the same
>>> reason.
>>>
>>> Let me know if you disagree please.
>>>
>>> Thanks.
>>> Jorge
>>>
>>> -----Original Message-----
>>> From: Haoweiguo <haoweiguo@huawei.com>
>>> Date: Thursday, March 26, 2015 at 9:55 PM
>>> To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
>>> "thomas.morin@orange.com" <thomas.morin@orange.com>,
>> "bess@ietf.org"
>>> <bess@ietf.org>
>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>> algorithm
>>>
>>>> Hi Satya,
>>>> Pls see inline.
>>>>
>>>> Thanks,
>>>> weiguo
>>>>
>>>> ________________________________________
>>>> From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
>>>> Sent: Friday, March 27, 2015 12:11
>>>> To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>> algorithm
>>>>
>>>> Thomas,
>>>>
>>>> Thanks for your summary. It captures the essence of the discussion.
>>>> [weiguo]: Yes, i agree.
>>>>
>>>> One feature of the HRW scheme is that it computes the Active and the
>>>> Backup DF upfront.
>>>> Whenever the Active PE, say PE1, dies; the Backup PE, PE2, can become
>>>> the Active immediately upon receiving the withdraw of the ES route
>>> >from PE1.
>>>> It need not even have to wait for the 3 seconds time.
>>>> [weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
>>>> message:) But when PE1 leaves ESI, it can withdraw the ES route
>>>> immediately.
>>>> When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2
>>>> will both act as PE for some transiet time.
>>>>
>>>> On the other hand, the 3 second timer will serve as a bulk and batch
>>>> mechanism for the case when new PE(s) is(are) coming up, so we don=B9t
>>>> have to do a DF election ever so often.
>>>>
>>>> Haoweiguo,  I think the "loop" scenario is speculative.
>>>> I think everybody understands that, yes, there may be some transient
>>>> duplication of bum traffic; however, there is no "loop" problem to be
>>>> solved.
>>>> [weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>>> case. In multi-homed device case, say CE1 is multi-homed to PE1 and
>>>> PE2 and both PE are DF PE. When CE1 sends traffic to PE1, the traffic
>>>> will loop back to CE1 through PE2. If CE1 can filter and drop the
>>>> traffic , it is fortunate, maybe no loop.
>>>> For multi-homed network scenario, say local network1(CE1 and CE2) is
>>>> multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>>> back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
>>>> allowed.
>>>>
>>>> Thanks,
>>>> --Satya
>>>>
>>>>
>>>>
>>>> On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
>>>>
>>>>> Hi Thomas,
>>>>> I know you are neutral, sorry for my reply vulnerable to be
>>>>> miunderstood.
>>>>>
>>>>> Thanks,
>>>>> weiguo
>>>>>
>>>>> ________________________________________
>>>>> From: thomas.morin@orange.com [thomas.morin@orange.com]
>>>>> Sent: Friday, March 27, 2015 11:59
>>>>> To: Haoweiguo; bess@ietf.org
>>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on
>>>>> Paxos algorithm
>>>>>
>>>>> Hi Weiguo,
>>>>>
>>>>> 2015-03-26, Haoweiguo:
>>>>>> Thomas:
>>>>>>> - RFC7432 may have transient periods where the DF election state
>>>>>>> is not yet synchronized between the two peers:
>>>>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic
>>>>>> loop
>>>>> Note well that I didn't write that there can be transient _loops_ .
>>>>> I tried to capture the exchange you had all, and what I gather is
>>>>> that the split-horizon procedure _prevents_loops_, independently of
>>>>> any transient period where DF state is not synchronized and which may
>>>>> lead to transient _duplicates_.
>>>>>
>>>>> -Thomas
>>>>>
>>>>>> 26/03/2015 20:05, John E Drake :
>>>>>>> Weiguo,
>>>>>>>
>>>>>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I hav=
e
>>>>>>> detailed, is a non-solution to a non-problem with tremendous
>>>>>>> control plane cost.
>>>>>>>
>>>>>>> Yours Irrespectively,
>>>>>>>
>>>>>>> John
>>>>>>>
>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> Pls see below.
>>>>>>>
>>>>>>> Thanks,
>>>>>>>
>>>>>>> weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>> *Sent:* Friday, March 27, 2015 6:00
>>>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> To recap,
>>>>>>>
>>>>>>> We have established that your proposal is untenable because of its
>>>>>>> control plane load.
>>>>>>>
>>>>>>> We have established that your proposal is based upon a flawed
>>>>>>> understanding of the DF election in RFC 7432.
>>>>>>>
>>>>>>> [weiguo]: In ethernet world, traffic loop is serious than short
>>>>>>> timer traffic disruption. If you want to implement  transiet
>>>>>>> traffic loop process, i will modify my draft to solve your issue.
>>>>>>>
>>>>>>> If i am the developer, i will prefer short timer traffic
>>>>>>> disruption based on current EVPN protocol.
>>>>>>>
>>>>>>> What you are now arguing is that your draft prevents two or more
>>>>>>> PEs from being DF simultaneously. This is clearly nonsense.
>>>>>>>
>>>>>>> [weiguo]: I will modify the draft problem statements, and use the
>>>>>>> same  handshaking solution to solve it.
>>>>>>>
>>>>>>> Furthermore, we have established that having two or more DFs for
>>>>>>> what even you admit is a brief transient leads to duplicate
>>>>>>> traffic, which is acceptable, but not loops, your assertion to the
>> contrary.
>>>>>>> [weiguo]: It is transient loop and traffic duplication issue.
>>>>>>>
>>>>>>> Yours Irrespectively,
>>>>>>>
>>>>>>> John
>>>>>>>
>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> John,
>>>>>>>
>>>>>>> As your understanding of the EVPN draft, the DF election mechanism
>>>>>>> has  more serious side effect, it will have short time traffic
>>>>>>> loop,i.e.,  dual DF PEs will exist for a short time. I think dual
>>>>>>> DF PEs is  absolutely not tolerated, because native ethernet header
>>>>>>> has no TTL,  up to several hundred ms traffic loop normally not
>>>>>>> tolerated in  commertial networks.
>>>>>>>
>>>>>>> As your understanding, the PEs should do as following:
>>>>>>>
>>>>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but
>>>>>>> have rarely deployment.
>>>>>>>
>>>>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3
>>>>>>> joins ESI and trigger DF re-election. When reception timer expires:
>>>>>>>
>>>>>>> PE1 upgrades to DF PE.
>>>>>>>
>>>>>>> After reception timer+ ES route transmission timer:
>>>>>>>
>>>>>>> PE2 downloads to non-DF PE.
>>>>>>>
>>>>>>> So in timer clock sync case, dual DF PEs will exist at least
>>>>>>> transmission timer.
>>>>>>>
>>>>>>> If NTP is used for timer sync, because it has bad accuracy, dual
>>>>>>> DF PEs will exist more longer timer.
>>>>>>>
>>>>>>> So as your understanding for DF election, the drawback is more clea=
r.
>>>>>>>
>>>>>>> Thanks,
>>>>>>>
>>>>>>> weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>> *Sent:* Friday, March 27, 2015 4:41
>>>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> Weiguo,
>>>>>>>
>>>>>>> We have already established that your proposal is untenable
>>>>>>> because of  its control plane load.
>>>>>>>
>>>>>>> What we are now discussing is that your proposal is based upon a
>>>>>>> misunderstanding of the algorithm in RFC 7432.  You are assuming
>>>>>>> that
>>>>>>> PE1 will advertise an ES route and then wait for the configured
>>>>>>> interval before performing the DF election while PE2 and PE3 will
>>>>>>> perform the DF election as soon as they receive the ES route from
>>>>>>> PE1.
>>>>>>> This is not what RFC 7432 says.
>>>>>>>
>>>>>>> Rather, what is says is that the advertisement of the ES route by
>>>>>>> PE1 and its receipt by PE2 and PE3 causes all three PEs to start
>>>>>>> the configured interval timer  - =B33. When the timer expires, each
>>>>>>> PE builds an ordered list of the IP addresses of all the PE nodes
>>>>>>> connected to the Ethernet segment (including itself), in
>>>>>>> increasing numeric value.=B2
>>>>>>>
>>>>>>> Yours Irrespectively,
>>>>>>>
>>>>>>> John
>>>>>>>
>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> Pls read my detail replies to Satya. If you still can't catch it,
>>>>>>> pls read my draft and EVPN base protocol,  thanks
>>>>>>>
>>>>>>> weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>> *Sent:* Friday, March 27, 2015 1:28
>>>>>>> *To:* Patrice Brissette (pbrisset)
>>>>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>>> <mailto:bess@ietf.org>
>>>>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>> Paxos algorithm
>>>>>>>
>>>>>>> I think Patrice is correct.  Your proposal doesn't solve the
>>>>>>> problem and it does so at huge cost.
>>>>>>>
>>>>>>> Sent from my iPhone
>>>>>>>
>>>>>>>
>>>>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>>>>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>>>>>>
>>>>>>>       Weiguo,
>>>>>>>
>>>>>>>       I=B9m not sure I=B9m following here.
>>>>>>>
>>>>>>>       Don=B9t you have the same issue with your handshaking mechani=
sm?
>>>>>>>
>>>>>>>       If you don=B9t know your peer, how can you handshake?
>>>>>>>
>>>>>>>       Regards,
>>>>>>>
>>>>>>>       Patrice
>>>>>>>
>>>>>>>       Image removed by sender.
>>>>>>>
>>>>>>>       *Patrice Brissette*
>>>>>>>       TECHNICAL LEADER.ENGINEERING
>>>>>>>
>>>>>>>       pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>>>       Phone: *+1 613 254 3336*
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>       *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>>>       Canada
>>>>>>>       Cisco.com <http://www.cisco.com/global/CA/>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>       Image removed by sender.Think before you print.
>>>>>>>
>>>>>>>       This email may contain confidential and privileged material f=
or
>>>>>>>       the sole use of the intended recipient. Any review, use,
>>>>>>>       distribution or disclosure by others is strictly prohibited. =
If
>>>>>>>       you are not the intended recipient (or authorized to receive =
for
>>>>>>>       the recipient), please contact the sender by reply email and
>>>>>>>       delete all copies of this message.
>>>>>>>
>>>>>>>       Please click here
>>>>>>>
>>>>>>> <http://www.cisco.com/web/about/doing_business/legal/cri/index.h
>> tml>
>>>>>>>       for Company Registration Information.
>>>>>>>
>>>>>>>
>>>>>>>       *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>>> <mailto:haoweiguo@huawei.com>>
>>>>>>>       *Date: *Thursday, March 26, 2015 at 10:24 AM
>>>>>>>       *To: *Patrice Brissette <pbrisset@cisco.com
>>>>>>>       <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.ne=
t
>>>>>>>       <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>>>>       <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>>>       <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>=
>
>>>>>>>       *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES
>>>>>>> based on
>>>>>>>       Paxos algorithm
>>>>>>>
>>>>>>>           Hi Patrice,
>>>>>>>
>>>>>>>           Up to reception timer traffic disruption in transient
>>>>>>> phase is
>>>>>>>           one of the issues.
>>>>>>>
>>>>>>>           Thanks,
>>>>>>>
>>>>>>>           weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>>           *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>>>>>>           <mailto:pbrisset@cisco.com>]
>>>>>>>           *Sent:* Thursday, March 26, 2015 20:54
>>>>>>>           *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>>>>>>           bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>           *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>>>           based on Paxos algorithm
>>>>>>>
>>>>>>>           Weiguo,
>>>>>>>
>>>>>>>           You mention "But if your draft have not solved all
>>>>>>> issues=B2,
>>>>>>>
>>>>>>>           Can you explain what Satya=B9s draft is not solving?
>>>>>>>
>>>>>>>           Regards,
>>>>>>>
>>>>>>>           Patrice
>>>>>>>
>>>>>>>           Image removed by sender.
>>>>>>>
>>>>>>>           *Patrice Brissette*
>>>>>>>           TECHNICAL LEADER.ENGINEERING
>>>>>>>
>>>>>>>           pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>>>           Phone: *+1 613 254 3336*
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>           *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE=
*
>>>>>>>           Canada
>>>>>>>           Cisco.com <http://www.cisco.com/global/CA/>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>           Image removed by sender.Think before you print.
>>>>>>>
>>>>>>>           This email may contain confidential and privileged materi=
al
>>>>>>>           for the sole use of the intended recipient. Any review, u=
se,
>>>>>>>           distribution or disclosure by others is strictly prohibit=
ed.
>>>>>>>           If you are not the intended recipient (or authorized to
>>>>>>>           receive for the recipient), please contact the sender by
>>>>>>> reply
>>>>>>>           email and delete all copies of this message.
>>>>>>>
>>>>>>>           Please click here
>>>>>>>
>>>>>>> <http://www.cisco.com/web/about/doing_business/legal/cri/index.h
>> tml>
>>>>>>>           for Company Registration Information.
>>>>>>>
>>>>>>>
>>>>>>>           *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>>>           <mailto:haoweiguo@huawei.com>>
>>>>>>>           *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>>>>>>           *To: *John E Drake <jdrake@juniper.net
>>>>>>>           <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.=
com
>>>>>>>           <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>>>           <mailto:bess@ietf.org>" <bess@ietf.org
>>>>>>> <mailto:bess@ietf.org>>
>>>>>>>           *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>>>           based on Paxos algorithm
>>>>>>>
>>>>>>>               Hi John,
>>>>>>>
>>>>>>>                Firstly i think EVPN community should reach
>>>>>>> consensus on
>>>>>>>               the issues of current DF election mechanism. All thes=
e
>>>>>>>               issues should be resolved in a single new DF election
>>>>>>>               draft,rather than in multiple separate drafts. If you=
r
>>>>>>>               draft can solve all these issues and stable, i have n=
o
>>>>>>>               question for its progressing. But if your draft have =
not
>>>>>>>               solved all issues, i think it had better combine with
>>>>>>>               other drafts to provide a comprehensive solution. I
>>>>>>> think
>>>>>>>               the issues listed in draft-rabadan-bess-evpn-ac-df-01
>>>>>>> and
>>>>>>>               draft-hao-bess-evpn-df-handshaking-00 is valid, it
>>>>>>> should
>>>>>>>               be resolved. So i think although your new Hash algori=
thm
>>>>>>>               for DF election is good, it only includes partial
>>>>>>>               enhancements, maybe it still needs some time for
>>>>>>> consensus.
>>>>>>>
>>>>>>>               Thanks,
>>>>>>>
>>>>>>>               weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>>               *From:*John E Drake [jdrake@juniper.net
>>>>>>>               <mailto:jdrake@juniper.net>]
>>>>>>>               *Sent:* Thursday, March 26, 2015 7:16
>>>>>>>               *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>               <mailto:bess@ietf.org>
>>>>>>>               *Subject:* RE: [bess] Handshaking among PEs in an
>>>>>>> EVPN ES
>>>>>>>               based on Paxos algorithm
>>>>>>>
>>>>>>>               Weiguo,
>>>>>>>
>>>>>>>               Your proposal introduces a control plane processing l=
oad
>>>>>>>               that is O(#EVIs * PEs) per DF election and given that
>>>>>>>               there can be 4K EVIs per ES, this looks like a
>>>>>>>               **substantial** load. Furthermore,
>>>>>>>
>>>>>>>               you can=B9t  use the ES route to co-ordinate DF elect=
ion
>>>>>>>               because you would need to carry your new extended
>>>>>>>               community for each EVI and they would not all fit.  Y=
ou
>>>>>>>               also can=B9t use the Per EVI Ethernet AD route becaus=
e
>>>>>>> that
>>>>>>>               is processed by all PEs in the EVI.
>>>>>>>
>>>>>>>               I think that from a practical perspective the new DF
>>>>>>>               election proposed in Satya=B9s draft is sufficiently
>>>>>>> stable
>>>>>>>               that it renders your draft moot, even if it could be
>>>>>>> made
>>>>>>>               to work.
>>>>>>>
>>>>>>>               Yours Irrespectively,
>>>>>>>
>>>>>>>               John
>>>>>>>
>>>>>>>               *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf=
 Of
>>>>>>>               *Haoweiguo
>>>>>>>               *Sent:* Wednesday, March 25, 2015 5:34 PM
>>>>>>>               *To:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>               <mailto:bess@ietf.org>
>>>>>>>               *Subject:* Re: [bess] Handshaking among PEs in an
>>>>>>> EVPN ES
>>>>>>>               based on Paxos algorithm
>>>>>>>
>>>>>>>               Hi Ali,
>>>>>>>
>>>>>>>               Thanks for your information. I scanned through this
>>>>>>> draft,
>>>>>>>               it really introduces inter-chassis message for DF
>>>>>>> election
>>>>>>>               handshaking, the requirements in this draft is to
>>>>>>>               eliminate transiet Loop and traffic duplication. Curr=
ent
>>>>>>>               EVPN DF election mechanism already eliminated loop an=
d
>>>>>>>               traffic duplication by configuring long reception
>>>>>>> timer on
>>>>>>>               each multi-homed PE, but up to reception timer traffi=
c
>>>>>>>               disruption issue still exist. EVPN for DCI is an
>>>>>>> important
>>>>>>>               use case for EVPN, up to reception timer traffic
>>>>>>>               disruption can't be tolerated for service providers, =
it
>>>>>>>               should be improved.
>>>>>>>
>>>>>>>               Also for accuracy, i think handshaking state machine =
on
>>>>>>>               each multi-homed PE is also needed. From solution
>>>>>>>               perspective, in my draft, no inter-chassis message is
>>>>>>>               introduced, only one new extended community is
>>>>>>> introduced,
>>>>>>>               i think the process is comparatively simple than your
>>>>>>>               following draft.
>>>>>>>
>>>>>>>               Current EVPN DF election has some drawbacks, so there
>>>>>>> are
>>>>>>>               three new drafts about DF election emerged. I think B=
ESS
>>>>>>>               WG can consider these three drafts in global view, a
>>>>>>>               single,comprehensive new DF election draft is hoped.
>>>>>>>
>>>>>>>               thanks.
>>>>>>>
>>>>>>>               Thanks,
>>>>>>>
>>>>>>>               weiguo
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------------
>>>>>>> ---
>>>>>>> -
>>>>>>> -
>>>>>>>
>>>>>>>               *From:*BESS [bess-bounces@ietf.org
>>>>>>>               <mailto:bess-bounces@ietf.org>] on behalf of Ali Saja=
ssi
>>>>>>>               (sajassi) [sajassi@cisco.com <mailto:sajassi@cisco.co=
m>]
>>>>>>>               *Sent:* Thursday, March 26, 2015 0:21
>>>>>>>               *To:* bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>               *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>>>>>>               based on Paxos algorithm
>>>>>>>
>>>>>>>               FYI- First published July 4, 2011
>>>>>>>
>>>>>>>
>>>>>>> https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-r
>>>>>>> out
>>>>>>> e
>>>>>>> /
>>>>>>>
>>>>>>>               -Ali
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> BESS mailing list
>>>>>>> BESS@ietf.org
>>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> ______________________________________________________
>> ______________
>>>>>> ___
>>>>>> _
>>>>>> _
>>>>>> ________________________________________________
>>>>>>
>>>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>>>> confidentielles ou privilegiees et ne doivent donc  pas etre
>>>>>> diffuses, exploites ou copies sans autorisation. Si vous avez recu
>>>>>> ce message par erreur, veuillez le signaler  a l'expediteur et le
>>>>>> detruire ainsi que les pieces jointes. Les messages electroniques
>>>>>> etant susceptibles d'alteration,  Orange decline toute
>>>>>> responsabilite si ce message a ete altere, deforme ou falsifie.
>>>>>> Merci.
>>>>>>
>>>>>> This message and its attachments may contain confidential or
>>>>>> privileged information that may be protected by law;  they should
>>>>>> not be distributed, used or copied without authorisation.
>>>>>> If you have received this email in error, please notify the sender
>>>>>> and delete this message and its attachments.
>>>>>> As emails may be altered, Orange is not liable for messages that
>>>>>> have been modified, changed or falsified.
>>>>>> Thank you.
>>>>>>
>>>>>> _______________________________________________
>>>>>> BESS mailing list
>>>>>> BESS@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>>>
>>>>>
>>>>> _______________________________________________________
>> ______________
>>>>> ___
>>>>> _
>>>>> _
>>>>> _______________________________________________
>>>>>
>>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>>> confidentielles ou privilegiees et ne doivent donc pas etre diffuses,
>>>>> exploites ou copies sans autorisation. Si vous avez recu ce message
>>>>> par erreur, veuillez le signaler a l'expediteur et le detruire ainsi
>>>>> que les pieces jointes. Les messages electroniques etant susceptibles
>>>>> d'alteration, Orange decline toute responsabilite si ce message a ete
>>>>> altere, deforme ou falsifie. Merci.
>>>>>
>>>>> This message and its attachments may contain confidential or
>>>>> privileged information that may be protected by law; they should not
>>>>> be distributed, used or copied without authorisation.
>>>>> If you have received this email in error, please notify the sender
>>>>> and delete this message and its attachments.
>>>>> As emails may be altered, Orange is not liable for messages that have
>>>>> been modified, changed or falsified.
>>>>> Thank you.


___________________________________________________________________________=
______________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confiden=
tielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu=
 ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages el=
ectroniques etant susceptibles d'alteration,
Orange decline toute responsabilite si ce message a ete altere, deforme ou =
falsifie. Merci.

This message and its attachments may contain confidential or privileged inf=
ormation that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and dele=
te this message and its attachments.
As emails may be altered, Orange is not liable for messages that have been =
modified, changed or falsified.
Thank you.

_______________________________________________
BESS mailing list
BESS@ietf.org
https://www.ietf.org/mailman/listinfo/bess=


From nobody Fri Mar 27 09:01:01 2015
Return-Path: <haoweiguo@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6ED9E1A1B59 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 09:00:59 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.611
X-Spam-Level: 
X-Spam-Status: No, score=-3.611 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id htzV6nCxN91i for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 09:00:54 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 94C8C1A1BBD for <bess@ietf.org>; Fri, 27 Mar 2015 09:00:52 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml404-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQT97389; Fri, 27 Mar 2015 16:00:50 +0000 (GMT)
Received: from NKGEML403-HUB.china.huawei.com (10.98.56.34) by lhreml404-hub.china.huawei.com (10.201.5.218) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 27 Mar 2015 16:00:31 +0000
Received: from NKGEML501-MBS.china.huawei.com ([169.254.2.244]) by nkgeml403-hub.china.huawei.com ([10.98.56.34]) with mapi id 14.03.0158.001; Sat, 28 Mar 2015 00:00:28 +0800
From: Haoweiguo <haoweiguo@huawei.com>
To: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8Qf23Bh/Vhyp0qijaKjediZQp0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HYAABS9Z9AAD2dSgAAN/i8VACtqLYAARXXAWQ==
Date: Fri, 27 Mar 2015 16:00:27 +0000
Message-ID: <DD5FC8DE455C3348B94340C0AB5517334F857801@nkgeml501-mbs.china.huawei.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com> <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com>, <D13AC565.6C964%jorge.rabadan@alcatel-lucent.com>
In-Reply-To: <D13AC565.6C964%jorge.rabadan@alcatel-lucent.com>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.47.144.149]
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/f3a7TE68Xk-dSAROJG3cKhyX4L0>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 16:00:59 -0000

Hi Jorge,
The picture in email is just for a simplified explanation. Source MAC can b=
e the layer 3 device connecting to CE3, CE3 will not drop the frame so easi=
ly. This is an absolutely loop.

Thanks,
weiguo

________________________________________
From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
Sent: Friday, March 27, 2015 23:39
To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com; bess@ietf=
.org
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algo=
rithm

Hi Weiguo,

I disagree with your explanation below.

This is the flow assuming PE1-PE2 *might* be both DF and PE3 is DF:

CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3=97=97>DROP

If the MAC SA of the frame is CE3=92s MAC, CE3 will discard.
Even if CE3 forwards to CE4->PE4, PE4 is NDF and will discard the frame.
Remember MH network ES' as per your diagram below, only support
single-active MH.


Thanks.
Jorge

-----Original Message-----
From: Haoweiguo <haoweiguo@huawei.com>
Date: Friday, March 27, 2015 at 8:11 AM
To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
(satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
<thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos
algorithm

>Hi Jorge,
>Let give you another indefinitely forwarding loop case, the picture is as
>follows:
>
>CE3------------CE4
>  |                        |
>PE3                   PE4
>
>
>
>PE1                   PE2
> |                         |
>CE1------------CE2
>
>If PE1 and PE2 are dual DF PE in transiet time, assuing PE3 and PE4 are
>in stable state ,PE3 is DF PE, PE4 is non-DF PE, the BUM traffic from CE3
>to CE1, the forwarding loop path is as follows;
>CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->PE4----
>>forever......
>
>Do you think the harm is enough?
>Thanks,
>weiguo
>________________________________________
>From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>Sent: Friday, March 27, 2015 22:59
>To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>bess@ietf.org
>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>algorithm
>
>Hi Weiguo,
>
>OK, thanks for clarifying, that is the part I did not understand in your
>loop :-)
>
>To me a loop means you send a frame and that frame keeps being forwarded
>by the same nodes indefinitely till you take a corrective action.
>In this case, CE3 might get a few packets during the transient state, but
>those packets will have its own MAC SA so CE3 will discard them. I don=92t
>think that is harmful for CE3 in most of the cases. And if it is, you just
>need to make sure the timers are long enough and the DF election properly
>implemented to avoid the transient DF-DF state.
>
>The handshaking is definitely overkilling IMHO.
>
>Thanks.
>Jorge
>
>-----Original Message-----
>From: Haoweiguo <haoweiguo@huawei.com>
>Date: Friday, March 27, 2015 at 7:29 AM
>To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
>(satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
><thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>algorithm
>
>>Hi Jorge,
>>Sorry, maybe i have not explain clearly, pls allow me to clarify it more
>>clearly.
>>In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2 is
>>single homed to remote PE3. CE1 and CE2 forms one layer 2 network, so it
>>multi-homed network scenario.
>>           CE3
>>             |
>>            PE3
>>
>>
>>PE1                  PE2
>>  |                       |
>>CE1-----------CE2
>>
>>In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends
>>BUM traffic to CE1, the traffic will reach PE1 and PE2, CE1 will receive
>>two copy of the traffic from CE3, also loop is formed, the form path:
>>
>>CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
>>
>>I think the traffic loop is absolutely not allowed. In MHN case, the loop
>>will formed due to dual-DF PE in transiet period.
>>
>>Thanks,
>>weiguo
>>
>>
>>
>>________________________________________
>>From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>>Sent: Friday, March 27, 2015 20:26
>>To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>>bess@ietf.org
>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>algorithm
>>
>>Hi Weiguo,
>>
>>About this:
>>
>>[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
>>and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>>loop back to CE1 through PE2. If CE1 can filter and drop the traffic , it
>>is fortunate, maybe no loop.
>>For multi-homed network scenario, say local network1(CE1 and CE2) is
>>multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>back
>>through PE1-->PE2-->CE2, it is very serious. It is absolutely not
>>allowed.
>>
>>
>>[JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2
>>will
>>not send the traffic back to CE1 due to the ESI-label and split-horizon
>>procedures.
>>In MH network case PE2 will NOT send the traffic to CE2 for the same
>>reason.
>>
>>Let me know if you disagree please.
>>
>>Thanks.
>>Jorge
>>
>>-----Original Message-----
>>From: Haoweiguo <haoweiguo@huawei.com>
>>Date: Thursday, March 26, 2015 at 9:55 PM
>>To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
>>"thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org"
>><bess@ietf.org>
>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>algorithm
>>
>>>Hi Satya,
>>>Pls see inline.
>>>
>>>Thanks,
>>>weiguo
>>>
>>>________________________________________
>>>From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
>>>Sent: Friday, March 27, 2015 12:11
>>>To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
>>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>algorithm
>>>
>>>Thomas,
>>>
>>>Thanks for your summary. It captures the essence of the discussion.
>>>[weiguo]: Yes, i agree.
>>>
>>>One feature of the HRW scheme is that it computes the Active and the
>>>Backup DF upfront.
>>>Whenever the Active PE, say PE1, dies;
>>>the Backup PE, PE2, can become the
>>>Active immediately upon receiving the withdraw of the ES route from PE1.
>>>It need not even have to wait for the 3 seconds time.
>>>[weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
>>>message:) But when PE1 leaves ESI, it can withdraw the ES route
>>>immediately.
>>>When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2 will
>>>both act as PE for some transiet time.
>>>
>>>On the other hand, the 3 second timer will serve as a bulk and batch
>>>mechanism for the case when new PE(s) is(are) coming up, so we don=B9t
>>>have
>>>to do a DF election ever so often.
>>>
>>>Haoweiguo,  I think the "loop" scenario is speculative.
>>>I think everybody understands that, yes, there may be some transient
>>>duplication of bum traffic; however, there is no "loop" problem to be
>>>solved.
>>>[weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>>case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
>>>and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>>>loop back to CE1 through PE2. If CE1 can filter and drop the traffic ,
>>>it
>>>is fortunate, maybe no loop.
>>>For multi-homed network scenario, say local network1(CE1 and CE2) is
>>>multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>>back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
>>>allowed.
>>>
>>>Thanks,
>>>--Satya
>>>
>>>
>>>
>>>On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
>>>
>>>>Hi Thomas,
>>>>I know you are neutral, sorry for my reply vulnerable to be
>>>>miunderstood.
>>>>
>>>>Thanks,
>>>>weiguo
>>>>
>>>>________________________________________
>>>>From: thomas.morin@orange.com [thomas.morin@orange.com]
>>>>Sent: Friday, March 27, 2015 11:59
>>>>To: Haoweiguo; bess@ietf.org
>>>>Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>>algorithm
>>>>
>>>>Hi Weiguo,
>>>>
>>>>2015-03-26, Haoweiguo:
>>>> > Thomas:
>>>>> > - RFC7432 may have transient periods where the DF election state is
>>>>> > not yet synchronized between the two peers:
>>>> >
>>>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic loop
>>>>
>>>>Note well that I didn't write that there can be transient _loops_ .
>>>>I tried to capture the exchange you had all, and what I gather is that
>>>>the split-horizon procedure _prevents_loops_, independently of any
>>>>transient period where DF state is not synchronized and which may lead
>>>>to transient _duplicates_.
>>>>
>>>>-Thomas
>>>>
>>>>>
>>>>> 26/03/2015 20:05, John E Drake :
>>>>>>
>>>>>> Weiguo,
>>>>>>
>>>>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I have
>>>>>> detailed, is a non-solution to a non-problem with tremendous control
>>>>>> plane cost.
>>>>>>
>>>>>> Yours Irrespectively,
>>>>>>
>>>>>> John
>>>>>>
>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>> Paxos algorithm
>>>>>>
>>>>>> Pls see below.
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> weiguo
>>>>>>
>>>>>>
>>>>>>---------------------------------------------------------------------
>>>>>>-
>>>>>>-
>>>>>>-
>>>>>>
>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>> *Sent:* Friday, March 27, 2015 6:00
>>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>> Paxos algorithm
>>>>>>
>>>>>> To recap,
>>>>>>
>>>>>> We have established that your proposal is untenable because of its
>>>>>> control plane load.
>>>>>>
>>>>>> We have established that your proposal is based upon a flawed
>>>>>> understanding of the DF election in RFC 7432.
>>>>>>
>>>>>> [weiguo]: In ethernet world, traffic loop is serious than short
>>>>>>timer
>>>>>> traffic disruption. If you want to implement  transiet traffic loop
>>>>>> process, i will modify my draft to solve your issue.
>>>>>>
>>>>>> If i am the developer, i will prefer short timer traffic disruption
>>>>>> based on current EVPN protocol.
>>>>>>
>>>>>> What you are now arguing is that your draft prevents two or more PEs
>>>>>> from being DF simultaneously. This is clearly nonsense.
>>>>>>
>>>>>> [weiguo]: I will modify the draft problem statements, and use the
>>>>>>same
>>>>>> handshaking solution to solve it.
>>>>>>
>>>>>> Furthermore, we have established that having two or more DFs for
>>>>>>what
>>>>>> even you admit is a brief transient leads to duplicate traffic,
>>>>>>which
>>>>>> is acceptable, but not loops, your assertion to the contrary.
>>>>>>
>>>>>> [weiguo]: It is transient loop and traffic duplication issue.
>>>>>>
>>>>>> Yours Irrespectively,
>>>>>>
>>>>>> John
>>>>>>
>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>> Paxos algorithm
>>>>>>
>>>>>> John,
>>>>>>
>>>>>> As your understanding of the EVPN draft, the DF election mechanism
>>>>>>has
>>>>>> more serious side effect, it will have short time traffic loop,i.e.,
>>>>>> dual DF PEs will exist for a short time. I think dual DF PEs is
>>>>>> absolutely not tolerated, because native ethernet header has no TTL,
>>>>>> up to several hundred ms traffic loop normally not tolerated in
>>>>>> commertial networks.
>>>>>>
>>>>>> As your understanding, the PEs should do as following:
>>>>>>
>>>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but have
>>>>>> rarely deployment.
>>>>>>
>>>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3 joins
>>>>>> ESI and trigger DF re-election. When reception timer expires:
>>>>>>
>>>>>> PE1 upgrades to DF PE.
>>>>>>
>>>>>> After reception timer+ ES route transmission timer:
>>>>>>
>>>>>> PE2 downloads to non-DF PE.
>>>>>>
>>>>>> So in timer clock sync case, dual DF PEs will exist at least
>>>>>> transmission timer.
>>>>>>
>>>>>> If NTP is used for timer sync, because it has bad accuracy, dual DF
>>>>>> PEs will exist more longer timer.
>>>>>>
>>>>>> So as your understanding for DF election, the drawback is more
>>>>>>clear.
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> weiguo
>>>>>>
>>>>>>
>>>>>>---------------------------------------------------------------------
>>>>>>-
>>>>>>-
>>>>>>-
>>>>>>
>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>> *Sent:* Friday, March 27, 2015 4:41
>>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>> Paxos algorithm
>>>>>>
>>>>>> Weiguo,
>>>>>>
>>>>>> We have already established that your proposal is untenable because
>>>>>>of
>>>>>> its control plane load.
>>>>>>
>>>>>> What we are now discussing is that your proposal is based upon a
>>>>>> misunderstanding of the algorithm in RFC 7432.  You are assuming
>>>>>>that
>>>>>> PE1 will advertise an ES route and then wait for the configured
>>>>>> interval before performing the DF election while PE2 and PE3 will
>>>>>> perform the DF election as soon as they receive the ES route from
>>>>>>PE1.
>>>>>> This is not what RFC 7432 says.
>>>>>>
>>>>>> Rather, what is says is that the advertisement of the ES route by
>>>>>>PE1
>>>>>> and its receipt by PE2 and PE3 causes all three PEs to start the
>>>>>> configured interval timer  - =B33. When the timer expires, each PE
>>>>>> builds an ordered list of the IP addresses of all the PE nodes
>>>>>> connected to the Ethernet segment (including itself), in increasing
>>>>>> numeric value.=B2
>>>>>>
>>>>>> Yours Irrespectively,
>>>>>>
>>>>>> John
>>>>>>
>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>> Paxos algorithm
>>>>>>
>>>>>> Pls read my detail replies to Satya. If you still can't catch it,
>>>>>>pls
>>>>>> read my draft and EVPN base protocol,  thanks
>>>>>>
>>>>>> weiguo
>>>>>>
>>>>>>
>>>>>>---------------------------------------------------------------------
>>>>>>-
>>>>>>-
>>>>>>-
>>>>>>
>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>> *Sent:* Friday, March 27, 2015 1:28
>>>>>> *To:* Patrice Brissette (pbrisset)
>>>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>> <mailto:bess@ietf.org>
>>>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>> Paxos algorithm
>>>>>>
>>>>>> I think Patrice is correct.  Your proposal doesn't solve the problem
>>>>>> and it does so at huge cost.
>>>>>>
>>>>>> Sent from my iPhone
>>>>>>
>>>>>>
>>>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>>>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>>>>>
>>>>>>      Weiguo,
>>>>>>
>>>>>>      I=B9m not sure I=B9m following here.
>>>>>>
>>>>>>      Don=B9t you have the same issue with your handshaking mechanism=
?
>>>>>>
>>>>>>      If you don=B9t know your peer, how can you handshake?
>>>>>>
>>>>>>      Regards,
>>>>>>
>>>>>>      Patrice
>>>>>>
>>>>>>      Image removed by sender.
>>>>>>
>>>>>>      *Patrice Brissette*
>>>>>>      TECHNICAL LEADER.ENGINEERING
>>>>>>
>>>>>>      pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>>      Phone: *+1 613 254 3336*
>>>>>>
>>>>>>
>>>>>>
>>>>>>      *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>>      Canada
>>>>>>      Cisco.com <http://www.cisco.com/global/CA/>
>>>>>>
>>>>>>
>>>>>>
>>>>>>      Image removed by sender.Think before you print.
>>>>>>
>>>>>>      This email may contain confidential and privileged material for
>>>>>>      the sole use of the intended recipient. Any review, use,
>>>>>>      distribution or disclosure by others is strictly prohibited. If
>>>>>>      you are not the intended recipient (or authorized to receive
>>>>>>for
>>>>>>      the recipient), please contact the sender by reply email and
>>>>>>      delete all copies of this message.
>>>>>>
>>>>>>      Please click here
>>>>>>
>>>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>>>>      for Company Registration Information.
>>>>>>
>>>>>>
>>>>>>      *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>><mailto:haoweiguo@huawei.com>>
>>>>>>      *Date: *Thursday, March 26, 2015 at 10:24 AM
>>>>>>      *To: *Patrice Brissette <pbrisset@cisco.com
>>>>>>      <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>>>>>      <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>>>      <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>>      <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>>>>      *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES based
>>>>>>on
>>>>>>      Paxos algorithm
>>>>>>
>>>>>>          Hi Patrice,
>>>>>>
>>>>>>          Up to reception timer traffic disruption in transient phase
>>>>>>is
>>>>>>          one of the issues.
>>>>>>
>>>>>>          Thanks,
>>>>>>
>>>>>>          weiguo
>>>>>>
>>>>>>
>>>>>>---------------------------------------------------------------------
>>>>>>-
>>>>>>-
>>>>>>-
>>>>>>
>>>>>>          *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>>>>>          <mailto:pbrisset@cisco.com>]
>>>>>>          *Sent:* Thursday, March 26, 2015 20:54
>>>>>>          *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>>>>>          bess@ietf.org <mailto:bess@ietf.org>
>>>>>>          *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>>          based on Paxos algorithm
>>>>>>
>>>>>>          Weiguo,
>>>>>>
>>>>>>          You mention "But if your draft have not solved all issues=
=B2,
>>>>>>
>>>>>>          Can you explain what Satya=B9s draft is not solving?
>>>>>>
>>>>>>          Regards,
>>>>>>
>>>>>>          Patrice
>>>>>>
>>>>>>          Image removed by sender.
>>>>>>
>>>>>>          *Patrice Brissette*
>>>>>>          TECHNICAL LEADER.ENGINEERING
>>>>>>
>>>>>>          pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>>          Phone: *+1 613 254 3336*
>>>>>>
>>>>>>
>>>>>>
>>>>>>          *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>>          Canada
>>>>>>          Cisco.com <http://www.cisco.com/global/CA/>
>>>>>>
>>>>>>
>>>>>>
>>>>>>          Image removed by sender.Think before you print.
>>>>>>
>>>>>>          This email may contain confidential and privileged material
>>>>>>          for the sole use of the intended recipient. Any review,
>>>>>>use,
>>>>>>          distribution or disclosure by others is strictly
>>>>>>prohibited.
>>>>>>          If you are not the intended recipient (or authorized to
>>>>>>          receive for the recipient), please contact the sender by
>>>>>>reply
>>>>>>          email and delete all copies of this message.
>>>>>>
>>>>>>          Please click here
>>>>>>
>>>>>><http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>>>>          for Company Registration Information.
>>>>>>
>>>>>>
>>>>>>          *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>>          <mailto:haoweiguo@huawei.com>>
>>>>>>          *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>>>>>          *To: *John E Drake <jdrake@juniper.net
>>>>>>          <mailto:jdrake@juniper.net>>, Ali Sajassi
>>>>>><sajassi@cisco.com
>>>>>>          <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>>          <mailto:bess@ietf.org>" <bess@ietf.org
>>>>>><mailto:bess@ietf.org>>
>>>>>>          *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>>          based on Paxos algorithm
>>>>>>
>>>>>>              Hi John,
>>>>>>
>>>>>>               Firstly i think EVPN community should reach consensus
>>>>>>on
>>>>>>              the issues of current DF election mechanism. All these
>>>>>>              issues should be resolved in a single new DF election
>>>>>>              draft,rather than in multiple separate drafts. If your
>>>>>>              draft can solve all these issues and stable, i have no
>>>>>>              question for its progressing. But if your draft have
>>>>>>not
>>>>>>              solved all issues, i think it had better combine with
>>>>>>              other drafts to provide a comprehensive solution. I
>>>>>>think
>>>>>>              the issues listed in draft-rabadan-bess-evpn-ac-df-01
>>>>>>and
>>>>>>              draft-hao-bess-evpn-df-handshaking-00 is valid, it
>>>>>>should
>>>>>>              be resolved. So i think although your new Hash
>>>>>>algorithm
>>>>>>              for DF election is good, it only includes partial
>>>>>>              enhancements, maybe it still needs some time for
>>>>>>consensus.
>>>>>>
>>>>>>              Thanks,
>>>>>>
>>>>>>              weiguo
>>>>>>
>>>>>>
>>>>>>---------------------------------------------------------------------
>>>>>>-
>>>>>>-
>>>>>>-
>>>>>>
>>>>>>              *From:*John E Drake [jdrake@juniper.net
>>>>>>              <mailto:jdrake@juniper.net>]
>>>>>>              *Sent:* Thursday, March 26, 2015 7:16
>>>>>>              *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>>              <mailto:bess@ietf.org>
>>>>>>              *Subject:* RE: [bess] Handshaking among PEs in an EVPN
>>>>>>ES
>>>>>>              based on Paxos algorithm
>>>>>>
>>>>>>              Weiguo,
>>>>>>
>>>>>>              Your proposal introduces a control plane processing
>>>>>>load
>>>>>>              that is O(#EVIs * PEs) per DF election and given that
>>>>>>              there can be 4K EVIs per ES, this looks like a
>>>>>>              **substantial** load. Furthermore,
>>>>>>
>>>>>>              you can=B9t  use the ES route to co-ordinate DF electio=
n
>>>>>>              because you would need to carry your new extended
>>>>>>              community for each EVI and they would not all fit.  You
>>>>>>              also can=B9t use the Per EVI Ethernet AD route because
>>>>>>that
>>>>>>              is processed by all PEs in the EVI.
>>>>>>
>>>>>>              I think that from a practical perspective the new DF
>>>>>>              election proposed in Satya=B9s draft is sufficiently
>>>>>>stable
>>>>>>              that it renders your draft moot, even if it could be
>>>>>>made
>>>>>>              to work.
>>>>>>
>>>>>>              Yours Irrespectively,
>>>>>>
>>>>>>              John
>>>>>>
>>>>>>              *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf
>>>>>>Of
>>>>>>              *Haoweiguo
>>>>>>              *Sent:* Wednesday, March 25, 2015 5:34 PM
>>>>>>              *To:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>>              <mailto:bess@ietf.org>
>>>>>>              *Subject:* Re: [bess] Handshaking among PEs in an EVPN
>>>>>>ES
>>>>>>              based on Paxos algorithm
>>>>>>
>>>>>>              Hi Ali,
>>>>>>
>>>>>>              Thanks for your information. I scanned through this
>>>>>>draft,
>>>>>>              it really introduces inter-chassis message for DF
>>>>>>election
>>>>>>              handshaking, the requirements in this draft is to
>>>>>>              eliminate transiet Loop and traffic duplication.
>>>>>>Current
>>>>>>              EVPN DF election mechanism already eliminated loop and
>>>>>>              traffic duplication by configuring long reception timer
>>>>>>on
>>>>>>              each multi-homed PE, but up to reception timer traffic
>>>>>>              disruption issue still exist. EVPN for DCI is an
>>>>>>important
>>>>>>              use case for EVPN, up to reception timer traffic
>>>>>>              disruption can't be tolerated for service providers, it
>>>>>>              should be improved.
>>>>>>
>>>>>>              Also for accuracy, i think handshaking state machine on
>>>>>>              each multi-homed PE is also needed. From solution
>>>>>>              perspective, in my draft, no inter-chassis message is
>>>>>>              introduced, only one new extended community is
>>>>>>introduced,
>>>>>>              i think the process is comparatively simple than your
>>>>>>              following draft.
>>>>>>
>>>>>>              Current EVPN DF election has some drawbacks, so there
>>>>>>are
>>>>>>              three new drafts about DF election emerged. I think
>>>>>>BESS
>>>>>>              WG can consider these three drafts in global view, a
>>>>>>              single,comprehensive new DF election draft is hoped.
>>>>>>
>>>>>>              thanks.
>>>>>>
>>>>>>              Thanks,
>>>>>>
>>>>>>              weiguo
>>>>>>
>>>>>>
>>>>>>---------------------------------------------------------------------
>>>>>>-
>>>>>>-
>>>>>>-
>>>>>>
>>>>>>              *From:*BESS [bess-bounces@ietf.org
>>>>>>              <mailto:bess-bounces@ietf.org>] on behalf of Ali
>>>>>>Sajassi
>>>>>>              (sajassi) [sajassi@cisco.com
>>>>>><mailto:sajassi@cisco.com>]
>>>>>>              *Sent:* Thursday, March 26, 2015 0:21
>>>>>>              *To:* bess@ietf.org <mailto:bess@ietf.org>
>>>>>>              *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>>>>>              based on Paxos algorithm
>>>>>>
>>>>>>              FYI- First published July 4, 2011
>>>>>>
>>>>>>
>>>>>>https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-rou
>>>>>>t
>>>>>>e
>>>>>>/
>>>>>>
>>>>>>              -Ali
>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> BESS mailing list
>>>>>> BESS@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>______________________________________________________________________
>>>>>_
>>>>>_
>>>>>_
>>>>>________________________________________________
>>>>>
>>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>>>confidentielles ou privilegiees et ne doivent donc
>>>>> pas etre diffuses, exploites ou copies sans autorisation. Si vous
>>>>>avez
>>>>>recu ce message par erreur, veuillez le signaler
>>>>> a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>>>>messages electroniques etant susceptibles d'alteration,
>>>>> Orange decline toute responsabilite si ce message a ete altere,
>>>>>deforme
>>>>>ou falsifie. Merci.
>>>>>
>>>>> This message and its attachments may contain confidential or
>>>>>privileged
>>>>>information that may be protected by law;
>>>>> they should not be distributed, used or copied without authorisation.
>>>>> If you have received this email in error, please notify the sender
>>>>>and
>>>>>delete this message and its attachments.
>>>>> As emails may be altered, Orange is not liable for messages that have
>>>>>been modified, changed or falsified.
>>>>> Thank you.
>>>>>
>>>>> _______________________________________________
>>>>> BESS mailing list
>>>>> BESS@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>>
>>>>
>>>>
>>>>_______________________________________________________________________
>>>>_
>>>>_
>>>>_
>>>>_______________________________________________
>>>>
>>>>Ce message et ses pieces jointes peuvent contenir des informations
>>>>confidentielles ou privilegiees et ne doivent donc
>>>>pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>>>>recu ce message par erreur, veuillez le signaler
>>>>a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>>>messages
>>>>electroniques etant susceptibles d'alteration,
>>>>Orange decline toute responsabilite si ce message a ete altere, deforme
>>>>ou falsifie. Merci.
>>>>
>>>>This message and its attachments may contain confidential or privileged
>>>>information that may be protected by law;
>>>>they should not be distributed, used or copied without authorisation.
>>>>If you have received this email in error, please notify the sender and
>>>>delete this message and its attachments.
>>>>As emails may be altered, Orange is not liable for messages that have
>>>>been modified, changed or falsified.
>>>>Thank you.
>>>>_______________________________________________
>>>>BESS mailing list
>>>>BESS@ietf.org
>>>>https://www.ietf.org/mailman/listinfo/bess
>>>_______________________________________________
>>>BESS mailing list
>>>BESS@ietf.org
>>>https://www.ietf.org/mailman/listinfo/bess
>>_______________________________________________
>>BESS mailing list
>>BESS@ietf.org
>>https://www.ietf.org/mailman/listinfo/bess=


From nobody Fri Mar 27 09:54:32 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9D7DB1AD2EE for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 09:54:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id o3uhBFZdsOkl for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 09:54:25 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpgre-esg-01.alcatel-lucent.com [135.245.210.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A18681AD2EC for <bess@ietf.org>; Fri, 27 Mar 2015 09:54:24 -0700 (PDT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (unknown [135.239.2.122]) by Websense Email Security Gateway with ESMTPS id CFAE9CFDF1B79; Fri, 27 Mar 2015 16:54:19 +0000 (GMT)
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id t2RGsM2l029441 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 17:54:22 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR712WXCHHUB03.zeu.alcatel-lucent.com ([135.239.2.74]) with mapi id 14.03.0195.001; Fri, 27 Mar 2015 17:54:22 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Haoweiguo <haoweiguo@huawei.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8QfEaq8IGmBqka8wmUCt0qdaZ0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HOAABL1oIAADZ/DgAAMKZcAACYGUQAAPKPZgACGEVUA
Date: Fri, 27 Mar 2015 16:54:22 +0000
Message-ID: <D13AD8D4.6C9A5%jorge.rabadan@alcatel-lucent.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com> <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com> <D13AC565.6C964%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857801@nkgeml501-mbs.china.huawei.com>
In-Reply-To: <DD5FC8DE455C3348B94340C0AB5517334F857801@nkgeml501-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.39]
Content-Type: text/plain; charset="utf-8"
Content-ID: <032F6C18404F6041A0740B2DCD9B89F4@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/FcznGVPsT0uaC2l3hDOIrw5cgF0>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 16:54:30 -0000
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From nobody Fri Mar 27 10:13:53 2015
Return-Path: <jheitz@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1107D1B2A05 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 10:13:50 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -13.911
X-Spam-Level: 
X-Spam-Status: No, score=-13.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id e-3kIUJHktHh for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 10:13:44 -0700 (PDT)
Received: from alln-iport-5.cisco.com (alln-iport-5.cisco.com [173.37.142.92]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8C8161B2A02 for <bess@ietf.org>; Fri, 27 Mar 2015 10:13:44 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=34095; q=dns/txt; s=iport; t=1427476425; x=1428686025; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=NgsNbuy60l3d1RN5ciH7bZBwTBmM9c4Z3rLF+IZJWjE=; b=jLgzY0CFUcNMwRK2hptjXyZ5S9dpOIYUk5tvbQDNmyJzfpAjkih9xNUi P68uBpQroWvpho5ZXirJyqMNBhbMaVa+paTSWVEvYDpISRK0l2peAVCGm 4YRe2FwsAvTvVsla8l96o6IyMzsWB96b2M/LH/mrzC2c3tm4LhoRTpr6N M=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0APBQB6jhVV/5hdJa1ZA4MGUlrFHRkKhXUCgTdMAQEBAQEBfYQUAQEBAwEBAQEXAQw/CAsFBwICAgEIEQMBAQEBJwcbDAsUCQgCBA4FFAeIDAgNs2yYXwEBAQEBAQEBAQEBAQEBAQEBAQEBARcEiySEDQgBEQEODw0WCwUCBQYLgwaBFgWGGoo9g2+GAYEbOoJ3glqJQ4NIIoIBARyBUG8BgQMHF4EhAQEB
X-IronPort-AV: E=Sophos;i="5.11,479,1422921600"; d="scan'208";a="136036185"
Received: from rcdn-core-1.cisco.com ([173.37.93.152]) by alln-iport-5.cisco.com with ESMTP; 27 Mar 2015 17:13:43 +0000
Received: from xhc-rcd-x06.cisco.com (xhc-rcd-x06.cisco.com [173.37.183.80]) by rcdn-core-1.cisco.com (8.14.5/8.14.5) with ESMTP id t2RHDgDi001410 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 27 Mar 2015 17:13:42 GMT
Received: from xmb-aln-x02.cisco.com ([169.254.5.130]) by xhc-rcd-x06.cisco.com ([173.37.183.80]) with mapi id 14.03.0195.001; Fri, 27 Mar 2015 12:13:42 -0500
From: "Jakob Heitz (jheitz)" <jheitz@cisco.com>
To: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ1HVe3FMFl30a0u0Bv7Fz5r7fJ0ucdUAgACbTgCAACnuAIAAAsqAgAAf4ACAACCzAIAAFSOAgAAPjACAAAZ+AIAABLkAgAAvJACAAB1OgIAAD/gAgAADOwCAAAKcAIAAAO2AgAAMOoCAAH4LAIAAIlCAgAAIYICAAANSAIAAB8OAgAAF6oCAAA8QAP//sZU+
Date: Fri, 27 Mar 2015 17:13:41 +0000
Message-ID: <9EC1BDF0-E375-4767-9732-216505C19B97@cisco.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857294@nkgeml501-mbs.china.huawei.com> <D1396B6C.4EDD3%pbrisset@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F8573CC@nkgeml501-mbs.china.huawei.com> <D13990E2.4EE3E%pbrisset@cisco.com> <FB7DD2A6-C205-4F9D-A70B-A73A564E1F99@juniper.net> <DD5FC8DE455C3348B94340C0AB5517334F857483@nkgeml501-mbs.china.huawei.com> <BLUPR05MB56203C9BEDD75197CAE5489C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F8574F0@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562B3E4F4C04DF1FE86CDE7C7080@BLUPR05MB562.namprd05.prod.outlook.com> <DD5FC8DE455C3348B94340C0AB5517334F857541@nkgeml501-mbs.china.huawei.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com> <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com> <D13AC565.6C964%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857801@nkgeml501-mbs.china.huawei.com>, <D13AD8D4.6C9A5%jorge.rabadan@alcatel-lucent.com>
In-Reply-To: <D13AD8D4.6C9A5%jorge.rabadan@alcatel-lucent.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/CdXHSzdFiKWxiLjgiCz0-c1rZyU>
Cc: Haoweiguo <haoweiguo@huawei.com>, "thomas.morin@orange.com" <thomas.morin@orange.com>, "Satya Mohanty \(satyamoh\)" <satyamoh@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 17:13:50 -0000

LAG (Link Aggregation Group) is what stops the loop. The CE will not forwar=
d back to the PE.
PE-->CE-->PE is prevented, because the links from CE to PE is a LAG.

--
Jakob Heitz.


> On Mar 27, 2015, at 11:54 AM, Rabadan, Jorge (Jorge) <jorge.rabadan@alcat=
el-lucent.com> wrote:
>=20
> Hi Weiguo,
>=20
> As I said:
> "Even if CE3 forwards to CE4->PE4, PE4 is NDF and will discard the frame.
> Remember MH network ES' as per your diagram below, only support
> single-active MH.=94
>=20
> So there is NO LOOP.
>=20
> If you have any other scenario, let=92s talk offline. We are boring peopl=
e
> ;-)
>=20
>=20
> Thanks.
> Jorge
>=20
> -----Original Message-----
> From: Haoweiguo <haoweiguo@huawei.com>
> Date: Friday, March 27, 2015 at 9:00 AM
> To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
> (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
> <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>=20
>> Hi Jorge,
>> The picture in email is just for a simplified explanation. Source MAC ca=
n
>> be the layer 3 device connecting to CE3, CE3 will not drop the frame so
>> easily. This is an absolutely loop.
>>=20
>> Thanks,
>> weiguo
>>=20
>> ________________________________________
>> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>> Sent: Friday, March 27, 2015 23:39
>> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>> bess@ietf.org
>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>> algorithm
>>=20
>> Hi Weiguo,
>>=20
>> I disagree with your explanation below.
>>=20
>> This is the flow assuming PE1-PE2 *might* be both DF and PE3 is DF:
>>=20
>> CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3=97=97>DROP
>>=20
>> If the MAC SA of the frame is CE3=92s MAC, CE3 will discard.
>> Even if CE3 forwards to CE4->PE4, PE4 is NDF and will discard the frame.
>> Remember MH network ES' as per your diagram below, only support
>> single-active MH.
>>=20
>>=20
>> Thanks.
>> Jorge
>>=20
>> -----Original Message-----
>> From: Haoweiguo <haoweiguo@huawei.com>
>> Date: Friday, March 27, 2015 at 8:11 AM
>> To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
>> (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
>> <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
>> Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>> algorithm
>>=20
>>> Hi Jorge,
>>> Let give you another indefinitely forwarding loop case, the picture is =
as
>>> follows:
>>>=20
>>> CE3------------CE4
>>> |                        |
>>> PE3                   PE4
>>>=20
>>>=20
>>>=20
>>> PE1                   PE2
>>> |                         |
>>> CE1------------CE2
>>>=20
>>> If PE1 and PE2 are dual DF PE in transiet time, assuing PE3 and PE4 are
>>> in stable state ,PE3 is DF PE, PE4 is non-DF PE, the BUM traffic from C=
E3
>>> to CE1, the forwarding loop path is as follows;
>>> CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->PE4-=
--
>>> -
>>>> forever......
>>>=20
>>> Do you think the harm is enough?
>>> Thanks,
>>> weiguo
>>> ________________________________________
>>> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>>> Sent: Friday, March 27, 2015 22:59
>>> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>>> bess@ietf.org
>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>> algorithm
>>>=20
>>> Hi Weiguo,
>>>=20
>>> OK, thanks for clarifying, that is the part I did not understand in you=
r
>>> loop :-)
>>>=20
>>> To me a loop means you send a frame and that frame keeps being forwarde=
d
>>> by the same nodes indefinitely till you take a corrective action.
>>> In this case, CE3 might get a few packets during the transient state, b=
ut
>>> those packets will have its own MAC SA so CE3 will discard them. I don=
=92t
>>> think that is harmful for CE3 in most of the cases. And if it is, you
>>> just
>>> need to make sure the timers are long enough and the DF election proper=
ly
>>> implemented to avoid the transient DF-DF state.
>>>=20
>>> The handshaking is definitely overkilling IMHO.
>>>=20
>>> Thanks.
>>> Jorge
>>>=20
>>> -----Original Message-----
>>> From: Haoweiguo <haoweiguo@huawei.com>
>>> Date: Friday, March 27, 2015 at 7:29 AM
>>> To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
>>> (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
>>> <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>> algorithm
>>>=20
>>>> Hi Jorge,
>>>> Sorry, maybe i have not explain clearly, pls allow me to clarify it mo=
re
>>>> clearly.
>>>> In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2
>>>> is
>>>> single homed to remote PE3. CE1 and CE2 forms one layer 2 network, so =
it
>>>> multi-homed network scenario.
>>>>          CE3
>>>>            |
>>>>           PE3
>>>>=20
>>>>=20
>>>> PE1                  PE2
>>>> |                       |
>>>> CE1-----------CE2
>>>>=20
>>>> In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 send=
s
>>>> BUM traffic to CE1, the traffic will reach PE1 and PE2, CE1 will recei=
ve
>>>> two copy of the traffic from CE3, also loop is formed, the form path:
>>>>=20
>>>> CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
>>>>=20
>>>> I think the traffic loop is absolutely not allowed. In MHN case, the
>>>> loop
>>>> will formed due to dual-DF PE in transiet period.
>>>>=20
>>>> Thanks,
>>>> weiguo
>>>>=20
>>>>=20
>>>>=20
>>>> ________________________________________
>>>> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>>>> Sent: Friday, March 27, 2015 20:26
>>>> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>>>> bess@ietf.org
>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>> algorithm
>>>>=20
>>>> Hi Weiguo,
>>>>=20
>>>> About this:
>>>>=20
>>>> [weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>>> case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE=
2
>>>> and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>>>> loop back to CE1 through PE2. If CE1 can filter and drop the traffic ,
>>>> it
>>>> is fortunate, maybe no loop.
>>>> For multi-homed network scenario, say local network1(CE1 and CE2) is
>>>> multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>>> back
>>>> through PE1-->PE2-->CE2, it is very serious. It is absolutely not
>>>> allowed.
>>>>=20
>>>>=20
>>>> [JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2
>>>> will
>>>> not send the traffic back to CE1 due to the ESI-label and split-horizo=
n
>>>> procedures.
>>>> In MH network case PE2 will NOT send the traffic to CE2 for the same
>>>> reason.
>>>>=20
>>>> Let me know if you disagree please.
>>>>=20
>>>> Thanks.
>>>> Jorge
>>>>=20
>>>> -----Original Message-----
>>>> From: Haoweiguo <haoweiguo@huawei.com>
>>>> Date: Thursday, March 26, 2015 at 9:55 PM
>>>> To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
>>>> "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org"
>>>> <bess@ietf.org>
>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>> algorithm
>>>>=20
>>>>> Hi Satya,
>>>>> Pls see inline.
>>>>>=20
>>>>> Thanks,
>>>>> weiguo
>>>>>=20
>>>>> ________________________________________
>>>>> From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
>>>>> Sent: Friday, March 27, 2015 12:11
>>>>> To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
>>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxo=
s
>>>>> algorithm
>>>>>=20
>>>>> Thomas,
>>>>>=20
>>>>> Thanks for your summary. It captures the essence of the discussion.
>>>>> [weiguo]: Yes, i agree.
>>>>>=20
>>>>> One feature of the HRW scheme is that it computes the Active and the
>>>>> Backup DF upfront.
>>>>> Whenever the Active PE, say PE1, dies;
>>>>> the Backup PE, PE2, can become the
>>>>> Active immediately upon receiving the withdraw of the ES route from
>>>>> PE1.
>>>>> It need not even have to wait for the 3 seconds time.
>>>>> [weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
>>>>> message:) But when PE1 leaves ESI, it can withdraw the ES route
>>>>> immediately.
>>>>> When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2
>>>>> will
>>>>> both act as PE for some transiet time.
>>>>>=20
>>>>> On the other hand, the 3 second timer will serve as a bulk and batch
>>>>> mechanism for the case when new PE(s) is(are) coming up, so we don=B9=
t
>>>>> have
>>>>> to do a DF election ever so often.
>>>>>=20
>>>>> Haoweiguo,  I think the "loop" scenario is speculative.
>>>>> I think everybody understands that, yes, there may be some transient
>>>>> duplication of bum traffic; however, there is no "loop" problem to be
>>>>> solved.
>>>>> [weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>>>> case. In multi-homed device case, say CE1 is multi-homed to PE1 and P=
E2
>>>>> and both PE are DF PE. When CE1 sends traffic to PE1, the traffic wil=
l
>>>>> loop back to CE1 through PE2. If CE1 can filter and drop the traffic =
,
>>>>> it
>>>>> is fortunate, maybe no loop.
>>>>> For multi-homed network scenario, say local network1(CE1 and CE2) is
>>>>> multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>>>> back through PE1-->PE2-->CE2, it is very serious. It is absolutely no=
t
>>>>> allowed.
>>>>>=20
>>>>> Thanks,
>>>>> --Satya
>>>>>=20
>>>>>=20
>>>>>=20
>>>>>> On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
>>>>>>=20
>>>>>> Hi Thomas,
>>>>>> I know you are neutral, sorry for my reply vulnerable to be
>>>>>> miunderstood.
>>>>>>=20
>>>>>> Thanks,
>>>>>> weiguo
>>>>>>=20
>>>>>> ________________________________________
>>>>>> From: thomas.morin@orange.com [thomas.morin@orange.com]
>>>>>> Sent: Friday, March 27, 2015 11:59
>>>>>> To: Haoweiguo; bess@ietf.org
>>>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Pax=
os
>>>>>> algorithm
>>>>>>=20
>>>>>> Hi Weiguo,
>>>>>>=20
>>>>>> 2015-03-26, Haoweiguo:
>>>>>>> Thomas:
>>>>>>>> - RFC7432 may have transient periods where the DF election state
>>>>>>> is
>>>>>>>> not yet synchronized between the two peers:
>>>>>>>=20
>>>>>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic
>>>>>>> loop
>>>>>>=20
>>>>>> Note well that I didn't write that there can be transient _loops_ .
>>>>>> I tried to capture the exchange you had all, and what I gather is th=
at
>>>>>> the split-horizon procedure _prevents_loops_, independently of any
>>>>>> transient period where DF state is not synchronized and which may le=
ad
>>>>>> to transient _duplicates_.
>>>>>>=20
>>>>>> -Thomas
>>>>>>=20
>>>>>>>=20
>>>>>>> 26/03/2015 20:05, John E Drake :
>>>>>>>>=20
>>>>>>>> Weiguo,
>>>>>>>>=20
>>>>>>>> I guess I wasn=B9t clear.  I think you draft, for the reasons I ha=
ve
>>>>>>>> detailed, is a non-solution to a non-problem with tremendous
>>>>>>>> control
>>>>>>>> plane cost.
>>>>>>>>=20
>>>>>>>> Yours Irrespectively,
>>>>>>>>=20
>>>>>>>> John
>>>>>>>>=20
>>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
>>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>=20
>>>>>>>> Pls see below.
>>>>>>>>=20
>>>>>>>> Thanks,
>>>>>>>>=20
>>>>>>>> weiguo
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>> ------------------------------------------------------------------=
--
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>=20
>>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>>> *Sent:* Friday, March 27, 2015 6:00
>>>>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>=20
>>>>>>>> To recap,
>>>>>>>>=20
>>>>>>>> We have established that your proposal is untenable because of its
>>>>>>>> control plane load.
>>>>>>>>=20
>>>>>>>> We have established that your proposal is based upon a flawed
>>>>>>>> understanding of the DF election in RFC 7432.
>>>>>>>>=20
>>>>>>>> [weiguo]: In ethernet world, traffic loop is serious than short
>>>>>>>> timer
>>>>>>>> traffic disruption. If you want to implement  transiet traffic loo=
p
>>>>>>>> process, i will modify my draft to solve your issue.
>>>>>>>>=20
>>>>>>>> If i am the developer, i will prefer short timer traffic disruptio=
n
>>>>>>>> based on current EVPN protocol.
>>>>>>>>=20
>>>>>>>> What you are now arguing is that your draft prevents two or more
>>>>>>>> PEs
>>>>>>>> from being DF simultaneously. This is clearly nonsense.
>>>>>>>>=20
>>>>>>>> [weiguo]: I will modify the draft problem statements, and use the
>>>>>>>> same
>>>>>>>> handshaking solution to solve it.
>>>>>>>>=20
>>>>>>>> Furthermore, we have established that having two or more DFs for
>>>>>>>> what
>>>>>>>> even you admit is a brief transient leads to duplicate traffic,
>>>>>>>> which
>>>>>>>> is acceptable, but not loops, your assertion to the contrary.
>>>>>>>>=20
>>>>>>>> [weiguo]: It is transient loop and traffic duplication issue.
>>>>>>>>=20
>>>>>>>> Yours Irrespectively,
>>>>>>>>=20
>>>>>>>> John
>>>>>>>>=20
>>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
>>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>=20
>>>>>>>> John,
>>>>>>>>=20
>>>>>>>> As your understanding of the EVPN draft, the DF election mechanism
>>>>>>>> has
>>>>>>>> more serious side effect, it will have short time traffic
>>>>>>>> loop,i.e.,
>>>>>>>> dual DF PEs will exist for a short time. I think dual DF PEs is
>>>>>>>> absolutely not tolerated, because native ethernet header has no
>>>>>>>> TTL,
>>>>>>>> up to several hundred ms traffic loop normally not tolerated in
>>>>>>>> commertial networks.
>>>>>>>>=20
>>>>>>>> As your understanding, the PEs should do as following:
>>>>>>>>=20
>>>>>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but
>>>>>>>> have
>>>>>>>> rarely deployment.
>>>>>>>>=20
>>>>>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3
>>>>>>>> joins
>>>>>>>> ESI and trigger DF re-election. When reception timer expires:
>>>>>>>>=20
>>>>>>>> PE1 upgrades to DF PE.
>>>>>>>>=20
>>>>>>>> After reception timer+ ES route transmission timer:
>>>>>>>>=20
>>>>>>>> PE2 downloads to non-DF PE.
>>>>>>>>=20
>>>>>>>> So in timer clock sync case, dual DF PEs will exist at least
>>>>>>>> transmission timer.
>>>>>>>>=20
>>>>>>>> If NTP is used for timer sync, because it has bad accuracy, dual D=
F
>>>>>>>> PEs will exist more longer timer.
>>>>>>>>=20
>>>>>>>> So as your understanding for DF election, the drawback is more
>>>>>>>> clear.
>>>>>>>>=20
>>>>>>>> Thanks,
>>>>>>>>=20
>>>>>>>> weiguo
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>> ------------------------------------------------------------------=
--
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>=20
>>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>>> *Sent:* Friday, March 27, 2015 4:41
>>>>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>=20
>>>>>>>> Weiguo,
>>>>>>>>=20
>>>>>>>> We have already established that your proposal is untenable becaus=
e
>>>>>>>> of
>>>>>>>> its control plane load.
>>>>>>>>=20
>>>>>>>> What we are now discussing is that your proposal is based upon a
>>>>>>>> misunderstanding of the algorithm in RFC 7432.  You are assuming
>>>>>>>> that
>>>>>>>> PE1 will advertise an ES route and then wait for the configured
>>>>>>>> interval before performing the DF election while PE2 and PE3 will
>>>>>>>> perform the DF election as soon as they receive the ES route from
>>>>>>>> PE1.
>>>>>>>> This is not what RFC 7432 says.
>>>>>>>>=20
>>>>>>>> Rather, what is says is that the advertisement of the ES route by
>>>>>>>> PE1
>>>>>>>> and its receipt by PE2 and PE3 causes all three PEs to start the
>>>>>>>> configured interval timer  - =B33. When the timer expires, each PE
>>>>>>>> builds an ordered list of the IP addresses of all the PE nodes
>>>>>>>> connected to the Ethernet segment (including itself), in increasin=
g
>>>>>>>> numeric value.=B2
>>>>>>>>=20
>>>>>>>> Yours Irrespectively,
>>>>>>>>=20
>>>>>>>> John
>>>>>>>>=20
>>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
>>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>=20
>>>>>>>> Pls read my detail replies to Satya. If you still can't catch it,
>>>>>>>> pls
>>>>>>>> read my draft and EVPN base protocol,  thanks
>>>>>>>>=20
>>>>>>>> weiguo
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>> ------------------------------------------------------------------=
--
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>=20
>>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>>> *Sent:* Friday, March 27, 2015 1:28
>>>>>>>> *To:* Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>> <mailto:bess@ietf.org>
>>>>>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>=20
>>>>>>>> I think Patrice is correct.  Your proposal doesn't solve the
>>>>>>>> problem
>>>>>>>> and it does so at huge cost.
>>>>>>>>=20
>>>>>>>> Sent from my iPhone
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>>>>>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>>>>>>>=20
>>>>>>>>     Weiguo,
>>>>>>>>=20
>>>>>>>>     I=B9m not sure I=B9m following here.
>>>>>>>>=20
>>>>>>>>     Don=B9t you have the same issue with your handshaking mechanis=
m?
>>>>>>>>=20
>>>>>>>>     If you don=B9t know your peer, how can you handshake?
>>>>>>>>=20
>>>>>>>>     Regards,
>>>>>>>>=20
>>>>>>>>     Patrice
>>>>>>>>=20
>>>>>>>>     Image removed by sender.
>>>>>>>>=20
>>>>>>>>     *Patrice Brissette*
>>>>>>>>     TECHNICAL LEADER.ENGINEERING
>>>>>>>>=20
>>>>>>>>     pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>>>>     Phone: *+1 613 254 3336*
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>     *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>>>>     Canada
>>>>>>>>     Cisco.com <http://www.cisco.com/global/CA/>
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>     Image removed by sender.Think before you print.
>>>>>>>>=20
>>>>>>>>     This email may contain confidential and privileged material
>>>>>>>> for
>>>>>>>>     the sole use of the intended recipient. Any review, use,
>>>>>>>>     distribution or disclosure by others is strictly prohibited.
>>>>>>>> If
>>>>>>>>     you are not the intended recipient (or authorized to receive
>>>>>>>> for
>>>>>>>>     the recipient), please contact the sender by reply email and
>>>>>>>>     delete all copies of this message.
>>>>>>>>=20
>>>>>>>>     Please click here
>>>>>>>>=20
>>>>>>>> <http://www.cisco.com/web/about/doing_business/legal/cri/index.htm=
l>
>>>>>>>>     for Company Registration Information.
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>     *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>>>> <mailto:haoweiguo@huawei.com>>
>>>>>>>>     *Date: *Thursday, March 26, 2015 at 10:24 AM
>>>>>>>>     *To: *Patrice Brissette <pbrisset@cisco.com
>>>>>>>>     <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>>>>>>>     <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>>>>>     <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>>>>     <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>>>>>>     *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES
>>>>>>>> based
>>>>>>>> on
>>>>>>>>     Paxos algorithm
>>>>>>>>=20
>>>>>>>>         Hi Patrice,
>>>>>>>>=20
>>>>>>>>         Up to reception timer traffic disruption in transient
>>>>>>>> phase
>>>>>>>> is
>>>>>>>>         one of the issues.
>>>>>>>>=20
>>>>>>>>         Thanks,
>>>>>>>>=20
>>>>>>>>         weiguo
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>> ------------------------------------------------------------------=
--
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>=20
>>>>>>>>         *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>>>>>>>         <mailto:pbrisset@cisco.com>]
>>>>>>>>         *Sent:* Thursday, March 26, 2015 20:54
>>>>>>>>         *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>>>>>>>         bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>>         *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>>>>         based on Paxos algorithm
>>>>>>>>=20
>>>>>>>>         Weiguo,
>>>>>>>>=20
>>>>>>>>         You mention "But if your draft have not solved all
>>>>>>>> issues=B2,
>>>>>>>>=20
>>>>>>>>         Can you explain what Satya=B9s draft is not solving?
>>>>>>>>=20
>>>>>>>>         Regards,
>>>>>>>>=20
>>>>>>>>         Patrice
>>>>>>>>=20
>>>>>>>>         Image removed by sender.
>>>>>>>>=20
>>>>>>>>         *Patrice Brissette*
>>>>>>>>         TECHNICAL LEADER.ENGINEERING
>>>>>>>>=20
>>>>>>>>         pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>>>>         Phone: *+1 613 254 3336*
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>         *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>>>>         Canada
>>>>>>>>         Cisco.com <http://www.cisco.com/global/CA/>
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>         Image removed by sender.Think before you print.
>>>>>>>>=20
>>>>>>>>         This email may contain confidential and privileged
>>>>>>>> material
>>>>>>>>         for the sole use of the intended recipient. Any review,
>>>>>>>> use,
>>>>>>>>         distribution or disclosure by others is strictly
>>>>>>>> prohibited.
>>>>>>>>         If you are not the intended recipient (or authorized to
>>>>>>>>         receive for the recipient), please contact the sender by
>>>>>>>> reply
>>>>>>>>         email and delete all copies of this message.
>>>>>>>>=20
>>>>>>>>         Please click here
>>>>>>>>=20
>>>>>>>> <http://www.cisco.com/web/about/doing_business/legal/cri/index.htm=
l>
>>>>>>>>         for Company Registration Information.
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>         *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>>>>         <mailto:haoweiguo@huawei.com>>
>>>>>>>>         *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>>>>>>>         *To: *John E Drake <jdrake@juniper.net
>>>>>>>>         <mailto:jdrake@juniper.net>>, Ali Sajassi
>>>>>>>> <sajassi@cisco.com
>>>>>>>>         <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>>>>         <mailto:bess@ietf.org>" <bess@ietf.org
>>>>>>>> <mailto:bess@ietf.org>>
>>>>>>>>         *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>>>>         based on Paxos algorithm
>>>>>>>>=20
>>>>>>>>             Hi John,
>>>>>>>>=20
>>>>>>>>              Firstly i think EVPN community should reach consensus
>>>>>>>> on
>>>>>>>>             the issues of current DF election mechanism. All these
>>>>>>>>             issues should be resolved in a single new DF election
>>>>>>>>             draft,rather than in multiple separate drafts. If your
>>>>>>>>             draft can solve all these issues and stable, i have no
>>>>>>>>             question for its progressing. But if your draft have
>>>>>>>> not
>>>>>>>>             solved all issues, i think it had better combine with
>>>>>>>>             other drafts to provide a comprehensive solution. I
>>>>>>>> think
>>>>>>>>             the issues listed in draft-rabadan-bess-evpn-ac-df-01
>>>>>>>> and
>>>>>>>>             draft-hao-bess-evpn-df-handshaking-00 is valid, it
>>>>>>>> should
>>>>>>>>             be resolved. So i think although your new Hash
>>>>>>>> algorithm
>>>>>>>>             for DF election is good, it only includes partial
>>>>>>>>             enhancements, maybe it still needs some time for
>>>>>>>> consensus.
>>>>>>>>=20
>>>>>>>>             Thanks,
>>>>>>>>=20
>>>>>>>>             weiguo
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>> ------------------------------------------------------------------=
--
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>=20
>>>>>>>>             *From:*John E Drake [jdrake@juniper.net
>>>>>>>>             <mailto:jdrake@juniper.net>]
>>>>>>>>             *Sent:* Thursday, March 26, 2015 7:16
>>>>>>>>             *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>>             <mailto:bess@ietf.org>
>>>>>>>>             *Subject:* RE: [bess] Handshaking among PEs in an EVPN
>>>>>>>> ES
>>>>>>>>             based on Paxos algorithm
>>>>>>>>=20
>>>>>>>>             Weiguo,
>>>>>>>>=20
>>>>>>>>             Your proposal introduces a control plane processing
>>>>>>>> load
>>>>>>>>             that is O(#EVIs * PEs) per DF election and given that
>>>>>>>>             there can be 4K EVIs per ES, this looks like a
>>>>>>>>             **substantial** load. Furthermore,
>>>>>>>>=20
>>>>>>>>             you can=B9t  use the ES route to co-ordinate DF electi=
on
>>>>>>>>             because you would need to carry your new extended
>>>>>>>>             community for each EVI and they would not all fit.
>>>>>>>> You
>>>>>>>>             also can=B9t use the Per EVI Ethernet AD route because
>>>>>>>> that
>>>>>>>>             is processed by all PEs in the EVI.
>>>>>>>>=20
>>>>>>>>             I think that from a practical perspective the new DF
>>>>>>>>             election proposed in Satya=B9s draft is sufficiently
>>>>>>>> stable
>>>>>>>>             that it renders your draft moot, even if it could be
>>>>>>>> made
>>>>>>>>             to work.
>>>>>>>>=20
>>>>>>>>             Yours Irrespectively,
>>>>>>>>=20
>>>>>>>>             John
>>>>>>>>=20
>>>>>>>>             *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf
>>>>>>>> Of
>>>>>>>>             *Haoweiguo
>>>>>>>>             *Sent:* Wednesday, March 25, 2015 5:34 PM
>>>>>>>>             *To:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>>             <mailto:bess@ietf.org>
>>>>>>>>             *Subject:* Re: [bess] Handshaking among PEs in an EVPN
>>>>>>>> ES
>>>>>>>>             based on Paxos algorithm
>>>>>>>>=20
>>>>>>>>             Hi Ali,
>>>>>>>>=20
>>>>>>>>             Thanks for your information. I scanned through this
>>>>>>>> draft,
>>>>>>>>             it really introduces inter-chassis message for DF
>>>>>>>> election
>>>>>>>>             handshaking, the requirements in this draft is to
>>>>>>>>             eliminate transiet Loop and traffic duplication.
>>>>>>>> Current
>>>>>>>>             EVPN DF election mechanism already eliminated loop and
>>>>>>>>             traffic duplication by configuring long reception
>>>>>>>> timer
>>>>>>>> on
>>>>>>>>             each multi-homed PE, but up to reception timer traffic
>>>>>>>>             disruption issue still exist. EVPN for DCI is an
>>>>>>>> important
>>>>>>>>             use case for EVPN, up to reception timer traffic
>>>>>>>>             disruption can't be tolerated for service providers,
>>>>>>>> it
>>>>>>>>             should be improved.
>>>>>>>>=20
>>>>>>>>             Also for accuracy, i think handshaking state machine
>>>>>>>> on
>>>>>>>>             each multi-homed PE is also needed. From solution
>>>>>>>>             perspective, in my draft, no inter-chassis message is
>>>>>>>>             introduced, only one new extended community is
>>>>>>>> introduced,
>>>>>>>>             i think the process is comparatively simple than your
>>>>>>>>             following draft.
>>>>>>>>=20
>>>>>>>>             Current EVPN DF election has some drawbacks, so there
>>>>>>>> are
>>>>>>>>             three new drafts about DF election emerged. I think
>>>>>>>> BESS
>>>>>>>>             WG can consider these three drafts in global view, a
>>>>>>>>             single,comprehensive new DF election draft is hoped.
>>>>>>>>=20
>>>>>>>>             thanks.
>>>>>>>>=20
>>>>>>>>             Thanks,
>>>>>>>>=20
>>>>>>>>             weiguo
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>> ------------------------------------------------------------------=
--
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>=20
>>>>>>>>             *From:*BESS [bess-bounces@ietf.org
>>>>>>>>             <mailto:bess-bounces@ietf.org>] on behalf of Ali
>>>>>>>> Sajassi
>>>>>>>>             (sajassi) [sajassi@cisco.com
>>>>>>>> <mailto:sajassi@cisco.com>]
>>>>>>>>             *Sent:* Thursday, March 26, 2015 0:21
>>>>>>>>             *To:* bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>>             *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>>>>>>>             based on Paxos algorithm
>>>>>>>>=20
>>>>>>>>             FYI- First published July 4, 2011
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>> https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-=
ro
>>>>>>>> u
>>>>>>>> t
>>>>>>>> e
>>>>>>>> /
>>>>>>>>=20
>>>>>>>>             -Ali
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>>=20
>>>>>>>> _______________________________________________
>>>>>>>> BESS mailing list
>>>>>>>> BESS@ietf.org
>>>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>>=20
>>>>>>> ___________________________________________________________________=
__
>>>>>>> _
>>>>>>> _
>>>>>>> _
>>>>>>> _
>>>>>>> ________________________________________________
>>>>>>>=20
>>>>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>>>>> confidentielles ou privilegiees et ne doivent donc
>>>>>>> pas etre diffuses, exploites ou copies sans autorisation. Si vous
>>>>>>> avez
>>>>>>> recu ce message par erreur, veuillez le signaler
>>>>>>> a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>>>>>> messages electroniques etant susceptibles d'alteration,
>>>>>>> Orange decline toute responsabilite si ce message a ete altere,
>>>>>>> deforme
>>>>>>> ou falsifie. Merci.
>>>>>>>=20
>>>>>>> This message and its attachments may contain confidential or
>>>>>>> privileged
>>>>>>> information that may be protected by law;
>>>>>>> they should not be distributed, used or copied without
>>>>>>> authorisation.
>>>>>>> If you have received this email in error, please notify the sender
>>>>>>> and
>>>>>>> delete this message and its attachments.
>>>>>>> As emails may be altered, Orange is not liable for messages that
>>>>>>> have
>>>>>>> been modified, changed or falsified.
>>>>>>> Thank you.
>>>>>>>=20
>>>>>>> _______________________________________________
>>>>>>> BESS mailing list
>>>>>>> BESS@ietf.org
>>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>>>=20
>>>>>>=20
>>>>>> ____________________________________________________________________=
__
>>>>>> _
>>>>>> _
>>>>>> _
>>>>>> _
>>>>>> _______________________________________________
>>>>>>=20
>>>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>>>> confidentielles ou privilegiees et ne doivent donc
>>>>>> pas etre diffuses, exploites ou copies sans autorisation. Si vous av=
ez
>>>>>> recu ce message par erreur, veuillez le signaler
>>>>>> a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>>>>> messages
>>>>>> electroniques etant susceptibles d'alteration,
>>>>>> Orange decline toute responsabilite si ce message a ete altere,
>>>>>> deforme
>>>>>> ou falsifie. Merci.
>>>>>>=20
>>>>>> This message and its attachments may contain confidential or
>>>>>> privileged
>>>>>> information that may be protected by law;
>>>>>> they should not be distributed, used or copied without authorisation=
.
>>>>>> If you have received this email in error, please notify the sender a=
nd
>>>>>> delete this message and its attachments.
>>>>>> As emails may be altered, Orange is not liable for messages that hav=
e
>>>>>> been modified, changed or falsified.
>>>>>> Thank you.
>>>>>> _______________________________________________
>>>>>> BESS mailing list
>>>>>> BESS@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>> _______________________________________________
>>>>> BESS mailing list
>>>>> BESS@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>> _______________________________________________
>>>> BESS mailing list
>>>> BESS@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/bess
>> _______________________________________________
>> BESS mailing list
>> BESS@ietf.org
>> https://www.ietf.org/mailman/listinfo/bess
>=20
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess


From nobody Fri Mar 27 16:09:02 2015
Return-Path: <thomas.morin@orange.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 920D21A00C2 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 16:08:58 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FREEMAIL_FROM=0.001, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001, UNPARSEABLE_RELAY=0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fHOdMXGOgKSX for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 16:08:52 -0700 (PDT)
Received: from relais-inet.francetelecom.com (relais-ias243.francetelecom.com [80.12.204.243]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 62D651A00E1 for <bess@ietf.org>; Fri, 27 Mar 2015 16:08:51 -0700 (PDT)
Received: from omfeda08.si.francetelecom.fr (unknown [xx.xx.xx.201]) by omfeda14.si.francetelecom.fr (ESMTP service) with ESMTP id B15072AC46F; Sat, 28 Mar 2015 00:08:49 +0100 (CET)
Received: from Exchangemail-eme1.itn.ftgroup (unknown [10.114.1.183]) by omfeda08.si.francetelecom.fr (ESMTP service) with ESMTP id 99F9D384069; Sat, 28 Mar 2015 00:08:49 +0100 (CET)
Received: from [10.193.116.37] (10.197.38.6) by PEXCVZYH02.corporate.adroot.infra.ftgroup (10.114.1.183) with Microsoft SMTP Server (TLS) id 14.3.224.2; Sat, 28 Mar 2015 00:08:48 +0100
Message-ID: <3719_1427497729_5515E301_3719_11416_1_5515E2FD.5070702@orange.com>
Date: Fri, 27 Mar 2015 19:08:45 -0400
From: <thomas.morin@orange.com>
Organization: Orange
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, Haoweiguo <haoweiguo@huawei.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "bess@ietf.org" <bess@ietf.org>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com> <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com> <D13AC565.6C964%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857801@nkgeml501-mbs.china.huawei.com> <D13AD8D4.6C9A5%jorge.rabadan@alcatel-lucent.com>
In-Reply-To: <D13AD8D4.6C9A5%jorge.rabadan@alcatel-lucent.com>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit
X-Originating-IP: [10.197.38.6]
X-PMX-Version: 6.2.1.2478543, Antispam-Engine: 2.7.2.2107409, Antispam-Data: 2015.3.27.185417
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/O5dB7dhvk3BWZWJlJAo8rg0FC3g>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 23:08:58 -0000

Hi Jorge,

Jorge:
> If you have any other scenario, let’s talk offline. We are boring people
> ;-)

I think its nice to keep the discussion on the list, which is here for a 
reason (that is, not just for polls for adoption and WG last calls...).

I'm sure these clarifications can be useful for some of the 
participants, and the others can very much skip the thread.

And there is, I think, perhaps one more thing worth spelling out.

RFC7432 explains that for multi-homing in active-active mode, LAG is 
required.
My understanding of the reason why, in this case, it is not supported to 
connect the bridged
network via multiple CEs is that there is no technique to build a LAG 
between two CEs and two PEs.

Best,

-Thomas



Rabadan, Jorge :
> As I said:
> "Even if CE3 forwards to CE4->PE4, PE4 is NDF and will discard the frame.
> Remember MH network ES' as per your diagram below, only support
> single-active MH.”
>
> So there is NO LOOP.
>
>
>
>
> Thanks.
> Jorge
>
> -----Original Message-----
> From: Haoweiguo <haoweiguo@huawei.com>
> Date: Friday, March 27, 2015 at 9:00 AM
> To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
> (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
> <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
> algorithm
>
>> Hi Jorge,
>> The picture in email is just for a simplified explanation. Source MAC can
>> be the layer 3 device connecting to CE3, CE3 will not drop the frame so
>> easily. This is an absolutely loop.
>>
>> Thanks,
>> weiguo
>>
>> ________________________________________
>> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>> Sent: Friday, March 27, 2015 23:39
>> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>> bess@ietf.org
>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>> algorithm
>>
>> Hi Weiguo,
>>
>> I disagree with your explanation below.
>>
>> This is the flow assuming PE1-PE2 *might* be both DF and PE3 is DF:
>>
>> CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3——>DROP
>>
>> If the MAC SA of the frame is CE3’s MAC, CE3 will discard.
>> Even if CE3 forwards to CE4->PE4, PE4 is NDF and will discard the frame.
>> Remember MH network ES' as per your diagram below, only support
>> single-active MH.
>>
>>
>> Thanks.
>> Jorge
>>
>> -----Original Message-----
>> From: Haoweiguo <haoweiguo@huawei.com>
>> Date: Friday, March 27, 2015 at 8:11 AM
>> To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
>> (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
>> <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
>> Subject: RE: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>> algorithm
>>
>>> Hi Jorge,
>>> Let give you another indefinitely forwarding loop case, the picture is as
>>> follows:
>>>
>>> CE3------------CE4
>>>   |                        |
>>> PE3                   PE4
>>>
>>>
>>>
>>> PE1                   PE2
>>> |                         |
>>> CE1------------CE2
>>>
>>> If PE1 and PE2 are dual DF PE in transiet time, assuing PE3 and PE4 are
>>> in stable state ,PE3 is DF PE, PE4 is non-DF PE, the BUM traffic from CE3
>>> to CE1, the forwarding loop path is as follows;
>>> CE3--->PE3------>PE2--->CE2---->CE1---->PE1---->PE3-->CE3-->CE4--->PE4---
>>> -
>>>> forever......
>>> Do you think the harm is enough?
>>> Thanks,
>>> weiguo
>>> ________________________________________
>>> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>>> Sent: Friday, March 27, 2015 22:59
>>> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>>> bess@ietf.org
>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>> algorithm
>>>
>>> Hi Weiguo,
>>>
>>> OK, thanks for clarifying, that is the part I did not understand in your
>>> loop :-)
>>>
>>> To me a loop means you send a frame and that frame keeps being forwarded
>>> by the same nodes indefinitely till you take a corrective action.
>>> In this case, CE3 might get a few packets during the transient state, but
>>> those packets will have its own MAC SA so CE3 will discard them. I don’t
>>> think that is harmful for CE3 in most of the cases. And if it is, you
>>> just
>>> need to make sure the timers are long enough and the DF election properly
>>> implemented to avoid the transient DF-DF state.
>>>
>>> The handshaking is definitely overkilling IMHO.
>>>
>>> Thanks.
>>> Jorge
>>>
>>> -----Original Message-----
>>> From: Haoweiguo <haoweiguo@huawei.com>
>>> Date: Friday, March 27, 2015 at 7:29 AM
>>> To: Jorge Rabadan <jorge.rabadan@alcatel-lucent.com>, "Satya Mohanty
>>> (satyamoh)" <satyamoh@cisco.com>, "thomas.morin@orange.com"
>>> <thomas.morin@orange.com>, "bess@ietf.org" <bess@ietf.org>
>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>> algorithm
>>>
>>>> Hi Jorge,
>>>> Sorry, maybe i have not explain clearly, pls allow me to clarify it more
>>>> clearly.
>>>> In MHN case, assuming CE1 and CE2 are multti-homed to PE1 and PE2, CE2
>>>> is
>>>> single homed to remote PE3. CE1 and CE2 forms one layer 2 network, so it
>>>> multi-homed network scenario.
>>>>            CE3
>>>>              |
>>>>             PE3
>>>>
>>>>
>>>> PE1                  PE2
>>>>   |                       |
>>>> CE1-----------CE2
>>>>
>>>> In transiet period, PE1 and PE2 are both DF PE. At this time, CE3 sends
>>>> BUM traffic to CE1, the traffic will reach PE1 and PE2, CE1 will receive
>>>> two copy of the traffic from CE3, also loop is formed, the form path:
>>>>
>>>> CE3---->PE3---->PE2---->CE2---->CE1---->PE1--->PE3--->CE3
>>>>
>>>> I think the traffic loop is absolutely not allowed. In MHN case, the
>>>> loop
>>>> will formed due to dual-DF PE in transiet period.
>>>>
>>>> Thanks,
>>>> weiguo
>>>>
>>>>
>>>>
>>>> ________________________________________
>>>> From: Rabadan, Jorge (Jorge) [jorge.rabadan@alcatel-lucent.com]
>>>> Sent: Friday, March 27, 2015 20:26
>>>> To: Haoweiguo; Satya Mohanty (satyamoh); thomas.morin@orange.com;
>>>> bess@ietf.org
>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>> algorithm
>>>>
>>>> Hi Weiguo,
>>>>
>>>> About this:
>>>>
>>>> [weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>>> case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
>>>> and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>>>> loop back to CE1 through PE2. If CE1 can filter and drop the traffic ,
>>>> it
>>>> is fortunate, maybe no loop.
>>>> For multi-homed network scenario, say local network1(CE1 and CE2) is
>>>> multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>>> back
>>>> through PE1-->PE2-->CE2, it is very serious. It is absolutely not
>>>> allowed.
>>>>
>>>>
>>>> [JORGE] In a MH device case: CE1->BUM->PE1->(ESI-label+BUM)->PE2. PE2
>>>> will
>>>> not send the traffic back to CE1 due to the ESI-label and split-horizon
>>>> procedures.
>>>> In MH network case PE2 will NOT send the traffic to CE2 for the same
>>>> reason.
>>>>
>>>> Let me know if you disagree please.
>>>>
>>>> Thanks.
>>>> Jorge
>>>>
>>>> -----Original Message-----
>>>> From: Haoweiguo <haoweiguo@huawei.com>
>>>> Date: Thursday, March 26, 2015 at 9:55 PM
>>>> To: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>,
>>>> "thomas.morin@orange.com" <thomas.morin@orange.com>, "bess@ietf.org"
>>>> <bess@ietf.org>
>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>> algorithm
>>>>
>>>>> Hi Satya,
>>>>> Pls see inline.
>>>>>
>>>>> Thanks,
>>>>> weiguo
>>>>>
>>>>> ________________________________________
>>>>> From: Satya Mohanty (satyamoh) [satyamoh@cisco.com]
>>>>> Sent: Friday, March 27, 2015 12:11
>>>>> To: Haoweiguo; thomas.morin@orange.com; bess@ietf.org
>>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>>> algorithm
>>>>>
>>>>> Thomas,
>>>>>
>>>>> Thanks for your summary. It captures the essence of the discussion.
>>>>> [weiguo]: Yes, i agree.
>>>>>
>>>>> One feature of the HRW scheme is that it computes the Active and the
>>>>> Backup DF upfront.
>>>>> Whenever the Active PE, say PE1, dies;
>>>>> the Backup PE, PE2, can become the
>>>>> Active immediately upon receiving the withdraw of the ES route from
>>>>> PE1.
>>>>> It need not even have to wait for the 3 seconds time.
>>>>> [weiguo]: I think maybe your negligence, dies PE1 can't send withdraw
>>>>> message:) But when PE1 leaves ESI, it can withdraw the ES route
>>>>> immediately.
>>>>> When PE1 dies and recovers, assuming PE1 is DF PE, then PE1 and PE2
>>>>> will
>>>>> both act as PE for some transiet time.
>>>>>
>>>>> On the other hand, the 3 second timer will serve as a bulk and batch
>>>>> mechanism for the case when new PE(s) is(are) coming up, so we don¹t
>>>>> have
>>>>> to do a DF election ever so often.
>>>>>
>>>>> Haoweiguo,  I think the "loop" scenario is speculative.
>>>>> I think everybody understands that, yes, there may be some transient
>>>>> duplication of bum traffic; however, there is no "loop" problem to be
>>>>> solved.
>>>>> [weiguo]: Traffic duplication and loop issue both exist in dual DF PE
>>>>> case. In multi-homed device case, say CE1 is multi-homed to PE1 and PE2
>>>>> and both PE are DF PE. When CE1 sends traffic to PE1, the traffic will
>>>>> loop back to CE1 through PE2. If CE1 can filter and drop the traffic ,
>>>>> it
>>>>> is fortunate, maybe no loop.
>>>>> For multi-homed network scenario, say local network1(CE1 and CE2) is
>>>>> multi-homed to PE1 and PE2. Similarly, the traffic from CE1 will loop
>>>>> back through PE1-->PE2-->CE2, it is very serious. It is absolutely not
>>>>> allowed.
>>>>>
>>>>> Thanks,
>>>>> --Satya
>>>>>
>>>>>
>>>>>
>>>>> On 3/26/15 9:08 PM, "Haoweiguo" <haoweiguo@huawei.com> wrote:
>>>>>
>>>>>> Hi Thomas,
>>>>>> I know you are neutral, sorry for my reply vulnerable to be
>>>>>> miunderstood.
>>>>>>
>>>>>> Thanks,
>>>>>> weiguo
>>>>>>
>>>>>> ________________________________________
>>>>>> From: thomas.morin@orange.com [thomas.morin@orange.com]
>>>>>> Sent: Friday, March 27, 2015 11:59
>>>>>> To: Haoweiguo; bess@ietf.org
>>>>>> Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos
>>>>>> algorithm
>>>>>>
>>>>>> Hi Weiguo,
>>>>>>
>>>>>> 2015-03-26, Haoweiguo:
>>>>>>> Thomas:
>>>>>>>> - RFC7432 may have transient periods where the DF election state
>>>>>>> is
>>>>>>>> not yet synchronized between the two peers:
>>>>>>> [weiguo]: Yes, i think RFC 7432 has transient periods of traffic
>>>>>>> loop
>>>>>> Note well that I didn't write that there can be transient _loops_ .
>>>>>> I tried to capture the exchange you had all, and what I gather is that
>>>>>> the split-horizon procedure _prevents_loops_, independently of any
>>>>>> transient period where DF state is not synchronized and which may lead
>>>>>> to transient _duplicates_.
>>>>>>
>>>>>> -Thomas
>>>>>>
>>>>>>> 26/03/2015 20:05, John E Drake :
>>>>>>>> Weiguo,
>>>>>>>>
>>>>>>>> I guess I wasn¹t clear.  I think you draft, for the reasons I have
>>>>>>>> detailed, is a non-solution to a non-problem with tremendous
>>>>>>>> control
>>>>>>>> plane cost.
>>>>>>>>
>>>>>>>> Yours Irrespectively,
>>>>>>>>
>>>>>>>> John
>>>>>>>>
>>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>>> *Sent:* Thursday, March 26, 2015 6:17 PM
>>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>
>>>>>>>> Pls see below.
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>>
>>>>>>>> weiguo
>>>>>>>>
>>>>>>>>
>>>>>>>> --------------------------------------------------------------------
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>
>>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>>> *Sent:* Friday, March 27, 2015 6:00
>>>>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>
>>>>>>>> To recap,
>>>>>>>>
>>>>>>>> We have established that your proposal is untenable because of its
>>>>>>>> control plane load.
>>>>>>>>
>>>>>>>> We have established that your proposal is based upon a flawed
>>>>>>>> understanding of the DF election in RFC 7432.
>>>>>>>>
>>>>>>>> [weiguo]: In ethernet world, traffic loop is serious than short
>>>>>>>> timer
>>>>>>>> traffic disruption. If you want to implement  transiet traffic loop
>>>>>>>> process, i will modify my draft to solve your issue.
>>>>>>>>
>>>>>>>> If i am the developer, i will prefer short timer traffic disruption
>>>>>>>> based on current EVPN protocol.
>>>>>>>>
>>>>>>>> What you are now arguing is that your draft prevents two or more
>>>>>>>> PEs
>>>>>>>> from being DF simultaneously. This is clearly nonsense.
>>>>>>>>
>>>>>>>> [weiguo]: I will modify the draft problem statements, and use the
>>>>>>>> same
>>>>>>>> handshaking solution to solve it.
>>>>>>>>
>>>>>>>> Furthermore, we have established that having two or more DFs for
>>>>>>>> what
>>>>>>>> even you admit is a brief transient leads to duplicate traffic,
>>>>>>>> which
>>>>>>>> is acceptable, but not loops, your assertion to the contrary.
>>>>>>>>
>>>>>>>> [weiguo]: It is transient loop and traffic duplication issue.
>>>>>>>>
>>>>>>>> Yours Irrespectively,
>>>>>>>>
>>>>>>>> John
>>>>>>>>
>>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>>> *Sent:* Thursday, March 26, 2015 5:37 PM
>>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>
>>>>>>>> John,
>>>>>>>>
>>>>>>>> As your understanding of the EVPN draft, the DF election mechanism
>>>>>>>> has
>>>>>>>> more serious side effect, it will have short time traffic
>>>>>>>> loop,i.e.,
>>>>>>>> dual DF PEs will exist for a short time. I think dual DF PEs is
>>>>>>>> absolutely not tolerated, because native ethernet header has no
>>>>>>>> TTL,
>>>>>>>> up to several hundred ms traffic loop normally not tolerated in
>>>>>>>> commertial networks.
>>>>>>>>
>>>>>>>> As your understanding, the PEs should do as following:
>>>>>>>>
>>>>>>>> 1. Accurate timer sync. NTP accuracy is bad, 1588v2 is good but
>>>>>>>> have
>>>>>>>> rarely deployment.
>>>>>>>>
>>>>>>>> Assuming PE1,PE2 and PE3 have consistent timer clock, when PE3
>>>>>>>> joins
>>>>>>>> ESI and trigger DF re-election. When reception timer expires:
>>>>>>>>
>>>>>>>> PE1 upgrades to DF PE.
>>>>>>>>
>>>>>>>> After reception timer+ ES route transmission timer:
>>>>>>>>
>>>>>>>> PE2 downloads to non-DF PE.
>>>>>>>>
>>>>>>>> So in timer clock sync case, dual DF PEs will exist at least
>>>>>>>> transmission timer.
>>>>>>>>
>>>>>>>> If NTP is used for timer sync, because it has bad accuracy, dual DF
>>>>>>>> PEs will exist more longer timer.
>>>>>>>>
>>>>>>>> So as your understanding for DF election, the drawback is more
>>>>>>>> clear.
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>>
>>>>>>>> weiguo
>>>>>>>>
>>>>>>>>
>>>>>>>> --------------------------------------------------------------------
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>
>>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>>> *Sent:* Friday, March 27, 2015 4:41
>>>>>>>> *To:* Haoweiguo; Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>
>>>>>>>> Weiguo,
>>>>>>>>
>>>>>>>> We have already established that your proposal is untenable because
>>>>>>>> of
>>>>>>>> its control plane load.
>>>>>>>>
>>>>>>>> What we are now discussing is that your proposal is based upon a
>>>>>>>> misunderstanding of the algorithm in RFC 7432.  You are assuming
>>>>>>>> that
>>>>>>>> PE1 will advertise an ES route and then wait for the configured
>>>>>>>> interval before performing the DF election while PE2 and PE3 will
>>>>>>>> perform the DF election as soon as they receive the ES route from
>>>>>>>> PE1.
>>>>>>>> This is not what RFC 7432 says.
>>>>>>>>
>>>>>>>> Rather, what is says is that the advertisement of the ES route by
>>>>>>>> PE1
>>>>>>>> and its receipt by PE2 and PE3 causes all three PEs to start the
>>>>>>>> configured interval timer  - ³3. When the timer expires, each PE
>>>>>>>> builds an ordered list of the IP addresses of all the PE nodes
>>>>>>>> connected to the Ethernet segment (including itself), in increasing
>>>>>>>> numeric value.²
>>>>>>>>
>>>>>>>> Yours Irrespectively,
>>>>>>>>
>>>>>>>> John
>>>>>>>>
>>>>>>>> *From:*Haoweiguo [mailto:haoweiguo@huawei.com]
>>>>>>>> *Sent:* Thursday, March 26, 2015 3:26 PM
>>>>>>>> *To:* John E Drake; Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Ali Sajassi (sajassi); bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>> *Subject:* RE: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>
>>>>>>>> Pls read my detail replies to Satya. If you still can't catch it,
>>>>>>>> pls
>>>>>>>> read my draft and EVPN base protocol,  thanks
>>>>>>>>
>>>>>>>> weiguo
>>>>>>>>
>>>>>>>>
>>>>>>>> --------------------------------------------------------------------
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>
>>>>>>>> *From:*John E Drake [jdrake@juniper.net]
>>>>>>>> *Sent:* Friday, March 27, 2015 1:28
>>>>>>>> *To:* Patrice Brissette (pbrisset)
>>>>>>>> *Cc:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>> <mailto:bess@ietf.org>
>>>>>>>> *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES based on
>>>>>>>> Paxos algorithm
>>>>>>>>
>>>>>>>> I think Patrice is correct.  Your proposal doesn't solve the
>>>>>>>> problem
>>>>>>>> and it does so at huge cost.
>>>>>>>>
>>>>>>>> Sent from my iPhone
>>>>>>>>
>>>>>>>>
>>>>>>>> On Mar 26, 2015, at 10:34 AM, Patrice Brissette (pbrisset)
>>>>>>>> <pbrisset@cisco.com <mailto:pbrisset@cisco.com>> wrote:
>>>>>>>>
>>>>>>>>       Weiguo,
>>>>>>>>
>>>>>>>>       I¹m not sure I¹m following here.
>>>>>>>>
>>>>>>>>       Don¹t you have the same issue with your handshaking mechanism?
>>>>>>>>
>>>>>>>>       If you don¹t know your peer, how can you handshake?
>>>>>>>>
>>>>>>>>       Regards,
>>>>>>>>
>>>>>>>>       Patrice
>>>>>>>>
>>>>>>>>       Image removed by sender.
>>>>>>>>
>>>>>>>>       *Patrice Brissette*
>>>>>>>>       TECHNICAL LEADER.ENGINEERING
>>>>>>>>
>>>>>>>>       pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>>>>       Phone: *+1 613 254 3336*
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>       *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>>>>       Canada
>>>>>>>>       Cisco.com <http://www.cisco.com/global/CA/>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>       Image removed by sender.Think before you print.
>>>>>>>>
>>>>>>>>       This email may contain confidential and privileged material
>>>>>>>> for
>>>>>>>>       the sole use of the intended recipient. Any review, use,
>>>>>>>>       distribution or disclosure by others is strictly prohibited.
>>>>>>>> If
>>>>>>>>       you are not the intended recipient (or authorized to receive
>>>>>>>> for
>>>>>>>>       the recipient), please contact the sender by reply email and
>>>>>>>>       delete all copies of this message.
>>>>>>>>
>>>>>>>>       Please click here
>>>>>>>>
>>>>>>>> <http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>>>>>>       for Company Registration Information.
>>>>>>>>
>>>>>>>>
>>>>>>>>       *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>>>> <mailto:haoweiguo@huawei.com>>
>>>>>>>>       *Date: *Thursday, March 26, 2015 at 10:24 AM
>>>>>>>>       *To: *Patrice Brissette <pbrisset@cisco.com
>>>>>>>>       <mailto:pbrisset@cisco.com>>, John E Drake <jdrake@juniper.net
>>>>>>>>       <mailto:jdrake@juniper.net>>, Ali Sajassi <sajassi@cisco.com
>>>>>>>>       <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>>>>       <mailto:bess@ietf.org>" <bess@ietf.org <mailto:bess@ietf.org>>
>>>>>>>>       *Subject: *RE: [bess] Handshaking among PEs in an EVPN ES
>>>>>>>> based
>>>>>>>> on
>>>>>>>>       Paxos algorithm
>>>>>>>>
>>>>>>>>           Hi Patrice,
>>>>>>>>
>>>>>>>>           Up to reception timer traffic disruption in transient
>>>>>>>> phase
>>>>>>>> is
>>>>>>>>           one of the issues.
>>>>>>>>
>>>>>>>>           Thanks,
>>>>>>>>
>>>>>>>>           weiguo
>>>>>>>>
>>>>>>>>
>>>>>>>> --------------------------------------------------------------------
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>
>>>>>>>>           *From:*Patrice Brissette (pbrisset) [pbrisset@cisco.com
>>>>>>>>           <mailto:pbrisset@cisco.com>]
>>>>>>>>           *Sent:* Thursday, March 26, 2015 20:54
>>>>>>>>           *To:* Haoweiguo; John E Drake; Ali Sajassi (sajassi);
>>>>>>>>           bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>>           *Subject:* Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>>>>           based on Paxos algorithm
>>>>>>>>
>>>>>>>>           Weiguo,
>>>>>>>>
>>>>>>>>           You mention "But if your draft have not solved all
>>>>>>>> issues²,
>>>>>>>>
>>>>>>>>           Can you explain what Satya¹s draft is not solving?
>>>>>>>>
>>>>>>>>           Regards,
>>>>>>>>
>>>>>>>>           Patrice
>>>>>>>>
>>>>>>>>           Image removed by sender.
>>>>>>>>
>>>>>>>>           *Patrice Brissette*
>>>>>>>>           TECHNICAL LEADER.ENGINEERING
>>>>>>>>
>>>>>>>>           pbrisset@cisco.com <mailto:pbrisset@cisco.com>
>>>>>>>>           Phone: *+1 613 254 3336*
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>           *Cisco Systems Canada Co. / Les Systemes Cisco Canada CIE*
>>>>>>>>           Canada
>>>>>>>>           Cisco.com <http://www.cisco.com/global/CA/>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>           Image removed by sender.Think before you print.
>>>>>>>>
>>>>>>>>           This email may contain confidential and privileged
>>>>>>>> material
>>>>>>>>           for the sole use of the intended recipient. Any review,
>>>>>>>> use,
>>>>>>>>           distribution or disclosure by others is strictly
>>>>>>>> prohibited.
>>>>>>>>           If you are not the intended recipient (or authorized to
>>>>>>>>           receive for the recipient), please contact the sender by
>>>>>>>> reply
>>>>>>>>           email and delete all copies of this message.
>>>>>>>>
>>>>>>>>           Please click here
>>>>>>>>
>>>>>>>> <http://www.cisco.com/web/about/doing_business/legal/cri/index.html>
>>>>>>>>           for Company Registration Information.
>>>>>>>>
>>>>>>>>
>>>>>>>>           *From: *Haoweiguo <haoweiguo@huawei.com
>>>>>>>>           <mailto:haoweiguo@huawei.com>>
>>>>>>>>           *Date: *Wednesday, March 25, 2015 at 10:38 PM
>>>>>>>>           *To: *John E Drake <jdrake@juniper.net
>>>>>>>>           <mailto:jdrake@juniper.net>>, Ali Sajassi
>>>>>>>> <sajassi@cisco.com
>>>>>>>>           <mailto:sajassi@cisco.com>>, "bess@ietf.org
>>>>>>>>           <mailto:bess@ietf.org>" <bess@ietf.org
>>>>>>>> <mailto:bess@ietf.org>>
>>>>>>>>           *Subject: *Re: [bess] Handshaking among PEs in an EVPN ES
>>>>>>>>           based on Paxos algorithm
>>>>>>>>
>>>>>>>>               Hi John,
>>>>>>>>
>>>>>>>>                Firstly i think EVPN community should reach consensus
>>>>>>>> on
>>>>>>>>               the issues of current DF election mechanism. All these
>>>>>>>>               issues should be resolved in a single new DF election
>>>>>>>>               draft,rather than in multiple separate drafts. If your
>>>>>>>>               draft can solve all these issues and stable, i have no
>>>>>>>>               question for its progressing. But if your draft have
>>>>>>>> not
>>>>>>>>               solved all issues, i think it had better combine with
>>>>>>>>               other drafts to provide a comprehensive solution. I
>>>>>>>> think
>>>>>>>>               the issues listed in draft-rabadan-bess-evpn-ac-df-01
>>>>>>>> and
>>>>>>>>               draft-hao-bess-evpn-df-handshaking-00 is valid, it
>>>>>>>> should
>>>>>>>>               be resolved. So i think although your new Hash
>>>>>>>> algorithm
>>>>>>>>               for DF election is good, it only includes partial
>>>>>>>>               enhancements, maybe it still needs some time for
>>>>>>>> consensus.
>>>>>>>>
>>>>>>>>               Thanks,
>>>>>>>>
>>>>>>>>               weiguo
>>>>>>>>
>>>>>>>>
>>>>>>>> --------------------------------------------------------------------
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>
>>>>>>>>               *From:*John E Drake [jdrake@juniper.net
>>>>>>>>               <mailto:jdrake@juniper.net>]
>>>>>>>>               *Sent:* Thursday, March 26, 2015 7:16
>>>>>>>>               *To:* Haoweiguo; Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>>               <mailto:bess@ietf.org>
>>>>>>>>               *Subject:* RE: [bess] Handshaking among PEs in an EVPN
>>>>>>>> ES
>>>>>>>>               based on Paxos algorithm
>>>>>>>>
>>>>>>>>               Weiguo,
>>>>>>>>
>>>>>>>>               Your proposal introduces a control plane processing
>>>>>>>> load
>>>>>>>>               that is O(#EVIs * PEs) per DF election and given that
>>>>>>>>               there can be 4K EVIs per ES, this looks like a
>>>>>>>>               **substantial** load. Furthermore,
>>>>>>>>
>>>>>>>>               you can¹t  use the ES route to co-ordinate DF election
>>>>>>>>               because you would need to carry your new extended
>>>>>>>>               community for each EVI and they would not all fit.
>>>>>>>> You
>>>>>>>>               also can¹t use the Per EVI Ethernet AD route because
>>>>>>>> that
>>>>>>>>               is processed by all PEs in the EVI.
>>>>>>>>
>>>>>>>>               I think that from a practical perspective the new DF
>>>>>>>>               election proposed in Satya¹s draft is sufficiently
>>>>>>>> stable
>>>>>>>>               that it renders your draft moot, even if it could be
>>>>>>>> made
>>>>>>>>               to work.
>>>>>>>>
>>>>>>>>               Yours Irrespectively,
>>>>>>>>
>>>>>>>>               John
>>>>>>>>
>>>>>>>>               *From:*BESS [mailto:bess-bounces@ietf.org] *On Behalf
>>>>>>>> Of
>>>>>>>>               *Haoweiguo
>>>>>>>>               *Sent:* Wednesday, March 25, 2015 5:34 PM
>>>>>>>>               *To:* Ali Sajassi (sajassi); bess@ietf.org
>>>>>>>>               <mailto:bess@ietf.org>
>>>>>>>>               *Subject:* Re: [bess] Handshaking among PEs in an EVPN
>>>>>>>> ES
>>>>>>>>               based on Paxos algorithm
>>>>>>>>
>>>>>>>>               Hi Ali,
>>>>>>>>
>>>>>>>>               Thanks for your information. I scanned through this
>>>>>>>> draft,
>>>>>>>>               it really introduces inter-chassis message for DF
>>>>>>>> election
>>>>>>>>               handshaking, the requirements in this draft is to
>>>>>>>>               eliminate transiet Loop and traffic duplication.
>>>>>>>> Current
>>>>>>>>               EVPN DF election mechanism already eliminated loop and
>>>>>>>>               traffic duplication by configuring long reception
>>>>>>>> timer
>>>>>>>> on
>>>>>>>>               each multi-homed PE, but up to reception timer traffic
>>>>>>>>               disruption issue still exist. EVPN for DCI is an
>>>>>>>> important
>>>>>>>>               use case for EVPN, up to reception timer traffic
>>>>>>>>               disruption can't be tolerated for service providers,
>>>>>>>> it
>>>>>>>>               should be improved.
>>>>>>>>
>>>>>>>>               Also for accuracy, i think handshaking state machine
>>>>>>>> on
>>>>>>>>               each multi-homed PE is also needed. From solution
>>>>>>>>               perspective, in my draft, no inter-chassis message is
>>>>>>>>               introduced, only one new extended community is
>>>>>>>> introduced,
>>>>>>>>               i think the process is comparatively simple than your
>>>>>>>>               following draft.
>>>>>>>>
>>>>>>>>               Current EVPN DF election has some drawbacks, so there
>>>>>>>> are
>>>>>>>>               three new drafts about DF election emerged. I think
>>>>>>>> BESS
>>>>>>>>               WG can consider these three drafts in global view, a
>>>>>>>>               single,comprehensive new DF election draft is hoped.
>>>>>>>>
>>>>>>>>               thanks.
>>>>>>>>
>>>>>>>>               Thanks,
>>>>>>>>
>>>>>>>>               weiguo
>>>>>>>>
>>>>>>>>
>>>>>>>> --------------------------------------------------------------------
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>> -
>>>>>>>>
>>>>>>>>               *From:*BESS [bess-bounces@ietf.org
>>>>>>>>               <mailto:bess-bounces@ietf.org>] on behalf of Ali
>>>>>>>> Sajassi
>>>>>>>>               (sajassi) [sajassi@cisco.com
>>>>>>>> <mailto:sajassi@cisco.com>]
>>>>>>>>               *Sent:* Thursday, March 26, 2015 0:21
>>>>>>>>               *To:* bess@ietf.org <mailto:bess@ietf.org>
>>>>>>>>               *Subject:* [bess] Handshaking among PEs in an EVPN ES
>>>>>>>>               based on Paxos algorithm
>>>>>>>>
>>>>>>>>               FYI- First published July 4, 2011
>>>>>>>>
>>>>>>>>
>>>>>>>> https://datatracker.ietf.org/doc/draft-sajassi-l2vpn-evpn-segment-ro
>>>>>>>> u
>>>>>>>> t
>>>>>>>> e
>>>>>>>> /
>>>>>>>>
>>>>>>>>               -Ali
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> BESS mailing list
>>>>>>>> BESS@ietf.org
>>>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _____________________________________________________________________
>>>>>>> _
>>>>>>> _
>>>>>>> _
>>>>>>> _
>>>>>>> ________________________________________________
>>>>>>>
>>>>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>>>>> confidentielles ou privilegiees et ne doivent donc
>>>>>>> pas etre diffuses, exploites ou copies sans autorisation. Si vous
>>>>>>> avez
>>>>>>> recu ce message par erreur, veuillez le signaler
>>>>>>> a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>>>>>> messages electroniques etant susceptibles d'alteration,
>>>>>>> Orange decline toute responsabilite si ce message a ete altere,
>>>>>>> deforme
>>>>>>> ou falsifie. Merci.
>>>>>>>
>>>>>>> This message and its attachments may contain confidential or
>>>>>>> privileged
>>>>>>> information that may be protected by law;
>>>>>>> they should not be distributed, used or copied without
>>>>>>> authorisation.
>>>>>>> If you have received this email in error, please notify the sender
>>>>>>> and
>>>>>>> delete this message and its attachments.
>>>>>>> As emails may be altered, Orange is not liable for messages that
>>>>>>> have
>>>>>>> been modified, changed or falsified.
>>>>>>> Thank you.
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> BESS mailing list
>>>>>>> BESS@ietf.org
>>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>>>>
>>>>>>
>>>>>> ______________________________________________________________________
>>>>>> _
>>>>>> _
>>>>>> _
>>>>>> _
>>>>>> _______________________________________________
>>>>>>
>>>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>>>> confidentielles ou privilegiees et ne doivent donc
>>>>>> pas etre diffuses, exploites ou copies sans autorisation. Si vous avez
>>>>>> recu ce message par erreur, veuillez le signaler
>>>>>> a l'expediteur et le detruire ainsi que les pieces jointes. Les
>>>>>> messages
>>>>>> electroniques etant susceptibles d'alteration,
>>>>>> Orange decline toute responsabilite si ce message a ete altere,
>>>>>> deforme
>>>>>> ou falsifie. Merci.
>>>>>>
>>>>>> This message and its attachments may contain confidential or
>>>>>> privileged
>>>>>> information that may be protected by law;
>>>>>> they should not be distributed, used or copied without authorisation.
>>>>>> If you have received this email in error, please notify the sender and
>>>>>> delete this message and its attachments.
>>>>>> As emails may be altered, Orange is not liable for messages that have
>>>>>> been modified, changed or falsified.
>>>>>> Thank you.
>>>>>> _______________________________________________
>>>>>> BESS mailing list
>>>>>> BESS@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>>> _______________________________________________
>>>>> BESS mailing list
>>>>> BESS@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/bess
>>>> _______________________________________________
>>>> BESS mailing list
>>>> BESS@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/bess
>> _______________________________________________
>> BESS mailing list
>> BESS@ietf.org
>> https://www.ietf.org/mailman/listinfo/bess


_________________________________________________________________________________________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration,
Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci.

This message and its attachments may contain confidential or privileged information that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and delete this message and its attachments.
As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified.
Thank you.


From nobody Fri Mar 27 20:52:58 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 211F61A1B74 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 20:52:57 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_56=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id D4ARnf4K3n59 for <bess@ietfa.amsl.com>; Fri, 27 Mar 2015 20:52:51 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpgre-esg-01.alcatel-lucent.com [135.245.210.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 904801A1B57 for <bess@ietf.org>; Fri, 27 Mar 2015 20:52:49 -0700 (PDT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (unknown [135.239.2.122]) by Websense Email Security Gateway with ESMTPS id DB03B6399C6F2; Sat, 28 Mar 2015 03:52:45 +0000 (GMT)
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id t2S3qk48016774 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Sat, 28 Mar 2015 04:52:46 +0100
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR711WXCHHUB02.zeu.alcatel-lucent.com ([135.239.2.112]) with mapi id 14.03.0195.001; Sat, 28 Mar 2015 04:52:46 +0100
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: "thomas.morin@orange.com" <thomas.morin@orange.com>, Haoweiguo <haoweiguo@huawei.com>, "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>, "bess@ietf.org" <bess@ietf.org>
Thread-Topic: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
Thread-Index: AQHQZ8QfEaq8IGmBqka8wmUCt0qdaZ0u4a8F//99WoCAAB/fAIAApme0//+Pb4CAAJHuo///hBwAABEPgy///6tiAIAAHU6AgACQ0dX//4JiAIAAhyEr//98Z4CAAI3P2v///HOAABL1oIAADZ/DgAAMKZcAACYGUQAAPKPZgACGEVUAAPBju4AB5YexgA==
Date: Sat, 28 Mar 2015 03:52:46 +0000
Message-ID: <D13B6F2F.6CAF1%jorge.rabadan@alcatel-lucent.com>
References: <D1384AA7.12E0C1%sajassi@cisco.com> <BLUPR05MB562A3B02E72B599D65C1F42C7090@BLUPR05MB562.namprd05.prod.outlook.com> <26983_1427424638_5514C57E_26983_8706_1_5514C57B.8010802@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857611@nkgeml501-mbs.china.huawei.com> <19169_1427428762_5514D59A_19169_8719_1_5514D596.2090308@orange.com> <DD5FC8DE455C3348B94340C0AB5517334F857632@nkgeml501-mbs.china.huawei.com> <D13A263C.2CB9E%satyamoh@cisco.com> <DD5FC8DE455C3348B94340C0AB5517334F857658@nkgeml501-mbs.china.huawei.com> <D13A9956.6C8BF%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857702@nkgeml501-mbs.china.huawei.com> <D13ABC18.6C903%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857788@nkgeml501-mbs.china.huawei.com> <D13AC565.6C964%jorge.rabadan@alcatel-lucent.com> <DD5FC8DE455C3348B94340C0AB5517334F857801@nkgeml501-mbs.china.huawei.com> <D13AD8D4.6C9A5%jorge.rabadan@alcatel-lucent.com> <3719_1427497729_5515E301_3719_11416_1_5515E2FD.5070702@orange.com>
In-Reply-To: <3719_1427497729_5515E301_3719_11416_1_5515E2FD.5070702@orange.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.38]
Content-Type: text/plain; charset="utf-8"
Content-ID: <A60C8BDAF2A7E1499315CCB2EC480B96@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/h2qNz2YkTrn2r4zmXsERnhbj7VI>
Subject: Re: [bess] Handshaking among PEs in an EVPN ES based on Paxos algorithm
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 28 Mar 2015 03:52:57 -0000
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From nobody Sun Mar 29 19:59:54 2015
Return-Path: <antoni.przygienda@ericsson.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A06771A8ADA for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 19:59:53 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.901
X-Spam-Level: 
X-Spam-Status: No, score=-0.901 tagged_above=-999 required=5 tests=[BAYES_50=0.8, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id MqIY3NMFecz4 for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 19:59:52 -0700 (PDT)
Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1DF4C1A8AD7 for <bess@ietf.org>; Sun, 29 Mar 2015 19:59:52 -0700 (PDT)
X-AuditID: c618062d-f79686d0000030a8-27-5518661fdb82
Received: from EUSAAHC004.ericsson.se (Unknown_Domain [147.117.188.84]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id 04.A5.12456.F1668155; Sun, 29 Mar 2015 22:52:48 +0200 (CEST)
Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC004.ericsson.se ([147.117.188.84]) with mapi id 14.03.0210.002; Sun, 29 Mar 2015 22:59:47 -0400
From: Antoni Przygienda <antoni.przygienda@ericsson.com>
To: "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>, Erik Nordmark <nordmark@acm.org>, "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAZ/NmAAAeuvZAAEFu+gAAIR4HQ
Date: Mon, 30 Mar 2015 02:59:47 +0000
Message-ID: <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se> <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com>
In-Reply-To: <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [147.117.188.10]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFnrHLMWRmVeSWpSXmKPExsUyuXRPiK5CmkSowd9DPBYrjs9ktvh2aQmL Rf/UnUwWsx//YHNg8bh8xduj9dleVo8lS34yBTBHcdmkpOZklqUW6dslcGU0nGxjKVgnVLFx 4lr2BsYPgl2MnBwSAiYSl4//YoawxSQu3FvP1sXIxSEkcJRRYsb8ViYIZzmjxOPlD1hAqtgE LCQuf3vKDJIQEVjEKHHl/002kASzgLLE4fnn2UFsYQEFifaOr2BjRQQUJeY/X8fYxcgBZLtJ rFkJVs4ioCpxfdVlVhCbV8BbYvf8V8wQy54xSpxbeRaslxOo/u/JC2AzGYHO+35qDRPELnGJ W0/mM0GcLSCxZM95qBdEJV4+/scKYStJTFp6jhVkL7OApsT6XfoQrYoSU7ofskPsFZQ4OfMJ ywRGsVlIps5C6JiFpGMWko4FjCyrGDlKi1PLctONDDYxAiPomASb7g7GPS8tDzEKcDAq8fAu sJUIFWJNLCuuzD3EKM3BoiTOu+jBwRAhgfTEktTs1NSC1KL4otKc1OJDjEwcnFINjK4cH153 hp/bqPHizqHvDpuPBWhfWO3MVK5f0by0/b2W/g6Of2//GfZcb3ip5FrFeefQ2f3rPI9mten1 7Pvz48a0vhD7+++UuequLji84abMVPvv356xr3gUoOdzSCZgWZ3afrX72lHbn+luO1tz/HHR TtNbrvNmKn6T3rZ49/3vYU2n/Swc95gqsRRnJBpqMRcVJwIAyYlhsYECAAA=
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/6oPp5bKKsw325pEr47PXG_xYlis>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 30 Mar 2015 02:59:53 -0000
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From nobody Sun Mar 29 21:18:29 2015
Return-Path: <wim.henderickx@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B50221A8F46 for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 21:18:26 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xRirW_cts7iC for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 21:18:25 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id F14AD1A8F44 for <bess@ietf.org>; Sun, 29 Mar 2015 21:18:24 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id ED2DB53A4CEA0; Mon, 30 Mar 2015 04:18:21 +0000 (GMT)
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2U4IMZk002789 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 30 Mar 2015 06:18:22 +0200
Received: from FR711WXCHMBA07.zeu.alcatel-lucent.com ([169.254.3.17]) by FR711WXCHHUB02.zeu.alcatel-lucent.com ([135.239.2.112]) with mapi id 14.03.0195.001; Mon, 30 Mar 2015 06:18:22 +0200
From: "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
To: Antoni Przygienda <antoni.przygienda@ericsson.com>, Erik Nordmark <nordmark@acm.org>, "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AQHQZ1QBsq1vN6Z2REGnqTtRu+9A/p0tzyWAgABl5ACABgXxgIAAN3kA
Date: Mon, 30 Mar 2015 04:18:21 +0000
Message-ID: <61BFC96E-350C-4F77-84E6-EFA6794B5EC0@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se> <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se>
In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se>
Accept-Language: nl-BE, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/15.8.1.150311
x-originating-ip: [135.239.27.38]
Content-Type: text/plain; charset="utf-8"
Content-ID: <C1817EA5A9C6F345942A8B1CB78DEF9F@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/kGu_eFUqFfzHoXFPvLA2hjVxGEA>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 30 Mar 2015 04:18:26 -0000
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From nobody Sun Mar 29 21:45:07 2015
Return-Path: <rraszuk@gmail.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AF6011A901E for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 21:45:06 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.677
X-Spam-Level: 
X-Spam-Status: No, score=-0.677 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, FM_FORGED_GMAIL=0.622, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, J_CHICKENPOX_12=0.6, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Ou1pi1TrSlLg for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 21:45:05 -0700 (PDT)
Received: from mail-ig0-x22a.google.com (mail-ig0-x22a.google.com [IPv6:2607:f8b0:4001:c05::22a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 610551A9023 for <bess@ietf.org>; Sun, 29 Mar 2015 21:45:04 -0700 (PDT)
Received: by igcau2 with SMTP id au2so63730139igc.0 for <bess@ietf.org>; Sun, 29 Mar 2015 21:45:03 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:sender:in-reply-to:references:date:message-id:subject :from:to:cc:content-type; bh=9N8K5GSuy1WUS9ZNiffTvNaFWzQjPbmsvIbb62IHE9w=; b=F/9fvGhrX2w8eUITK0zZK+OWg0LrqFFx7O4VLGvkdkRX2+gC2KL8fkrrt5Oh9Vaxf0 KIh1xnaiNmQIp1ff5jnhHvbDhzD0aQe/uBjKJC0YzLmoEggwAFmeh/BwEfixsfW2VR8u AC3nM9/2WVm6OQCVTBCb9I5ZhTkU45j7JZmOjxEyQvllEY2ZkOSDnIZvE7UDgUxkWMz2 UfZavDFCKzkH4O1wFx8A5EqhvkhQmhg3qAp7y1Bnh5wDuYEbKiNNUAjn1mU+JTNqJKfd YzQTNoWeN2JBfWOPWDGP40JmnPs34xM85zo+aUhG0wul4oyZL7pbqnCd3IWeD1LD/qHX vD5A==
MIME-Version: 1.0
X-Received: by 10.50.137.99 with SMTP id qh3mr15139098igb.9.1427690703806; Sun, 29 Mar 2015 21:45:03 -0700 (PDT)
Sender: rraszuk@gmail.com
Received: by 10.107.137.70 with HTTP; Sun, 29 Mar 2015 21:45:03 -0700 (PDT)
In-Reply-To: <61BFC96E-350C-4F77-84E6-EFA6794B5EC0@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se> <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se> <61BFC96E-350C-4F77-84E6-EFA6794B5EC0@alcatel-lucent.com>
Date: Sun, 29 Mar 2015 23:45:03 -0500
X-Google-Sender-Auth: _76zTFbXsT3lgj67YUsGTdO-Mzw
Message-ID: <CA+b+ERmVVaz4jm3+CRc7WMfrQn1TN5=sEiK-Qq+nngtw3yw--Q@mail.gmail.com>
From: Robert Raszuk <robert@raszuk.net>
To: "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Content-Type: multipart/alternative; boundary=001a11c3bcbc3feaaf05127a2566
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/CFBjIiEJMUvofBlwh6hW8hXxzHQ>
Cc: Erik Nordmark <nordmark@acm.org>, Antoni Przygienda <antoni.przygienda@ericsson.com>, "bess@ietf.org" <bess@ietf.org>, "Rabadan, Jorge \(Jorge\)" <jorge.rabadan@alcatel-lucent.com>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 30 Mar 2015 04:45:07 -0000

--001a11c3bcbc3feaaf05127a2566
Content-Type: text/plain; charset=UTF-8

Hi Wim,

What makes you say that in IPv4 there is no anycast ? All anycase I have
played so far is IPv4 :)

Cheers,
r.

On Sun, Mar 29, 2015 at 11:18 PM, Henderickx, Wim (Wim) <
wim.henderickx@alcatel-lucent.com> wrote:

> We will update the draft to highlight the IPv6 anycast behaviour better as
> pointed out by RObert. In IPv4 there is no anycast behaviour and as such
> there should be one option possible.
>
>
>
> On 30/03/15 04:59, "Antoni Przygienda" <antoni.przygienda@ericsson.com>
> wrote:
>
> >Yes, but of course I brought it up to show that 'the last one simply
> wins' as suggested by the draft is not enough IMO. A good architecture
> should probably keep track of what it served as answer and when the answer
> is invalid or a new, better one exists, provide a GARP.
> >
> >As well, when PE2 sends a newer MAC it may not be a good strategy to
> serve a GARP if PE1's MAC has already been offered. That could lead IMO to
> e.g. gateway chasing problems.
> >
> >--- tony
> >
> >
> >There are basically two types of people. People who accomplish things,
> and people who claim to have accomplished things. The first group is less
> crowded.
> >~~~ Mark Twain
> >
> >
> >> -----Original Message-----
> >> From: Henderickx, Wim (Wim) [mailto:wim.henderickx@alcatel-lucent.com]
> >> Sent: Thursday, March 26, 2015 6:01 AM
> >> To: Antoni Przygienda; Erik Nordmark; Rabadan, Jorge (Jorge)
> >> Cc: bess@ietf.org
> >> Subject: Re: [bess] ARP ND draft
> >>
> >> For this case you should sent a GARP with the new MAC/IP
> >>
> >>
> >>
> >>
> >> On 25/03/15 18:56, "Antoni Przygienda" <antoni.przygienda@ericsson.com>
> >> wrote:
> >>
> >> >> > b)It is worth explaining what is suggested behavior if eVPN
> >> >> > advertises the same IP with multiple MACs and what happens when
> >> >> > e.g. the served MAC vanishes
> >> >> >
> >> >> Doesn't the EVPN RFC already stating that the routes would be
> >> >> withdrawn in that case?
> >> >
> >> >The scenario I had in mind was when eVPN PE receives
> >> >
> >> >From PE2  IP1/M1  and  later
> >> >From PE3  IP1/M2
> >> >
> >> >while having answered with IP1/M1 per proxy alrady. Additionally, in
> >> >such situation ends up seeing
> >> >
> >> >From PE2   IP1/<no MAC>
> >> >
> >> >So the answer it gave is not valid anymore all of a sudden.
> >> >
> >> >--- tony
> _______________________________________________
> BESS mailing list
> BESS@ietf.org
> https://www.ietf.org/mailman/listinfo/bess
>

--001a11c3bcbc3feaaf05127a2566
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_default" style=3D"font-family:courier =
new,monospace;font-size:small">Hi Wim,</div><div class=3D"gmail_default" st=
yle=3D"font-family:courier new,monospace;font-size:small"><br></div><div cl=
ass=3D"gmail_default" style=3D"font-family:courier new,monospace;font-size:=
small">What makes you say that in IPv4 there is no anycast ? All anycase I =
have played so far is IPv4 :)</div><div class=3D"gmail_default" style=3D"fo=
nt-family:courier new,monospace;font-size:small"><br></div><div class=3D"gm=
ail_default" style=3D"font-family:courier new,monospace;font-size:small">Ch=
eers,</div><div class=3D"gmail_default" style=3D"font-family:courier new,mo=
nospace;font-size:small">r.</div></div><div class=3D"gmail_extra"><br><div =
class=3D"gmail_quote">On Sun, Mar 29, 2015 at 11:18 PM, Henderickx, Wim (Wi=
m) <span dir=3D"ltr">&lt;<a href=3D"mailto:wim.henderickx@alcatel-lucent.co=
m" target=3D"_blank">wim.henderickx@alcatel-lucent.com</a>&gt;</span> wrote=
:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-le=
ft:1px #ccc solid;padding-left:1ex">We will update the draft to highlight t=
he IPv6 anycast behaviour better as pointed out by RObert. In IPv4 there is=
 no anycast behaviour and as such there should be one option possible.<br>
<div class=3D"HOEnZb"><div class=3D"h5"><br>
<br>
<br>
On 30/03/15 04:59, &quot;Antoni Przygienda&quot; &lt;<a href=3D"mailto:anto=
ni.przygienda@ericsson.com">antoni.przygienda@ericsson.com</a>&gt; wrote:<b=
r>
<br>
&gt;Yes, but of course I brought it up to show that &#39;the last one simpl=
y wins&#39; as suggested by the draft is not enough IMO. A good architectur=
e should probably keep track of what it served as answer and when the answe=
r is invalid or a new, better one exists, provide a GARP.<br>
&gt;<br>
&gt;As well, when PE2 sends a newer MAC it may not be a good strategy to se=
rve a GARP if PE1&#39;s MAC has already been offered. That could lead IMO t=
o e.g. gateway chasing problems.<br>
&gt;<br>
&gt;--- tony<br>
&gt;<br>
&gt;<br>
&gt;There are basically two types of people. People who accomplish things, =
and people who claim to have accomplished things. The first group is less c=
rowded.<br>
&gt;~~~ Mark Twain<br>
&gt;<br>
&gt;<br>
&gt;&gt; -----Original Message-----<br>
&gt;&gt; From: Henderickx, Wim (Wim) [mailto:<a href=3D"mailto:wim.henderic=
kx@alcatel-lucent.com">wim.henderickx@alcatel-lucent.com</a>]<br>
&gt;&gt; Sent: Thursday, March 26, 2015 6:01 AM<br>
&gt;&gt; To: Antoni Przygienda; Erik Nordmark; Rabadan, Jorge (Jorge)<br>
&gt;&gt; Cc: <a href=3D"mailto:bess@ietf.org">bess@ietf.org</a><br>
&gt;&gt; Subject: Re: [bess] ARP ND draft<br>
&gt;&gt;<br>
&gt;&gt; For this case you should sent a GARP with the new MAC/IP<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; On 25/03/15 18:56, &quot;Antoni Przygienda&quot; &lt;<a href=3D"ma=
ilto:antoni.przygienda@ericsson.com">antoni.przygienda@ericsson.com</a>&gt;=
<br>
&gt;&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt; &gt;&gt; &gt; b)It is worth explaining what is suggested behavior =
if eVPN<br>
&gt;&gt; &gt;&gt; &gt; advertises the same IP with multiple MACs and what h=
appens when<br>
&gt;&gt; &gt;&gt; &gt; e.g. the served MAC vanishes<br>
&gt;&gt; &gt;&gt; &gt;<br>
&gt;&gt; &gt;&gt; Doesn&#39;t the EVPN RFC already stating that the routes =
would be<br>
&gt;&gt; &gt;&gt; withdrawn in that case?<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;The scenario I had in mind was when eVPN PE receives<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;From PE2=C2=A0 IP1/M1=C2=A0 and=C2=A0 later<br>
&gt;&gt; &gt;From PE3=C2=A0 IP1/M2<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;while having answered with IP1/M1 per proxy alrady. Additional=
ly, in<br>
&gt;&gt; &gt;such situation ends up seeing<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;From PE2=C2=A0 =C2=A0IP1/&lt;no MAC&gt;<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;So the answer it gave is not valid anymore all of a sudden.<br=
>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;--- tony<br>
_______________________________________________<br>
BESS mailing list<br>
<a href=3D"mailto:BESS@ietf.org">BESS@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/bess" target=3D"_blank">ht=
tps://www.ietf.org/mailman/listinfo/bess</a><br>
</div></div></blockquote></div><br></div>

--001a11c3bcbc3feaaf05127a2566--


From nobody Sun Mar 29 22:39:05 2015
Return-Path: <wim.henderickx@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B17F11A907B for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 22:39:04 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.309
X-Spam-Level: 
X-Spam-Status: No, score=-6.309 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZrtG1LIlHc59 for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 22:39:02 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2D1461A907D for <bess@ietf.org>; Sun, 29 Mar 2015 22:39:01 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id AF29747F4045D; Mon, 30 Mar 2015 05:38:56 +0000 (GMT)
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2U5cvVP006657 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 30 Mar 2015 07:38:57 +0200
Received: from FR711WXCHMBA07.zeu.alcatel-lucent.com ([169.254.3.17]) by FR711WXCHHUB01.zeu.alcatel-lucent.com ([135.239.2.111]) with mapi id 14.03.0195.001; Mon, 30 Mar 2015 07:38:57 +0200
From: "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
To: Robert Raszuk <robert@raszuk.net>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AQHQZ1QBsq1vN6Z2REGnqTtRu+9A/p0tzyWAgABl5ACABgXxgIAAN3kA///l8ICAADCVpg==
Date: Mon, 30 Mar 2015 05:38:56 +0000
Message-ID: <A606D412-1B8B-4DAB-B4B0-1B19F31FF50B@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se> <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se> <61BFC96E-350C-4F77-84E6-EFA6794B5EC0@alcatel-lucent.com>, <CA+b+ERmVVaz4jm3+CRc7WMfrQn1TN5=sEiK-Qq+nngtw3yw--Q@mail.gmail.com>
In-Reply-To: <CA+b+ERmVVaz4jm3+CRc7WMfrQn1TN5=sEiK-Qq+nngtw3yw--Q@mail.gmail.com>
Accept-Language: nl-BE, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Content-Type: multipart/alternative; boundary="_000_A606D4121B8B4DABB4B01B19F31FF50Balcatellucentcom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/jL5ZRArRsAQ73hKM3bdjO2-7nWw>
Cc: Erik Nordmark <nordmark@acm.org>, Antoni Przygienda <antoni.przygienda@ericsson.com>, "bess@ietf.org" <bess@ietf.org>, "Rabadan, Jorge \(Jorge\)" <jorge.rabadan@alcatel-lucent.com>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 30 Mar 2015 05:39:04 -0000

--_000_A606D4121B8B4DABB4B01B19F31FF50Balcatellucentcom_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

At interface level you get dad in most stacks I know.

Sent from my iPhone

On 30 Mar 2015, at 06:45, Robert Raszuk <robert@raszuk.net<mailto:robert@ra=
szuk.net>> wrote:

Hi Wim,

What makes you say that in IPv4 there is no anycast ? All anycase I have pl=
ayed so far is IPv4 :)

Cheers,
r.

On Sun, Mar 29, 2015 at 11:18 PM, Henderickx, Wim (Wim) <wim.henderickx@alc=
atel-lucent.com<mailto:wim.henderickx@alcatel-lucent.com>> wrote:
We will update the draft to highlight the IPv6 anycast behaviour better as =
pointed out by RObert. In IPv4 there is no anycast behaviour and as such th=
ere should be one option possible.



On 30/03/15 04:59, "Antoni Przygienda" <antoni.przygienda@ericsson.com<mail=
to:antoni.przygienda@ericsson.com>> wrote:

>Yes, but of course I brought it up to show that 'the last one simply wins'=
 as suggested by the draft is not enough IMO. A good architecture should pr=
obably keep track of what it served as answer and when the answer is invali=
d or a new, better one exists, provide a GARP.
>
>As well, when PE2 sends a newer MAC it may not be a good strategy to serve=
 a GARP if PE1's MAC has already been offered. That could lead IMO to e.g. =
gateway chasing problems.
>
>--- tony
>
>
>There are basically two types of people. People who accomplish things, and=
 people who claim to have accomplished things. The first group is less crow=
ded.
>~~~ Mark Twain
>
>
>> -----Original Message-----
>> From: Henderickx, Wim (Wim) [mailto:wim.henderickx@alcatel-lucent.com<ma=
ilto:wim.henderickx@alcatel-lucent.com>]
>> Sent: Thursday, March 26, 2015 6:01 AM
>> To: Antoni Przygienda; Erik Nordmark; Rabadan, Jorge (Jorge)
>> Cc: bess@ietf.org<mailto:bess@ietf.org>
>> Subject: Re: [bess] ARP ND draft
>>
>> For this case you should sent a GARP with the new MAC/IP
>>
>>
>>
>>
>> On 25/03/15 18:56, "Antoni Przygienda" <antoni.przygienda@ericsson.com<m=
ailto:antoni.przygienda@ericsson.com>>
>> wrote:
>>
>> >> > b)It is worth explaining what is suggested behavior if eVPN
>> >> > advertises the same IP with multiple MACs and what happens when
>> >> > e.g. the served MAC vanishes
>> >> >
>> >> Doesn't the EVPN RFC already stating that the routes would be
>> >> withdrawn in that case?
>> >
>> >The scenario I had in mind was when eVPN PE receives
>> >
>> >From PE2  IP1/M1  and  later
>> >From PE3  IP1/M2
>> >
>> >while having answered with IP1/M1 per proxy alrady. Additionally, in
>> >such situation ends up seeing
>> >
>> >From PE2   IP1/<no MAC>
>> >
>> >So the answer it gave is not valid anymore all of a sudden.
>> >
>> >--- tony
_______________________________________________
BESS mailing list
BESS@ietf.org<mailto:BESS@ietf.org>
https://www.ietf.org/mailman/listinfo/bess


--_000_A606D4121B8B4DABB4B01B19F31FF50Balcatellucentcom_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
</head>
<body dir=3D"auto">
<div>At interface level you get dad in most stacks I know.<br>
<br>
Sent from my iPhone</div>
<div><br>
On 30 Mar 2015, at 06:45, Robert Raszuk &lt;<a href=3D"mailto:robert@raszuk=
.net">robert@raszuk.net</a>&gt; wrote:<br>
<br>
</div>
<blockquote type=3D"cite">
<div>
<div dir=3D"ltr">
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
Hi Wim,</div>
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
<br>
</div>
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
What makes you say that in IPv4 there is no anycast ? All anycase I have pl=
ayed so far is IPv4 :)</div>
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
<br>
</div>
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
Cheers,</div>
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
r.</div>
</div>
<div class=3D"gmail_extra"><br>
<div class=3D"gmail_quote">On Sun, Mar 29, 2015 at 11:18 PM, Henderickx, Wi=
m (Wim) <span dir=3D"ltr">
&lt;<a href=3D"mailto:wim.henderickx@alcatel-lucent.com" target=3D"_blank">=
wim.henderickx@alcatel-lucent.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">
We will update the draft to highlight the IPv6 anycast behaviour better as =
pointed out by RObert. In IPv4 there is no anycast behaviour and as such th=
ere should be one option possible.<br>
<div class=3D"HOEnZb">
<div class=3D"h5"><br>
<br>
<br>
On 30/03/15 04:59, &quot;Antoni Przygienda&quot; &lt;<a href=3D"mailto:anto=
ni.przygienda@ericsson.com">antoni.przygienda@ericsson.com</a>&gt; wrote:<b=
r>
<br>
&gt;Yes, but of course I brought it up to show that 'the last one simply wi=
ns' as suggested by the draft is not enough IMO. A good architecture should=
 probably keep track of what it served as answer and when the answer is inv=
alid or a new, better one exists, provide
 a GARP.<br>
&gt;<br>
&gt;As well, when PE2 sends a newer MAC it may not be a good strategy to se=
rve a GARP if PE1's MAC has already been offered. That could lead IMO to e.=
g. gateway chasing problems.<br>
&gt;<br>
&gt;--- tony<br>
&gt;<br>
&gt;<br>
&gt;There are basically two types of people. People who accomplish things, =
and people who claim to have accomplished things. The first group is less c=
rowded.<br>
&gt;~~~ Mark Twain<br>
&gt;<br>
&gt;<br>
&gt;&gt; -----Original Message-----<br>
&gt;&gt; From: Henderickx, Wim (Wim) [mailto:<a href=3D"mailto:wim.henderic=
kx@alcatel-lucent.com">wim.henderickx@alcatel-lucent.com</a>]<br>
&gt;&gt; Sent: Thursday, March 26, 2015 6:01 AM<br>
&gt;&gt; To: Antoni Przygienda; Erik Nordmark; Rabadan, Jorge (Jorge)<br>
&gt;&gt; Cc: <a href=3D"mailto:bess@ietf.org">bess@ietf.org</a><br>
&gt;&gt; Subject: Re: [bess] ARP ND draft<br>
&gt;&gt;<br>
&gt;&gt; For this case you should sent a GARP with the new MAC/IP<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; On 25/03/15 18:56, &quot;Antoni Przygienda&quot; &lt;<a href=3D"ma=
ilto:antoni.przygienda@ericsson.com">antoni.przygienda@ericsson.com</a>&gt;=
<br>
&gt;&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt; &gt;&gt; &gt; b)It is worth explaining what is suggested behavior =
if eVPN<br>
&gt;&gt; &gt;&gt; &gt; advertises the same IP with multiple MACs and what h=
appens when<br>
&gt;&gt; &gt;&gt; &gt; e.g. the served MAC vanishes<br>
&gt;&gt; &gt;&gt; &gt;<br>
&gt;&gt; &gt;&gt; Doesn't the EVPN RFC already stating that the routes woul=
d be<br>
&gt;&gt; &gt;&gt; withdrawn in that case?<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;The scenario I had in mind was when eVPN PE receives<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;From PE2&nbsp; IP1/M1&nbsp; and&nbsp; later<br>
&gt;&gt; &gt;From PE3&nbsp; IP1/M2<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;while having answered with IP1/M1 per proxy alrady. Additional=
ly, in<br>
&gt;&gt; &gt;such situation ends up seeing<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;From PE2&nbsp; &nbsp;IP1/&lt;no MAC&gt;<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;So the answer it gave is not valid anymore all of a sudden.<br=
>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;--- tony<br>
_______________________________________________<br>
BESS mailing list<br>
<a href=3D"mailto:BESS@ietf.org">BESS@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/bess" target=3D"_blank">ht=
tps://www.ietf.org/mailman/listinfo/bess</a><br>
</div>
</div>
</blockquote>
</div>
<br>
</div>
</div>
</blockquote>
</body>
</html>

--_000_A606D4121B8B4DABB4B01B19F31FF50Balcatellucentcom_--


From nobody Sun Mar 29 23:11:09 2015
Return-Path: <wim.henderickx@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3D16A1A90A2 for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 23:11:08 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.309
X-Spam-Level: 
X-Spam-Status: No, score=-6.309 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QXsU-qgw3MCS for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 23:11:05 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpgre-esg-01.alcatel-lucent.com [135.245.210.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8403B1A90A1 for <bess@ietf.org>; Sun, 29 Mar 2015 23:11:03 -0700 (PDT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (unknown [135.239.2.122]) by Websense Email Security Gateway with ESMTPS id 281DE6B982C25; Mon, 30 Mar 2015 06:11:00 +0000 (GMT)
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id t2U6AxBW028315 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 30 Mar 2015 08:11:01 +0200
Received: from FR711WXCHMBA07.zeu.alcatel-lucent.com ([169.254.3.17]) by FR711WXCHHUB02.zeu.alcatel-lucent.com ([135.239.2.112]) with mapi id 14.03.0195.001; Mon, 30 Mar 2015 08:10:59 +0200
From: "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
To: Robert Raszuk <robert@raszuk.net>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AQHQZ1QBsq1vN6Z2REGnqTtRu+9A/p0tzyWAgABl5ACABgXxgIAAN3kA///l8ICAADCVpoAACPKA
Date: Mon, 30 Mar 2015 06:10:58 +0000
Message-ID: <15A9EAE0-D02F-4476-88FC-2E3BD66FA33D@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se> <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se> <61BFC96E-350C-4F77-84E6-EFA6794B5EC0@alcatel-lucent.com> <CA+b+ERmVVaz4jm3+CRc7WMfrQn1TN5=sEiK-Qq+nngtw3yw--Q@mail.gmail.com> <A606D412-1B8B-4DAB-B4B0-1B19F31FF50B@alcatel-lucent.com>
In-Reply-To: <A606D412-1B8B-4DAB-B4B0-1B19F31FF50B@alcatel-lucent.com>
Accept-Language: nl-BE, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/15.8.1.150311
x-originating-ip: [135.239.27.41]
Content-Type: multipart/alternative; boundary="_000_15A9EAE0D02F447688FC2E3BD66FA33Dalcatellucentcom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/l51paACBwtlycN8eFPAp5tjLxcU>
Cc: "Rabadan, Jorge \(Jorge\)" <jorge.rabadan@alcatel-lucent.com>, Erik Nordmark <nordmark@acm.org>, Antoni Przygienda <antoni.przygienda@ericsson.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 30 Mar 2015 06:11:08 -0000

--_000_15A9EAE0D02F447688FC2E3BD66FA33Dalcatellucentcom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_15A9EAE0D02F447688FC2E3BD66FA33Dalcatellucentcom_
Content-Type: text/html; charset="utf-8"
Content-ID: <8BF93658451F984EA0B96B3C98AE68D1@exchange.lucent.com>
Content-Transfer-Encoding: base64
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--_000_15A9EAE0D02F447688FC2E3BD66FA33Dalcatellucentcom_--


From nobody Sun Mar 29 23:17:58 2015
Return-Path: <wim.henderickx@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9201B1A90AA for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 23:17:57 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.309
X-Spam-Level: 
X-Spam-Status: No, score=-6.309 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6ODJ0ZP9Y9r8 for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 23:17:55 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpgre-esg-01.alcatel-lucent.com [135.245.210.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E868A1A90A5 for <bess@ietf.org>; Sun, 29 Mar 2015 23:17:54 -0700 (PDT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (unknown [135.239.2.122]) by Websense Email Security Gateway with ESMTPS id 6CF19B7B02CDE; Mon, 30 Mar 2015 06:17:51 +0000 (GMT)
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id t2U6HqmM000537 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 30 Mar 2015 08:17:52 +0200
Received: from FR711WXCHMBA07.zeu.alcatel-lucent.com ([169.254.3.17]) by FR711WXCHHUB02.zeu.alcatel-lucent.com ([135.239.2.112]) with mapi id 14.03.0195.001; Mon, 30 Mar 2015 08:17:52 +0200
From: "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
To: Robert Raszuk <robert@raszuk.net>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AQHQZ1QBsq1vN6Z2REGnqTtRu+9A/p0tzyWAgABl5ACABgXxgIAAN3kA///l8ICAADCVpoAACPKAgAAB7oA=
Date: Mon, 30 Mar 2015 06:17:51 +0000
Message-ID: <2B1EA39E-0759-4474-A64C-62B80ABC0B58@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se> <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se> <61BFC96E-350C-4F77-84E6-EFA6794B5EC0@alcatel-lucent.com> <CA+b+ERmVVaz4jm3+CRc7WMfrQn1TN5=sEiK-Qq+nngtw3yw--Q@mail.gmail.com> <A606D412-1B8B-4DAB-B4B0-1B19F31FF50B@alcatel-lucent.com> <15A9EAE0-D02F-4476-88FC-2E3BD66FA33D@alcatel-lucent.com>
In-Reply-To: <15A9EAE0-D02F-4476-88FC-2E3BD66FA33D@alcatel-lucent.com>
Accept-Language: nl-BE, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/15.8.1.150311
x-originating-ip: [135.239.27.40]
Content-Type: multipart/alternative; boundary="_000_2B1EA39E07594474A64C62B80ABC0B58alcatellucentcom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/lhl-_lFIBbLS4FnpIb6GkPOoT84>
Cc: "Rabadan, Jorge \(Jorge\)" <jorge.rabadan@alcatel-lucent.com>, Erik Nordmark <nordmark@acm.org>, Antoni Przygienda <antoni.przygienda@ericsson.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 30 Mar 2015 06:17:57 -0000

--_000_2B1EA39E07594474A64C62B80ABC0B58alcatellucentcom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_2B1EA39E07594474A64C62B80ABC0B58alcatellucentcom_
Content-Type: text/html; charset="utf-8"
Content-ID: <B1AEA3C5C7EBD24685D6D1099595271C@exchange.lucent.com>
Content-Transfer-Encoding: base64
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--_000_2B1EA39E07594474A64C62B80ABC0B58alcatellucentcom_--


From nobody Sun Mar 29 23:19:14 2015
Return-Path: <rraszuk@gmail.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 83A271A90AA for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 23:19:13 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.677
X-Spam-Level: 
X-Spam-Status: No, score=-0.677 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, FM_FORGED_GMAIL=0.622, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, J_CHICKENPOX_12=0.6, SPF_PASS=-0.001] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3VpX0cEYVArL for <bess@ietfa.amsl.com>; Sun, 29 Mar 2015 23:19:11 -0700 (PDT)
Received: from mail-ig0-x230.google.com (mail-ig0-x230.google.com [IPv6:2607:f8b0:4001:c05::230]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 65EB31A90A8 for <bess@ietf.org>; Sun, 29 Mar 2015 23:19:11 -0700 (PDT)
Received: by igcau2 with SMTP id au2so66370279igc.1 for <bess@ietf.org>; Sun, 29 Mar 2015 23:19:10 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:sender:in-reply-to:references:date:message-id:subject :from:to:cc:content-type; bh=/X9wCT1gzrQChZlqU/0jrkw7HCJiPpo+u5PSYbBBlqs=; b=i4zZ21m8SQQBnz4yvWoWOnUiCROykKEfTkhdtcMDnX5vq8gFuwhvH6fjcUD6RyxDlh sFOa7SfTO1VDSNweyaGDwylpyOOLo2MLZ7OuVeCxElSjKwzxhts87uKGoH5JjYXK+9yz 75GdpPvxsJbpcjESqOz1rYMEkIt/VJfcgijxHTT0+q39w4Lqz6kr5qoIfDV60nzVAtkX 6zPUsoahyeX5MZ9MmyCkcLEbys8Ao7nlEi/5Rb0ph/iFDKgcFiQGVG6Rw2xtzkTVjdaM PY5S99l/p+Aj5+zNtYgX9qRN6fr2cz6v2UlsqoeiQiUKCHubsxv3ycWbdxyxiRArnYK6 ek3A==
MIME-Version: 1.0
X-Received: by 10.50.66.141 with SMTP id f13mr15516821igt.9.1427696350850; Sun, 29 Mar 2015 23:19:10 -0700 (PDT)
Sender: rraszuk@gmail.com
Received: by 10.107.137.70 with HTTP; Sun, 29 Mar 2015 23:19:10 -0700 (PDT)
In-Reply-To: <15A9EAE0-D02F-4476-88FC-2E3BD66FA33D@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se> <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se> <61BFC96E-350C-4F77-84E6-EFA6794B5EC0@alcatel-lucent.com> <CA+b+ERmVVaz4jm3+CRc7WMfrQn1TN5=sEiK-Qq+nngtw3yw--Q@mail.gmail.com> <A606D412-1B8B-4DAB-B4B0-1B19F31FF50B@alcatel-lucent.com> <15A9EAE0-D02F-4476-88FC-2E3BD66FA33D@alcatel-lucent.com>
Date: Mon, 30 Mar 2015 01:19:10 -0500
X-Google-Sender-Auth: gH4XJ5T2H2or6gBmyfqBhi8e3xI
Message-ID: <CA+b+ER=RsuRBAsB=i+CgyTXOt_D9_J1QtSMWOYzn7moK1CV4DQ@mail.gmail.com>
From: Robert Raszuk <robert@raszuk.net>
To: "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Content-Type: multipart/alternative; boundary=047d7bdc0dd2d7170705127b75da
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/VzZFHvS1sjMD2dz_X9q3A1ZyyiE>
Cc: "Rabadan, Jorge \(Jorge\)" <jorge.rabadan@alcatel-lucent.com>, Erik Nordmark <nordmark@acm.org>, Antoni Przygienda <antoni.przygienda@ericsson.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 30 Mar 2015 06:19:13 -0000

--047d7bdc0dd2d7170705127b75da
Content-Type: text/plain; charset=UTF-8

Hi Wim,

> There is anycast at IPv4 level for sure but I am not ware this is
supported at arp level.

Precisely right. It needs to be documented and addressed if anyone is up to
proposing automated IP duplicate address detection and disabling.

RFC1546 is rather too old to consider here as solution :)

Cheers,
R.


On Mon, Mar 30, 2015 at 1:10 AM, Henderickx, Wim (Wim) <
wim.henderickx@alcatel-lucent.com> wrote:

>  To be clear: RFC4861 section 7.2.7 explains the anycast behaviour in
> IPv6.
> I am not aware of such thing at IPv4/ARP level. Do you have a pointer?
> There is anycast at IPv4 level for sure but I am not ware this is
> supported at arp level.
>
>   From: <Henderickx>, Wim Henderickx
> Date: Monday 30 March 2015 07:38
> To: Robert Raszuk
> Cc: Erik Nordmark, Antoni Przygienda, "bess@ietf.org", Jorge Rabadan
>
> Subject: Re: [bess] ARP ND draft
>
>   At interface level you get dad in most stacks I know.
>
> Sent from my iPhone
>
> On 30 Mar 2015, at 06:45, Robert Raszuk <robert@raszuk.net> wrote:
>
>    Hi Wim,
>
>  What makes you say that in IPv4 there is no anycast ? All anycase I have
> played so far is IPv4 :)
>
>  Cheers,
>  r.
>
> On Sun, Mar 29, 2015 at 11:18 PM, Henderickx, Wim (Wim) <
> wim.henderickx@alcatel-lucent.com> wrote:
>
>> We will update the draft to highlight the IPv6 anycast behaviour better
>> as pointed out by RObert. In IPv4 there is no anycast behaviour and as such
>> there should be one option possible.
>>
>>
>>
>> On 30/03/15 04:59, "Antoni Przygienda" <antoni.przygienda@ericsson.com>
>> wrote:
>>
>> >Yes, but of course I brought it up to show that 'the last one simply
>> wins' as suggested by the draft is not enough IMO. A good architecture
>> should probably keep track of what it served as answer and when the answer
>> is invalid or a new, better one exists, provide a GARP.
>> >
>> >As well, when PE2 sends a newer MAC it may not be a good strategy to
>> serve a GARP if PE1's MAC has already been offered. That could lead IMO to
>> e.g. gateway chasing problems.
>> >
>> >--- tony
>> >
>> >
>> >There are basically two types of people. People who accomplish things,
>> and people who claim to have accomplished things. The first group is less
>> crowded.
>> >~~~ Mark Twain
>> >
>> >
>> >> -----Original Message-----
>> >> From: Henderickx, Wim (Wim) [mailto:wim.henderickx@alcatel-lucent.com]
>> >> Sent: Thursday, March 26, 2015 6:01 AM
>> >> To: Antoni Przygienda; Erik Nordmark; Rabadan, Jorge (Jorge)
>> >> Cc: bess@ietf.org
>> >> Subject: Re: [bess] ARP ND draft
>> >>
>> >> For this case you should sent a GARP with the new MAC/IP
>> >>
>> >>
>> >>
>> >>
>> >> On 25/03/15 18:56, "Antoni Przygienda" <antoni.przygienda@ericsson.com
>> >
>> >> wrote:
>> >>
>> >> >> > b)It is worth explaining what is suggested behavior if eVPN
>> >> >> > advertises the same IP with multiple MACs and what happens when
>> >> >> > e.g. the served MAC vanishes
>> >> >> >
>> >> >> Doesn't the EVPN RFC already stating that the routes would be
>> >> >> withdrawn in that case?
>> >> >
>> >> >The scenario I had in mind was when eVPN PE receives
>> >> >
>> >> >From PE2  IP1/M1  and  later
>> >> >From PE3  IP1/M2
>> >> >
>> >> >while having answered with IP1/M1 per proxy alrady. Additionally, in
>> >> >such situation ends up seeing
>> >> >
>> >> >From PE2   IP1/<no MAC>
>> >> >
>> >> >So the answer it gave is not valid anymore all of a sudden.
>> >> >
>> >> >--- tony
>> _______________________________________________
>> BESS mailing list
>> BESS@ietf.org
>> https://www.ietf.org/mailman/listinfo/bess
>>
>
>

--047d7bdc0dd2d7170705127b75da
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_default" style=3D"font-family:&#39;cou=
rier new&#39;,monospace;font-size:small"><span style=3D"font-family:Calibri=
,sans-serif;font-size:12.8000001907349px">Hi Wim,</span><br></div><div clas=
s=3D"gmail_default" style=3D"font-family:&#39;courier new&#39;,monospace;fo=
nt-size:small"><br></div><div class=3D"gmail_default" style=3D"font-family:=
&#39;courier new&#39;,monospace;font-size:small"><span style=3D"font-family=
:Calibri,sans-serif;font-size:12.8000001907349px">&gt; There is anycast at =
IPv4 level for sure but=C2=A0I am not ware this is supported at arp level.<=
/span><br></div><div class=3D"gmail_default" style=3D"font-family:&#39;cour=
ier new&#39;,monospace;font-size:small"><span style=3D"font-family:Calibri,=
sans-serif;font-size:12.8000001907349px"><br></span></div><div class=3D"gma=
il_default" style=3D"font-family:&#39;courier new&#39;,monospace;font-size:=
small"><span style=3D"font-family:Calibri,sans-serif;font-size:12.800000190=
7349px">Precisely right. It needs to be documented and addressed if anyone =
is up to proposing automated IP duplicate address detection and disabling.<=
/span></div><div class=3D"gmail_default" style=3D"font-family:&#39;courier =
new&#39;,monospace;font-size:small"><span style=3D"font-family:Calibri,sans=
-serif;font-size:12.8000001907349px"><br></span></div><div class=3D"gmail_d=
efault" style=3D"font-family:&#39;courier new&#39;,monospace;font-size:smal=
l"><span style=3D"font-family:Calibri,sans-serif;font-size:12.8000001907349=
px">RFC1546 is rather too old to consider here as solution :)=C2=A0</span><=
/div><div class=3D"gmail_default" style=3D"font-family:&#39;courier new&#39=
;,monospace;font-size:small"><span style=3D"font-family:Calibri,sans-serif;=
font-size:12.8000001907349px"><br></span></div><div class=3D"gmail_default"=
 style=3D"font-family:&#39;courier new&#39;,monospace;font-size:small"><spa=
n style=3D"font-family:Calibri,sans-serif;font-size:12.8000001907349px">Che=
ers,</span></div><div class=3D"gmail_default"><font face=3D"Calibri, sans-s=
erif"><span style=3D"font-size:12.8000001907349px">R.</span></font></div><d=
iv class=3D"gmail_default"><font face=3D"Calibri, sans-serif"><span style=
=3D"font-size:12.8000001907349px"><br></span></font></div></div><div class=
=3D"gmail_extra"><br><div class=3D"gmail_quote">On Mon, Mar 30, 2015 at 1:1=
0 AM, Henderickx, Wim (Wim) <span dir=3D"ltr">&lt;<a href=3D"mailto:wim.hen=
derickx@alcatel-lucent.com" target=3D"_blank">wim.henderickx@alcatel-lucent=
.com</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"ma=
rgin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">



<div style=3D"word-wrap:break-word">
<div>
<div><font face=3D"Calibri,sans-serif">To be clear: RFC4861 section 7.2.7 e=
xplains the anycast behaviour in IPv6.</font></div>
<div><font face=3D"Calibri,sans-serif">I</font><font face=3D"Calibri,sans-s=
erif">=C2=A0am not aware of such thing at IPv4/ARP level. Do you=C2=A0have =
a pointer?</font></div>
<div><font face=3D"Calibri,sans-serif">There is anycast at IPv4 level for s=
ure but=C2=A0I am not ware this is supported at arp level.</font></div>
<div style=3D"color:rgb(0,0,0);font-family:Calibri,sans-serif;font-size:14p=
x">
<div></div>
</div>
</div>
<div style=3D"color:rgb(0,0,0);font-family:Calibri,sans-serif;font-size:14p=
x">
<br>
</div>
<span style=3D"color:rgb(0,0,0);font-family:Calibri,sans-serif;font-size:14=
px">
<div style=3D"font-family:Calibri;font-size:12pt;text-align:left;color:blac=
k;BORDER-BOTTOM:medium none;BORDER-LEFT:medium none;PADDING-BOTTOM:0in;PADD=
ING-LEFT:0in;PADDING-RIGHT:0in;BORDER-TOP:#b5c4df 1pt solid;BORDER-RIGHT:me=
dium none;PADDING-TOP:3pt">
<span style=3D"font-weight:bold">From: </span>&lt;Henderickx&gt;, Wim Hende=
rickx<br>
<span style=3D"font-weight:bold">Date: </span>Monday 30 March 2015 07:38<br=
>
<span style=3D"font-weight:bold">To: </span>Robert Raszuk<br>
<span style=3D"font-weight:bold">Cc: </span>Erik Nordmark, Antoni Przygiend=
a, &quot;<a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.org</=
a>&quot;, Jorge Rabadan<div><div class=3D"h5"><br>
<span style=3D"font-weight:bold">Subject: </span>Re: [bess] ARP ND draft<br=
>
</div></div></div><div><div class=3D"h5">
<div><br>
</div>
<div>
<div dir=3D"auto">
<div>At interface level you get dad in most stacks I know.<br>
<br>
Sent from my iPhone</div>
<div><br>
On 30 Mar 2015, at 06:45, Robert Raszuk &lt;<a href=3D"mailto:robert@raszuk=
.net" target=3D"_blank">robert@raszuk.net</a>&gt; wrote:<br>
<br>
</div>
<blockquote type=3D"cite">
<div>
<div dir=3D"ltr">
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
Hi Wim,</div>
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
<br>
</div>
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
What makes you say that in IPv4 there is no anycast ? All anycase I have pl=
ayed so far is IPv4 :)</div>
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
<br>
</div>
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
Cheers,</div>
<div class=3D"gmail_default" style=3D"font-family:courier new,monospace;fon=
t-size:small">
r.</div>
</div>
<div class=3D"gmail_extra"><br>
<div class=3D"gmail_quote">On Sun, Mar 29, 2015 at 11:18 PM, Henderickx, Wi=
m (Wim) <span dir=3D"ltr">
&lt;<a href=3D"mailto:wim.henderickx@alcatel-lucent.com" target=3D"_blank">=
wim.henderickx@alcatel-lucent.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">
We will update the draft to highlight the IPv6 anycast behaviour better as =
pointed out by RObert. In IPv4 there is no anycast behaviour and as such th=
ere should be one option possible.<br>
<div>
<div><br>
<br>
<br>
On 30/03/15 04:59, &quot;Antoni Przygienda&quot; &lt;<a href=3D"mailto:anto=
ni.przygienda@ericsson.com" target=3D"_blank">antoni.przygienda@ericsson.co=
m</a>&gt; wrote:<br>
<br>
&gt;Yes, but of course I brought it up to show that &#39;the last one simpl=
y wins&#39; as suggested by the draft is not enough IMO. A good architectur=
e should probably keep track of what it served as answer and when the answe=
r is invalid or a new, better one exists, provide
 a GARP.<br>
&gt;<br>
&gt;As well, when PE2 sends a newer MAC it may not be a good strategy to se=
rve a GARP if PE1&#39;s MAC has already been offered. That could lead IMO t=
o e.g. gateway chasing problems.<br>
&gt;<br>
&gt;--- tony<br>
&gt;<br>
&gt;<br>
&gt;There are basically two types of people. People who accomplish things, =
and people who claim to have accomplished things. The first group is less c=
rowded.<br>
&gt;~~~ Mark Twain<br>
&gt;<br>
&gt;<br>
&gt;&gt; -----Original Message-----<br>
&gt;&gt; From: Henderickx, Wim (Wim) [mailto:<a href=3D"mailto:wim.henderic=
kx@alcatel-lucent.com" target=3D"_blank">wim.henderickx@alcatel-lucent.com<=
/a>]<br>
&gt;&gt; Sent: Thursday, March 26, 2015 6:01 AM<br>
&gt;&gt; To: Antoni Przygienda; Erik Nordmark; Rabadan, Jorge (Jorge)<br>
&gt;&gt; Cc: <a href=3D"mailto:bess@ietf.org" target=3D"_blank">bess@ietf.o=
rg</a><br>
&gt;&gt; Subject: Re: [bess] ARP ND draft<br>
&gt;&gt;<br>
&gt;&gt; For this case you should sent a GARP with the new MAC/IP<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; On 25/03/15 18:56, &quot;Antoni Przygienda&quot; &lt;<a href=3D"ma=
ilto:antoni.przygienda@ericsson.com" target=3D"_blank">antoni.przygienda@er=
icsson.com</a>&gt;<br>
&gt;&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt; &gt;&gt; &gt; b)It is worth explaining what is suggested behavior =
if eVPN<br>
&gt;&gt; &gt;&gt; &gt; advertises the same IP with multiple MACs and what h=
appens when<br>
&gt;&gt; &gt;&gt; &gt; e.g. the served MAC vanishes<br>
&gt;&gt; &gt;&gt; &gt;<br>
&gt;&gt; &gt;&gt; Doesn&#39;t the EVPN RFC already stating that the routes =
would be<br>
&gt;&gt; &gt;&gt; withdrawn in that case?<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;The scenario I had in mind was when eVPN PE receives<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;From PE2=C2=A0 IP1/M1=C2=A0 and=C2=A0 later<br>
&gt;&gt; &gt;From PE3=C2=A0 IP1/M2<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;while having answered with IP1/M1 per proxy alrady. Additional=
ly, in<br>
&gt;&gt; &gt;such situation ends up seeing<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;From PE2=C2=A0 =C2=A0IP1/&lt;no MAC&gt;<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;So the answer it gave is not valid anymore all of a sudden.<br=
>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;--- tony<br>
_______________________________________________<br>
BESS mailing list<br>
<a href=3D"mailto:BESS@ietf.org" target=3D"_blank">BESS@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/bess" target=3D"_blank">ht=
tps://www.ietf.org/mailman/listinfo/bess</a><br>
</div>
</div>
</blockquote>
</div>
<br>
</div>
</div>
</blockquote>
</div>
</div>
</div></div></span>
</div>

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

--047d7bdc0dd2d7170705127b75da--


From nobody Mon Mar 30 00:12:19 2015
Return-Path: <antoni.przygienda@ericsson.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 038321A90F0 for <bess@ietfa.amsl.com>; Mon, 30 Mar 2015 00:12:18 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.6
X-Spam-Level: 
X-Spam-Status: No, score=-3.6 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uhTOtFovJq68 for <bess@ietfa.amsl.com>; Mon, 30 Mar 2015 00:12:15 -0700 (PDT)
Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CA3251A90E8 for <bess@ietf.org>; Mon, 30 Mar 2015 00:12:14 -0700 (PDT)
X-AuditID: c618062d-f79686d0000030a8-64-5518a146f5a4
Received: from EUSAAHC004.ericsson.se (Unknown_Domain [147.117.188.84]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id 7E.E9.12456.641A8155; Mon, 30 Mar 2015 03:05:10 +0200 (CEST)
Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC004.ericsson.se ([147.117.188.84]) with mapi id 14.03.0210.002; Mon, 30 Mar 2015 03:12:12 -0400
From: Antoni Przygienda <antoni.przygienda@ericsson.com>
To: Robert Raszuk <robert@raszuk.net>, "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAZ/NmAAAeuvZAAEFu+gAAIR4HQALLTFoAAAO63gAAB4cEAAAEeZwAAAElQAAAHRjGA
Date: Mon, 30 Mar 2015 07:12:12 +0000
Message-ID: <2E4BB27CAB87BF43B4207C0E55860F1828B205@eusaamb103.ericsson.se>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se> <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se> <61BFC96E-350C-4F77-84E6-EFA6794B5EC0@alcatel-lucent.com> <CA+b+ERmVVaz4jm3+CRc7WMfrQn1TN5=sEiK-Qq+nngtw3yw--Q@mail.gmail.com> <A606D412-1B8B-4DAB-B4B0-1B19F31FF50B@alcatel-lucent.com> <15A9EAE0-D02F-4476-88FC-2E3BD66FA33D@alcatel-lucent.com> <CA+b+ER=RsuRBAsB=i+CgyTXOt_D9_J1QtSMWOYzn7moK1CV4DQ@mail.gmail.com>
In-Reply-To: <CA+b+ER=RsuRBAsB=i+CgyTXOt_D9_J1QtSMWOYzn7moK1CV4DQ@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [147.117.188.9]
Content-Type: multipart/alternative; boundary="_000_2E4BB27CAB87BF43B4207C0E55860F1828B205eusaamb103ericsso_"
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFprAIsWRmVeSWpSXmKPExsUyuXRPiK7bQolQg3PNthYrjs9ktvh2aQmL Rf/UnUwWTQubmC1mP/7B5sDqcfmKt0frs72sHkuW/GTy2L1xAVMASxSXTUpqTmZZapG+XQJX xsGO1II9zxgren9NY2pg3PKAsYuRk0NCwETi/PnVLBC2mMSFe+vZuhi5OIQEjjJKtJxewwjh LGeU2HjtIRtIFZuAhcTlb0+ZQWwRgRSJLf27mUGKmAVaGCUWzT4ENkpYQEGiveMrVJGixPzn 6xgh7DyJ9582gNWwCKhKHOn9ADaUV8Bb4uaSjWBxIYFXLBJrt3mB2JwCgRKPn/WCxRmBzvt+ ag0TiM0sIC5x68l8JoizBSSW7DnPDGGLSrx8/I8VwlaU2Nc/nb2LkQOoPl/i6YYSiFWCEidn PmGZwCg6C8mkWQhVs5BUQYQ1Jdbv0oeoVpSY0v2QHcLWkGidM5cdWXwBI/sqRo7S4tSy3HQj g02MwCg8JsGmu4Nxz0vLQ4wCHIxKPLwLbCVChVgTy4orcw8xSnOwKInzLnpwMERIID2xJDU7 NbUgtSi+qDQntfgQIxMHp1QDY9jlsr8x7u5fgmZffFnkqSr5neXou8o34U3Pl9uE7xQ46v+6 Pul55m/rDIXvKzMYfj52+3w0d87DP0tOqs+uLtp+s2J+eU7qucqOdV2hhwxWv531sHxfyPfn RvK56to8Tu7vzV+VnzjRZFe+jF3twoJDskUiGSGX9WZ0mX2aEiKfxP/9M2/dHSWW4oxEQy3m ouJEAH5ImcmjAgAA
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/PPT1C62-pBDRqHlG0SZKeQg34lo>
Cc: Erik Nordmark <nordmark@acm.org>, "Rabadan, Jorge \(Jorge\)" <jorge.rabadan@alcatel-lucent.com>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 30 Mar 2015 07:12:18 -0000

--_000_2E4BB27CAB87BF43B4207C0E55860F1828B205eusaamb103ericsso_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_2E4BB27CAB87BF43B4207C0E55860F1828B205eusaamb103ericsso_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64

PGh0bWwgeG1sbnM6dj0idXJuOnNjaGVtYXMtbWljcm9zb2Z0LWNvbTp2bWwiIHhtbG5zOm89InVy
bjpzY2hlbWFzLW1pY3Jvc29mdC1jb206b2ZmaWNlOm9mZmljZSIgeG1sbnM6dz0idXJuOnNjaGVt
YXMtbWljcm9zb2Z0LWNvbTpvZmZpY2U6d29yZCIgeG1sbnM6bT0iaHR0cDovL3NjaGVtYXMubWlj
cm9zb2Z0LmNvbS9vZmZpY2UvMjAwNC8xMi9vbW1sIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcv
VFIvUkVDLWh0bWw0MCI+DQo8aGVhZD4NCjxtZXRhIGh0dHAtZXF1aXY9IkNvbnRlbnQtVHlwZSIg
Y29udGVudD0idGV4dC9odG1sOyBjaGFyc2V0PXV0Zi04Ij4NCjxtZXRhIG5hbWU9IkdlbmVyYXRv
ciIgY29udGVudD0iTWljcm9zb2Z0IFdvcmQgMTQgKGZpbHRlcmVkIG1lZGl1bSkiPg0KPHN0eWxl
PjwhLS0NCi8qIEZvbnQgRGVmaW5pdGlvbnMgKi8NCkBmb250LWZhY2UNCgl7Zm9udC1mYW1pbHk6
Q2FsaWJyaTsNCglwYW5vc2UtMToyIDE1IDUgMiAyIDIgNCAzIDIgNDt9DQpAZm9udC1mYWNlDQoJ
e2ZvbnQtZmFtaWx5OlRhaG9tYTsNCglwYW5vc2UtMToyIDExIDYgNCAzIDUgNCA0IDIgNDt9DQpA
Zm9udC1mYWNlDQoJe2ZvbnQtZmFtaWx5OiJCb29rbWFuIE9sZCBTdHlsZSI7DQoJcGFub3NlLTE6
MiA1IDYgNCA1IDUgNSAyIDIgNDt9DQovKiBTdHlsZSBEZWZpbml0aW9ucyAqLw0KcC5Nc29Ob3Jt
YWwsIGxpLk1zb05vcm1hbCwgZGl2Lk1zb05vcm1hbA0KCXttYXJnaW46MGluOw0KCW1hcmdpbi1i
b3R0b206LjAwMDFwdDsNCglmb250LXNpemU6MTIuMHB0Ow0KCWZvbnQtZmFtaWx5OiJUaW1lcyBO
ZXcgUm9tYW4iLCJzZXJpZiI7fQ0KYTpsaW5rLCBzcGFuLk1zb0h5cGVybGluaw0KCXttc28tc3R5
bGUtcHJpb3JpdHk6OTk7DQoJY29sb3I6Ymx1ZTsNCgl0ZXh0LWRlY29yYXRpb246dW5kZXJsaW5l
O30NCmE6dmlzaXRlZCwgc3Bhbi5Nc29IeXBlcmxpbmtGb2xsb3dlZA0KCXttc28tc3R5bGUtcHJp
b3JpdHk6OTk7DQoJY29sb3I6cHVycGxlOw0KCXRleHQtZGVjb3JhdGlvbjp1bmRlcmxpbmU7fQ0K
cC5Nc29BY2V0YXRlLCBsaS5Nc29BY2V0YXRlLCBkaXYuTXNvQWNldGF0ZQ0KCXttc28tc3R5bGUt
cHJpb3JpdHk6OTk7DQoJbXNvLXN0eWxlLWxpbms6IkJhbGxvb24gVGV4dCBDaGFyIjsNCgltYXJn
aW46MGluOw0KCW1hcmdpbi1ib3R0b206LjAwMDFwdDsNCglmb250LXNpemU6OC4wcHQ7DQoJZm9u
dC1mYW1pbHk6IlRhaG9tYSIsInNhbnMtc2VyaWYiO30NCnNwYW4uRW1haWxTdHlsZTE3DQoJe21z
by1zdHlsZS10eXBlOnBlcnNvbmFsLXJlcGx5Ow0KCWZvbnQtZmFtaWx5OiJDYWxpYnJpIiwic2Fu
cy1zZXJpZiI7DQoJY29sb3I6IzFGNDk3RDt9DQpzcGFuLkJhbGxvb25UZXh0Q2hhcg0KCXttc28t
c3R5bGUtbmFtZToiQmFsbG9vbiBUZXh0IENoYXIiOw0KCW1zby1zdHlsZS1wcmlvcml0eTo5OTsN
Cgltc28tc3R5bGUtbGluazoiQmFsbG9vbiBUZXh0IjsNCglmb250LWZhbWlseToiVGFob21hIiwi
c2Fucy1zZXJpZiI7fQ0KLk1zb0NocERlZmF1bHQNCgl7bXNvLXN0eWxlLXR5cGU6ZXhwb3J0LW9u
bHk7DQoJZm9udC1mYW1pbHk6IkNhbGlicmkiLCJzYW5zLXNlcmlmIjt9DQpAcGFnZSBXb3JkU2Vj
dGlvbjENCgl7c2l6ZTo4LjVpbiAxMS4waW47DQoJbWFyZ2luOjEuMGluIDEuMGluIDEuMGluIDEu
MGluO30NCmRpdi5Xb3JkU2VjdGlvbjENCgl7cGFnZTpXb3JkU2VjdGlvbjE7fQ0KLS0+PC9zdHls
ZT48IS0tW2lmIGd0ZSBtc28gOV0+PHhtbD4NCjxvOnNoYXBlZGVmYXVsdHMgdjpleHQ9ImVkaXQi
IHNwaWRtYXg9IjEwMjYiIC8+DQo8L3htbD48IVtlbmRpZl0tLT48IS0tW2lmIGd0ZSBtc28gOV0+
PHhtbD4NCjxvOnNoYXBlbGF5b3V0IHY6ZXh0PSJlZGl0Ij4NCjxvOmlkbWFwIHY6ZXh0PSJlZGl0
IiBkYXRhPSIxIiAvPg0KPC9vOnNoYXBlbGF5b3V0PjwveG1sPjwhW2VuZGlmXS0tPg0KPC9oZWFk
Pg0KPGJvZHkgbGFuZz0iRU4tVVMiIGxpbms9ImJsdWUiIHZsaW5rPSJwdXJwbGUiPg0KPGRpdiBj
bGFzcz0iV29yZFNlY3Rpb24xIj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJm
b250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fu
cy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5J4oCZbSBhbHNvIHNrZXB0aWNhbCB3aGV0aGVy
IElQIGR1cGxpY2F0ZSBkZXRlY3Rpb24gd291bGQgYmUgYSBnb29kIGRlZmF1bHQgdGhpbmcuIEVz
cGVjaWFsbHkgaW4gY2FzZSBvZiB3aGF0IEkgY2FsbCDigJhhbGlhc2VkIGRlZmF1bHQgZ2F0ZXdh
eeKAmSB3aGljaCBzZWN0aW9uIDEwLjENCiBzcGVjaWZpY2FsbHkgYWxsb3dzLCBpLmUuIGRlZmF1
bHQgR1cgSVAgYWRkcmVzcyBpcyBzYW1lIGJ1dCBlYWNoIFBFIG1heSB1c2UgYSBkaWZmZXJlbnQg
TUFDIHdoZW4gYWR2ZXJ0aXNpbmcgaXQgYW5kIGNvbnNlcXVlbnRseSByZXNwb25zZXMgZm9yIHNh
bWUgSVAgd2l0aCBkaWZmZXJlbnQgQVJQcyBtYXkgYmUgc2VlbiBpbiB0aGUgbmV0d29yay4gJm5i
c3A7WWVzLCBkZWZhdWx0IEdXIEV4dENvbW0gaXMgdGhlcmUgdG8gZGlmZmVyZW50aWF0ZSBzbyBp
dA0KIGNhbiBiZSBjYWxsZWQgYW4gZXhjZXB0aW9uIGJ1dCBuZXZlcnRoZWxlc3MuIDxvOnA+PC9v
OnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LXNp
emU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJp
ZiZxdW90Oztjb2xvcjojMUY0OTdEIj48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBj
bGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWls
eTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3
RCI+SSBhbHNvIHRob3VnaHQgYSB0YWQgYWJvdXQgVlJSUCBidXQgSSB0aGluayB0aGUgSVAgZHVw
bGljYXRlIGRldGVjdGlvbiB3aWxsIG5vdCBhcHBseSB0aGVyZSwgaXTigJlzIGFsbCBzYW1lIElQ
eC0mZ3Q7TUFDeCBmcm9tIGFsbCByb3V0ZXJzIHNvIGlmIGFueXRoaW5nLCBpdOKAmXMgbW9yZQ0K
IG9mIGEgTUFDIG1vdmUgdGhpbmcuIDxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJN
c29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90
O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj48bzpw
PiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHls
ZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90
O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+R2VuZXJhbGx5IEkgdGhpbmsgc29tZW9u
ZSB3aG8gd2FudHMgYSBzZWN1cmUsIHN0YWJsZSBlVlBOIHdhbnRzIElQIGR1cGxpY2F0ZSBkZXRl
Y3Rpb24sIHNvbWVvbmUgd2hvIHJ1bnMgYSB2ZXJ5IGR5bmFtaWMgbmV0d29yayB3aXRoIHRvbnMg
Z2F0ZXdheXMsIHBvc3NpYmx5DQogYW55Y2FzdCAmYW1wOyBmbG9hdGluZyBJUHMgd2lsbCBwcm9i
YWJseSBub3QgYmUgdG9vIGVuYW1vcmVkIHdpdGggaXQuIDxvOnA+PC9vOnA+PC9zcGFuPjwvcD4N
CjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQt
ZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjoj
MUY0OTdEIj48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFs
Ij48c3BhbiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJp
JnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+VGhhbmtzDQo8bzpw
PjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9u
dC1zaXplOjExLjBwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMt
c2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+PG86cD4mbmJzcDs8L286cD48L3NwYW4+PC9wPg0K
PHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1m
YW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMx
RjQ5N0QiPi0tLSB0b255DQo8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8ZGl2IHN0eWxlPSJtc28t
ZWxlbWVudDpwYXJhLWJvcmRlci1kaXY7Ym9yZGVyOm5vbmU7Ym9yZGVyLWJvdHRvbTpzb2xpZCB3
aW5kb3d0ZXh0IDEuMHB0O3BhZGRpbmc6MGluIDBpbiAxLjBwdCAwaW4iPg0KPHAgY2xhc3M9Ik1z
b05vcm1hbCIgc3R5bGU9ImJvcmRlcjpub25lO3BhZGRpbmc6MGluIj48aT48c3BhbiBzdHlsZT0i
Zm9udC1zaXplOjEwLjBwdDtmb250LWZhbWlseTomcXVvdDtCb29rbWFuIE9sZCBTdHlsZSZxdW90
OywmcXVvdDtzZXJpZiZxdW90Oztjb2xvcjojQTZBNkE2Ij48bzpwPiZuYnNwOzwvbzpwPjwvc3Bh
bj48L2k+PC9wPg0KPC9kaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48aT48c3BhbiBzdHlsZT0i
Zm9udC1zaXplOjkuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0Jvb2ttYW4gT2xkIFN0eWxlJnF1b3Q7
LCZxdW90O3NlcmlmJnF1b3Q7O2NvbG9yOiNBNkE2QTYiPjxhIGhyZWY9Imh0dHA6Ly93d3cuYnJh
aW55cXVvdGUuY29tL3F1b3Rlcy9xdW90ZXMvbS9tYXJrdHdhaW4zOTM1MzUuaHRtbCIgdGl0bGU9
InZpZXcgcXVvdGUiPjxzcGFuIHN0eWxlPSJjb2xvcjojQTZBNkE2O3RleHQtZGVjb3JhdGlvbjpu
b25lIj5UaGVyZSBhcmUNCiBiYXNpY2FsbHkgdHdvIHR5cGVzIG9mIHBlb3BsZS4gUGVvcGxlIHdo
byBhY2NvbXBsaXNoIHRoaW5ncywgYW5kIHBlb3BsZSB3aG8gY2xhaW0gdG8gaGF2ZSBhY2NvbXBs
aXNoZWQgdGhpbmdzLiBUaGUgZmlyc3QgZ3JvdXAgaXMgbGVzcyBjcm93ZGVkLjwvc3Bhbj48L2E+
PG86cD48L286cD48L3NwYW4+PC9pPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxpPjxzcGFu
IHN0eWxlPSJmb250LXNpemU6OC4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Qm9va21hbiBPbGQgU3R5
bGUmcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDs7Y29sb3I6I0E2QTZBNiI+fn5+DQo8YSBocmVmPSJo
dHRwOi8vd3d3LmJyYWlueXF1b3RlLmNvbS9xdW90ZXMvcXVvdGVzL20vbWFya3R3YWluMzkzNTM1
Lmh0bWwiIHRpdGxlPSJ2aWV3IHF1b3RlIj4NCjxzcGFuIHN0eWxlPSJjb2xvcjojQTZBNkE2O3Rl
eHQtZGVjb3JhdGlvbjpub25lIj5NYXJrIFR3YWluPC9zcGFuPjwvYT48bzpwPjwvbzpwPjwvc3Bh
bj48L2k+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZTox
MS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1
b3Q7O2NvbG9yOiMxRjQ5N0QiPjxvOnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxkaXYgc3R5
bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNvbGlkIGJsdWUgMS41cHQ7cGFkZGluZzowaW4g
MGluIDBpbiA0LjBwdCI+DQo8ZGl2Pg0KPGRpdiBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLXRv
cDpzb2xpZCAjQjVDNERGIDEuMHB0O3BhZGRpbmc6My4wcHQgMGluIDBpbiAwaW4iPg0KPHAgY2xh
c3M9Ik1zb05vcm1hbCI+PGI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZToxMC4wcHQ7Zm9udC1mYW1p
bHk6JnF1b3Q7VGFob21hJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPkZyb206PC9zcGFu
PjwvYj48c3BhbiBzdHlsZT0iZm9udC1zaXplOjEwLjBwdDtmb250LWZhbWlseTomcXVvdDtUYWhv
bWEmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+IHJyYXN6dWtAZ21haWwuY29tIFttYWls
dG86cnJhc3p1a0BnbWFpbC5jb21dDQo8Yj5PbiBCZWhhbGYgT2YgPC9iPlJvYmVydCBSYXN6dWs8
YnI+DQo8Yj5TZW50OjwvYj4gTW9uZGF5LCBNYXJjaCAzMCwgMjAxNSAxOjE5IEFNPGJyPg0KPGI+
VG86PC9iPiBIZW5kZXJpY2t4LCBXaW0gKFdpbSk8YnI+DQo8Yj5DYzo8L2I+IEVyaWsgTm9yZG1h
cms7IEFudG9uaSBQcnp5Z2llbmRhOyBiZXNzQGlldGYub3JnOyBSYWJhZGFuLCBKb3JnZSAoSm9y
Z2UpPGJyPg0KPGI+U3ViamVjdDo8L2I+IFJlOiBbYmVzc10gQVJQIE5EIGRyYWZ0PG86cD48L286
cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8L2Rpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+
Jm5ic3A7PC9vOnA+PC9wPg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3Bh
biBzdHlsZT0iZm9udC1zaXplOjkuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDss
JnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+SGkgV2ltLDwvc3Bhbj48c3BhbiBzdHlsZT0iZm9udC1m
YW1pbHk6JnF1b3Q7Q291cmllciBOZXcmcXVvdDsiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwv
ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LWZhbWls
eTomcXVvdDtDb3VyaWVyIE5ldyZxdW90OyI+PG86cD4mbmJzcDs8L286cD48L3NwYW4+PC9wPg0K
PC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6
ZTo5LjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYm
cXVvdDsiPiZndDsgVGhlcmUgaXMgYW55Y2FzdCBhdCBJUHY0IGxldmVsIGZvciBzdXJlIGJ1dCZu
YnNwO0kgYW0gbm90IHdhcmUgdGhpcyBpcyBzdXBwb3J0ZWQgYXQgYXJwIGxldmVsLjwvc3Bhbj48
c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7Q291cmllciBOZXcmcXVvdDsiPjxvOnA+PC9v
OnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFu
IHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtDb3VyaWVyIE5ldyZxdW90OyI+PG86cD4mbmJzcDs8
L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNw
YW4gc3R5bGU9ImZvbnQtc2l6ZTo5LjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7
LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPlByZWNpc2VseSByaWdodC4gSXQgbmVlZHMgdG8gYmUg
ZG9jdW1lbnRlZCBhbmQgYWRkcmVzc2VkIGlmIGFueW9uZSBpcyB1cCB0byBwcm9wb3NpbmcgYXV0
b21hdGVkIElQIGR1cGxpY2F0ZSBhZGRyZXNzIGRldGVjdGlvbiBhbmQgZGlzYWJsaW5nLjwvc3Bh
bj48c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7Q291cmllciBOZXcmcXVvdDsiPjxvOnA+
PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxz
cGFuIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtDb3VyaWVyIE5ldyZxdW90OyI+PG86cD4mbmJz
cDs8L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+
PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZTo5LjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1
b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPlJGQzE1NDYgaXMgcmF0aGVyIHRvbyBvbGQgdG8g
Y29uc2lkZXIgaGVyZSBhcyBzb2x1dGlvbiA6KSZuYnNwOzwvc3Bhbj48c3BhbiBzdHlsZT0iZm9u
dC1mYW1pbHk6JnF1b3Q7Q291cmllciBOZXcmcXVvdDsiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4N
CjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LWZh
bWlseTomcXVvdDtDb3VyaWVyIE5ldyZxdW90OyI+PG86cD4mbmJzcDs8L286cD48L3NwYW4+PC9w
Pg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQt
c2l6ZTo5LjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2Vy
aWYmcXVvdDsiPkNoZWVycyw8L3NwYW4+PHNwYW4gc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O0Nv
dXJpZXIgTmV3JnF1b3Q7Ij48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8
cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9udC1zaXplOjkuNXB0O2ZvbnQtZmFt
aWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+Ui48L3NwYW4+
PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48bzpw
PiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v
cm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+
T24gTW9uLCBNYXIgMzAsIDIwMTUgYXQgMToxMCBBTSwgSGVuZGVyaWNreCwgV2ltIChXaW0pICZs
dDs8YSBocmVmPSJtYWlsdG86d2ltLmhlbmRlcmlja3hAYWxjYXRlbC1sdWNlbnQuY29tIiB0YXJn
ZXQ9Il9ibGFuayI+d2ltLmhlbmRlcmlja3hAYWxjYXRlbC1sdWNlbnQuY29tPC9hPiZndDsgd3Jv
dGU6PG86cD48L286cD48L3A+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9y
bWFsIj48c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtz
YW5zLXNlcmlmJnF1b3Q7Ij5UbyBiZSBjbGVhcjogUkZDNDg2MSBzZWN0aW9uIDcuMi43IGV4cGxh
aW5zIHRoZSBhbnljYXN0IGJlaGF2aW91ciBpbiBJUHY2Ljwvc3Bhbj48bzpwPjwvbzpwPjwvcD4N
CjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LWZh
bWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPkkmbmJzcDth
bSBub3QgYXdhcmUgb2Ygc3VjaCB0aGluZyBhdCBJUHY0L0FSUCBsZXZlbC4gRG8geW91Jm5ic3A7
aGF2ZSBhIHBvaW50ZXI/PC9zcGFuPjxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAg
Y2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkm
cXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+VGhlcmUgaXMgYW55Y2FzdCBhdCBJUHY0IGxl
dmVsIGZvciBzdXJlIGJ1dCZuYnNwO0kgYW0gbm90IHdhcmUgdGhpcyBpcyBzdXBwb3J0ZWQgYXQg
YXJwIGxldmVsLjwvc3Bhbj48bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8ZGl2Pg0K
PHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1m
YW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOmJs
YWNrIj48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXYgc3R5bGU9ImJv
cmRlcjpub25lO2JvcmRlci10b3A6c29saWQgI0I1QzRERiAxLjBwdDtwYWRkaW5nOjMuMHB0IDBp
biAwaW4gMGluIj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxiPjxzcGFuIHN0eWxlPSJmb250LWZh
bWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6Ymxh
Y2siPkZyb206DQo8L3NwYW4+PC9iPjxzcGFuIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtDYWxp
YnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6YmxhY2siPiZsdDtIZW5kZXJp
Y2t4Jmd0OywgV2ltIEhlbmRlcmlja3g8YnI+DQo8Yj5EYXRlOiA8L2I+TW9uZGF5IDMwIE1hcmNo
IDIwMTUgMDc6Mzg8YnI+DQo8Yj5UbzogPC9iPlJvYmVydCBSYXN6dWs8YnI+DQo8Yj5DYzogPC9i
PkVyaWsgTm9yZG1hcmssIEFudG9uaSBQcnp5Z2llbmRhLCAmcXVvdDs8YSBocmVmPSJtYWlsdG86
YmVzc0BpZXRmLm9yZyIgdGFyZ2V0PSJfYmxhbmsiPmJlc3NAaWV0Zi5vcmc8L2E+JnF1b3Q7LCBK
b3JnZSBSYWJhZGFuPG86cD48L286cD48L3NwYW4+PC9wPg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFz
cz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90
OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOmJsYWNrIj48YnI+DQo8Yj5TdWJqZWN0OiA8
L2I+UmU6IFtiZXNzXSBBUlAgTkQgZHJhZnQ8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4N
CjwvZGl2Pg0KPC9kaXY+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs
Ij48c3BhbiBzdHlsZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJp
JnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6YmxhY2siPjxvOnA+Jm5ic3A7PC9v
OnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1z
b05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7
Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOmJsYWNrIj5BdCBpbnRl
cmZhY2UgbGV2ZWwgeW91IGdldCBkYWQgaW4gbW9zdCBzdGFja3MgSSBrbm93Ljxicj4NCjxicj4N
ClNlbnQgZnJvbSBteSBpUGhvbmU8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+
DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibWFyZ2luLWJvdHRvbToxMi4wcHQiPjxzcGFu
IHN0eWxlPSJmb250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDss
JnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjpibGFjayI+PGJyPg0KT24gMzAgTWFyIDIwMTUs
IGF0IDA2OjQ1LCBSb2JlcnQgUmFzenVrICZsdDs8YSBocmVmPSJtYWlsdG86cm9iZXJ0QHJhc3p1
ay5uZXQiIHRhcmdldD0iX2JsYW5rIj5yb2JlcnRAcmFzenVrLm5ldDwvYT4mZ3Q7IHdyb3RlOjxv
OnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGJsb2NrcXVvdGUgc3R5bGU9Im1hcmdpbi10
b3A6NS4wcHQ7bWFyZ2luLWJvdHRvbTo1LjBwdCI+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBj
bGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7Q291cmllciBO
ZXcmcXVvdDs7Y29sb3I6YmxhY2siPkhpIFdpbSw8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rp
dj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6
JnF1b3Q7Q291cmllciBOZXcmcXVvdDs7Y29sb3I6YmxhY2siPjxvOnA+Jm5ic3A7PC9vOnA+PC9z
cGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxl
PSJmb250LWZhbWlseTomcXVvdDtDb3VyaWVyIE5ldyZxdW90Oztjb2xvcjpibGFjayI+V2hhdCBt
YWtlcyB5b3Ugc2F5IHRoYXQgaW4gSVB2NCB0aGVyZSBpcyBubyBhbnljYXN0ID8gQWxsIGFueWNh
c2UgSSBoYXZlIHBsYXllZCBzbyBmYXIgaXMgSVB2NCA6KTxvOnA+PC9vOnA+PC9zcGFuPjwvcD4N
CjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LWZh
bWlseTomcXVvdDtDb3VyaWVyIE5ldyZxdW90Oztjb2xvcjpibGFjayI+PG86cD4mbmJzcDs8L286
cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4g
c3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O0NvdXJpZXIgTmV3JnF1b3Q7O2NvbG9yOmJsYWNrIj5D
aGVlcnMsPG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1z
b05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O0NvdXJpZXIgTmV3JnF1b3Q7
O2NvbG9yOmJsYWNrIj5yLjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8
ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7
Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2Nv
bG9yOmJsYWNrIj48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8ZGl2Pg0KPHAgY2xhc3M9
Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1
b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOmJsYWNrIj5PbiBT
dW4sIE1hciAyOSwgMjAxNSBhdCAxMToxOCBQTSwgSGVuZGVyaWNreCwgV2ltIChXaW0pICZsdDs8
YSBocmVmPSJtYWlsdG86d2ltLmhlbmRlcmlja3hAYWxjYXRlbC1sdWNlbnQuY29tIiB0YXJnZXQ9
Il9ibGFuayI+d2ltLmhlbmRlcmlja3hAYWxjYXRlbC1sdWNlbnQuY29tPC9hPiZndDsNCiB3cm90
ZTo8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHls
ZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90
O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6YmxhY2siPldlIHdpbGwgdXBkYXRlIHRoZSBkcmFmdCB0
byBoaWdobGlnaHQgdGhlIElQdjYgYW55Y2FzdCBiZWhhdmlvdXIgYmV0dGVyIGFzIHBvaW50ZWQg
b3V0IGJ5IFJPYmVydC4gSW4gSVB2NCB0aGVyZSBpcyBubyBhbnljYXN0IGJlaGF2aW91ciBhbmQg
YXMgc3VjaCB0aGVyZSBzaG91bGQNCiBiZSBvbmUgb3B0aW9uIHBvc3NpYmxlLjxvOnA+PC9vOnA+
PC9zcGFuPjwvcD4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5
bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVv
dDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOmJsYWNrIj48YnI+DQo8YnI+DQo8YnI+DQpPbiAzMC8w
My8xNSAwNDo1OSwgJnF1b3Q7QW50b25pIFByenlnaWVuZGEmcXVvdDsgJmx0OzxhIGhyZWY9Im1h
aWx0bzphbnRvbmkucHJ6eWdpZW5kYUBlcmljc3Nvbi5jb20iIHRhcmdldD0iX2JsYW5rIj5hbnRv
bmkucHJ6eWdpZW5kYUBlcmljc3Nvbi5jb208L2E+Jmd0OyB3cm90ZTo8YnI+DQo8YnI+DQomZ3Q7
WWVzLCBidXQgb2YgY291cnNlIEkgYnJvdWdodCBpdCB1cCB0byBzaG93IHRoYXQgJ3RoZSBsYXN0
IG9uZSBzaW1wbHkgd2lucycgYXMgc3VnZ2VzdGVkIGJ5IHRoZSBkcmFmdCBpcyBub3QgZW5vdWdo
IElNTy4gQSBnb29kIGFyY2hpdGVjdHVyZSBzaG91bGQgcHJvYmFibHkga2VlcCB0cmFjayBvZiB3
aGF0IGl0IHNlcnZlZCBhcyBhbnN3ZXIgYW5kIHdoZW4gdGhlIGFuc3dlciBpcyBpbnZhbGlkIG9y
IGEgbmV3LCBiZXR0ZXIgb25lIGV4aXN0cywgcHJvdmlkZQ0KIGEgR0FSUC48YnI+DQomZ3Q7PGJy
Pg0KJmd0O0FzIHdlbGwsIHdoZW4gUEUyIHNlbmRzIGEgbmV3ZXIgTUFDIGl0IG1heSBub3QgYmUg
YSBnb29kIHN0cmF0ZWd5IHRvIHNlcnZlIGEgR0FSUCBpZiBQRTEncyBNQUMgaGFzIGFscmVhZHkg
YmVlbiBvZmZlcmVkLiBUaGF0IGNvdWxkIGxlYWQgSU1PIHRvIGUuZy4gZ2F0ZXdheSBjaGFzaW5n
IHByb2JsZW1zLjxicj4NCiZndDs8YnI+DQomZ3Q7LS0tIHRvbnk8YnI+DQomZ3Q7PGJyPg0KJmd0
Ozxicj4NCiZndDtUaGVyZSBhcmUgYmFzaWNhbGx5IHR3byB0eXBlcyBvZiBwZW9wbGUuIFBlb3Bs
ZSB3aG8gYWNjb21wbGlzaCB0aGluZ3MsIGFuZCBwZW9wbGUgd2hvIGNsYWltIHRvIGhhdmUgYWNj
b21wbGlzaGVkIHRoaW5ncy4gVGhlIGZpcnN0IGdyb3VwIGlzIGxlc3MgY3Jvd2RlZC48YnI+DQom
Z3Q7fn5+IE1hcmsgVHdhaW48YnI+DQomZ3Q7PGJyPg0KJmd0Ozxicj4NCiZndDsmZ3Q7IC0tLS0t
T3JpZ2luYWwgTWVzc2FnZS0tLS0tPGJyPg0KJmd0OyZndDsgRnJvbTogSGVuZGVyaWNreCwgV2lt
IChXaW0pIFttYWlsdG86PGEgaHJlZj0ibWFpbHRvOndpbS5oZW5kZXJpY2t4QGFsY2F0ZWwtbHVj
ZW50LmNvbSIgdGFyZ2V0PSJfYmxhbmsiPndpbS5oZW5kZXJpY2t4QGFsY2F0ZWwtbHVjZW50LmNv
bTwvYT5dPGJyPg0KJmd0OyZndDsgU2VudDogVGh1cnNkYXksIE1hcmNoIDI2LCAyMDE1IDY6MDEg
QU08YnI+DQomZ3Q7Jmd0OyBUbzogQW50b25pIFByenlnaWVuZGE7IEVyaWsgTm9yZG1hcms7IFJh
YmFkYW4sIEpvcmdlIChKb3JnZSk8YnI+DQomZ3Q7Jmd0OyBDYzogPGEgaHJlZj0ibWFpbHRvOmJl
c3NAaWV0Zi5vcmciIHRhcmdldD0iX2JsYW5rIj5iZXNzQGlldGYub3JnPC9hPjxicj4NCiZndDsm
Z3Q7IFN1YmplY3Q6IFJlOiBbYmVzc10gQVJQIE5EIGRyYWZ0PGJyPg0KJmd0OyZndDs8YnI+DQom
Z3Q7Jmd0OyBGb3IgdGhpcyBjYXNlIHlvdSBzaG91bGQgc2VudCBhIEdBUlAgd2l0aCB0aGUgbmV3
IE1BQy9JUDxicj4NCiZndDsmZ3Q7PGJyPg0KJmd0OyZndDs8YnI+DQomZ3Q7Jmd0Ozxicj4NCiZn
dDsmZ3Q7PGJyPg0KJmd0OyZndDsgT24gMjUvMDMvMTUgMTg6NTYsICZxdW90O0FudG9uaSBQcnp5
Z2llbmRhJnF1b3Q7ICZsdDs8YSBocmVmPSJtYWlsdG86YW50b25pLnByenlnaWVuZGFAZXJpY3Nz
b24uY29tIiB0YXJnZXQ9Il9ibGFuayI+YW50b25pLnByenlnaWVuZGFAZXJpY3Nzb24uY29tPC9h
PiZndDs8YnI+DQomZ3Q7Jmd0OyB3cm90ZTo8YnI+DQomZ3Q7Jmd0Ozxicj4NCiZndDsmZ3Q7ICZn
dDsmZ3Q7ICZndDsgYilJdCBpcyB3b3J0aCBleHBsYWluaW5nIHdoYXQgaXMgc3VnZ2VzdGVkIGJl
aGF2aW9yIGlmIGVWUE48YnI+DQomZ3Q7Jmd0OyAmZ3Q7Jmd0OyAmZ3Q7IGFkdmVydGlzZXMgdGhl
IHNhbWUgSVAgd2l0aCBtdWx0aXBsZSBNQUNzIGFuZCB3aGF0IGhhcHBlbnMgd2hlbjxicj4NCiZn
dDsmZ3Q7ICZndDsmZ3Q7ICZndDsgZS5nLiB0aGUgc2VydmVkIE1BQyB2YW5pc2hlczxicj4NCiZn
dDsmZ3Q7ICZndDsmZ3Q7ICZndDs8YnI+DQomZ3Q7Jmd0OyAmZ3Q7Jmd0OyBEb2Vzbid0IHRoZSBF
VlBOIFJGQyBhbHJlYWR5IHN0YXRpbmcgdGhhdCB0aGUgcm91dGVzIHdvdWxkIGJlPGJyPg0KJmd0
OyZndDsgJmd0OyZndDsgd2l0aGRyYXduIGluIHRoYXQgY2FzZT88YnI+DQomZ3Q7Jmd0OyAmZ3Q7
PGJyPg0KJmd0OyZndDsgJmd0O1RoZSBzY2VuYXJpbyBJIGhhZCBpbiBtaW5kIHdhcyB3aGVuIGVW
UE4gUEUgcmVjZWl2ZXM8YnI+DQomZ3Q7Jmd0OyAmZ3Q7PGJyPg0KJmd0OyZndDsgJmd0O0Zyb20g
UEUyJm5ic3A7IElQMS9NMSZuYnNwOyBhbmQmbmJzcDsgbGF0ZXI8YnI+DQomZ3Q7Jmd0OyAmZ3Q7
RnJvbSBQRTMmbmJzcDsgSVAxL00yPGJyPg0KJmd0OyZndDsgJmd0Ozxicj4NCiZndDsmZ3Q7ICZn
dDt3aGlsZSBoYXZpbmcgYW5zd2VyZWQgd2l0aCBJUDEvTTEgcGVyIHByb3h5IGFscmFkeS4gQWRk
aXRpb25hbGx5LCBpbjxicj4NCiZndDsmZ3Q7ICZndDtzdWNoIHNpdHVhdGlvbiBlbmRzIHVwIHNl
ZWluZzxicj4NCiZndDsmZ3Q7ICZndDs8YnI+DQomZ3Q7Jmd0OyAmZ3Q7RnJvbSBQRTImbmJzcDsg
Jm5ic3A7SVAxLyZsdDtubyBNQUMmZ3Q7PGJyPg0KJmd0OyZndDsgJmd0Ozxicj4NCiZndDsmZ3Q7
ICZndDtTbyB0aGUgYW5zd2VyIGl0IGdhdmUgaXMgbm90IHZhbGlkIGFueW1vcmUgYWxsIG9mIGEg
c3VkZGVuLjxicj4NCiZndDsmZ3Q7ICZndDs8YnI+DQomZ3Q7Jmd0OyAmZ3Q7LS0tIHRvbnk8YnI+
DQpfX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXzxicj4NCkJF
U1MgbWFpbGluZyBsaXN0PGJyPg0KPGEgaHJlZj0ibWFpbHRvOkJFU1NAaWV0Zi5vcmciIHRhcmdl
dD0iX2JsYW5rIj5CRVNTQGlldGYub3JnPC9hPjxicj4NCjxhIGhyZWY9Imh0dHBzOi8vd3d3Lmll
dGYub3JnL21haWxtYW4vbGlzdGluZm8vYmVzcyIgdGFyZ2V0PSJfYmxhbmsiPmh0dHBzOi8vd3d3
LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vYmVzczwvYT48bzpwPjwvbzpwPjwvc3Bhbj48L3A+
DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHls
ZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90
O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6YmxhY2siPjxvOnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwv
cD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+
DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48bzpwPiZuYnNw
OzwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvYm9keT4NCjwvaHRtbD4NCg==

--_000_2E4BB27CAB87BF43B4207C0E55860F1828B205eusaamb103ericsso_--


From nobody Mon Mar 30 00:26:17 2015
Return-Path: <antoni.przygienda@ericsson.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 02D8D1A9102 for <bess@ietfa.amsl.com>; Mon, 30 Mar 2015 00:26:17 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.001
X-Spam-Level: 
X-Spam-Status: No, score=-3.001 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_12=0.6, J_CHICKENPOX_14=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fMy1JzSoGuee for <bess@ietfa.amsl.com>; Mon, 30 Mar 2015 00:26:15 -0700 (PDT)
Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 59A7C1A90F7 for <bess@ietf.org>; Mon, 30 Mar 2015 00:26:15 -0700 (PDT)
X-AuditID: c618062d-f79686d0000030a8-9b-5518a4908f2d
Received: from EUSAAHC004.ericsson.se (Unknown_Domain [147.117.188.84]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id A0.4A.12456.094A8155; Mon, 30 Mar 2015 03:19:12 +0200 (CEST)
Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC004.ericsson.se ([147.117.188.84]) with mapi id 14.03.0210.002; Mon, 30 Mar 2015 03:26:14 -0400
From: Antoni Przygienda <antoni.przygienda@ericsson.com>
To: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, Erik Nordmark <nordmark@acm.org>, "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAZ/NmAAAuXkQAAFTe+gA==
Date: Mon, 30 Mar 2015 07:26:13 +0000
Message-ID: <2E4BB27CAB87BF43B4207C0E55860F1828B24C@eusaamb103.ericsson.se>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <D138D9BF.6C4CE%jorge.rabadan@alcatel-lucent.com>
In-Reply-To: <D138D9BF.6C4CE%jorge.rabadan@alcatel-lucent.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [147.117.188.9]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFnrLLMWRmVeSWpSXmKPExsUyuXRPiO6EJRKhBg+uSVqsOD6T2eLbpSUs Fv1TdzJZzH78g82BxePyFW+P1md7WT2WLPnJFMAcxWWTkpqTWZZapG+XwJXx8PFstoJtuhW7 b+xlb2BcoNPFyMEhIWAi8eZwTBcjJ5ApJnHh3no2EFtI4CijRN9SkS5GLiB7OaPE/U3H2EES bAIWEpe/PWUGSYgILGKUWN7RBNbBLKAscXj+ebAiYQEFifaOr8wgtoiAosT85+sYIWwniX8N XUwgNouAqkTT1XawOK+At8T+A4eYILbNYpT4dmUu2FBOATuJszPngTUwAp33/dQaJohl4hK3 nsxngjhbQGLJnvPMELaoxMvH/1ghbEWJff3T2UG+ZBbQlFi/Sx+iVVFiSvdDdoi9ghInZz5h mcAoNgvJ1FkIHbOQdMxC0rGAkWUVI0dpcWpZbrqRwSZGYPwck2DT3cG456XlIUYBDkYlHt4F thKhQqyJZcWVuYcYpTlYlMR5Fz04GCIkkJ5YkpqdmlqQWhRfVJqTWnyIkYmDU6qBkeHZC+/s 6hna8tXbE69/6TviJ6ebURflvSj9xO7WeVwGu5Z82PJFii14mo76pnfzH81eGix7P+fhC75n +s/m3rr191dd+F+eRetyVqqoM/3JZdt4rOHFrof2Db+mlXJIOjm+tPOaVzB99rEo3l6VKO1g lTP1B2T7OQ4qeSc7vBTUYF88u53rvBJLcUaioRZzUXEiAP54Y5KAAgAA
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/g7rn8ZvpYPusg1qcnRTfyy5Y_lQ>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 30 Mar 2015 07:26:17 -0000

SGV5IEpvcmdlLCBJIHJlYWQgeW91ciBkcmFmdCBhZ2FpbiBpbiBhIGxhcmdlciBjb250ZXh0IHRo
YW4gc2ltcGxlICdpbmZvcm1hdGlvbmFsIGhvdyBJWFAgbmV0d29yayBydW5zIGVWUE4gQVJQIHN0
dWZmJy4gIEdlbmVyYWxseSBJIHRoaW5rIGl0IHNob3VsZCBiZSBleHRlbmRlZCBiZXlvbmQgJ3Nv
bWUgY3VzdG9tZXIncyB1c2UgY2FzZScgYW5kIHByb3ZpZGUgQkNQIG9yICdpbXBsZW1lbnRhdGlv
biByZWNvbW1lbmRhdGlvbnMnIG9yIHNvbWUgc3VjaCB0aGluZyBmb3IgdGhlIGdlbmVyaWMgaXNz
dWUgb2Ygc25vb3BpbmcgQVJQL05EL0RIQ1Avb3RoZXJzIHRvIHJlZHVjZSB0aGUgYW1vdW50IG9m
IEJVXyB0cmFmZmljIGFjcm9zcyB0aGUgUEVzLiAgQVJQL0RIQ1AvTkQgc25vb3BpbmcgaXMgYSBk
ZWxpY2F0ZSBmdW5jdGlvbmFsaXR5IHRoYXQgaXMgdmFzdGx5IHVuZGVyc3BlY2lmaWVkIGluIGVW
UE4gYmFzZSBhbmQgaXMgZXNzZW50aWFsIGZvciBhIGdvb2QgZVZQTiBpbXBsZW1lbnRhdGlvbiAo
aS5lLiByb2J1c3QsIGludGVyb3BlcmFibGUgJiBzY2FsYWJsZSkgSU1PLiANCg0KVGhhbmtzIGZv
ciBzaGFyaW5nIGFsbCB0aGlzIGV4cGVyaWVuY2UgQlRXLCBpbnNpZ2h0ZnVsIHJlYWQgdGhhdCBl
bmxpZ2h0ZW5lZCBidW5jaCBvZiBjb3JuZXIgY2FzZXMgSSBkaWRuJ3QgdGhpbmsgdGhyb3VnaC4g
DQoNCkNvbW1lbnRzIGlubGluZSANCg0KPiA+PiBhKUl0IGlzIHdvcnRoIGV4cGxhaW5pbmcgd2hh
dCBmbGF2b3Igb2YgQVJQIHByb3h5IGVWUE4gaW1wbGVtZW50cy4NCj4gPj4g4oCYcHJveHkgQVJQ
4oCZIEkgZm91bmQgb3V0IGhhcyBkaWZmZXJlbnQgZmxhdm9ycyBpbmNsdWRpbmcgZS5nLiB0aGUN
Cj4gPj4gcm91dGVyIHJlc3BvbmRpbmcgd2l0aCBpdHMgb3duIE1BQyB0byByZXF1ZXN0cyByZXBy
ZXNlbnRpbmcgcmVtb3RlDQo+ID4+IGhvc3RzLiBDdXN0b21lcnMgJiBvdGhlciBmb2xrcyBhcmUg
ZWFzaWx5IGNvbmZ1c2VkIGJ5IHRoZSBvdmVybG9hZGVkDQo+ID4+IOKAmHByb3h5IEFSUOKAmSB0
ZXJtIGluIGVWUE4gY29udGV4dC4NCj4gPj4NCj4gPlllcywgdGhhdCBpcyBhIHBhcnQgdGhhdCBp
cyB1bmRlcnNwZWNpZmllZCBmb3IgRVZQTi4gSSBhc3N1bWUgdGhhdCB0aGUNCj4gPnJlc3BvbnNl
IHdvdWxkIGluY2x1ZGUgdGhlIE1BQyBhZGRyZXNzIG9mIHRoZSBDRSBpbnN0ZWFkIG9mIHRoZQ0K
PiA+cm91dGVyJ3Mgb3duIE1BQyBhZGRyZXNzLg0KPiANCj4gW0pPUkdFXSBGcm9tIGRyYWZ0LXNu
ci1iZXNzLWV2cG4tcHJveHktYXJwLW5kLTAwOg0KPiANCj4gNC4yIFJlcGx5IHN1Yi1mdW5jdGlv
bg0KPiANCj4gLi4uDQo+IA0KPiAgICBhKSBXaGVuIHJlcGx5aW5nIHRvIEFSUCBSZXF1ZXN0IG9y
IE5TIG1lc3NhZ2VzLCB0aGUgUEUgU0hPVUxEIHVzZSB0aGUNCj4gICAgICAgUHJveHktQVJQL05E
IGVudHJ5IE1BQyBhZGRyZXNzIGFzIE1BQyBTQS4gVGhpcyBpcyByZWNvbW1lbmRlZCBzbw0KPiAg
ICAgICB0aGF0IHRoZSByZXNvbHZlZCBNQUMgY2FuIGJlIGxlYXJudCBpbiB0aGUgTUFDIEZJQiBv
ZiBwb3RlbnRpYWwNCj4gICAgICAgTGF5ZXItMiBzd2l0Y2hlcyBzZWF0aW5nIGJldHdlZW4gdGhl
IFBFIGFuZCB0aGUgQ0UgcmVxdWVzdGluZyB0aGUNCj4gICAgICAgQWRkcmVzcyBSZXNvbHV0aW9u
Lg0KDQogW1Rvbnkgc2FpejpdICBhZ3JlZWQuIEkgY2Fubm90IGltYWdpbmUgd2h5IGl0J3MgTk9U
IGEgTVVTVCAoZXhjZXB0IGRlZmF1bHQgR1cgY2FzZXMgd2hlcmUgaXQgbWFrZXMgc2Vuc2UgZS5n
LiBpbiBjYXNlIG9mIGFsaWFzZWQgY2FzZSB0byByZXNvbHZlIEdXIElQIHJlcXVlc3QpLiANCg0K
U21hbGwgb2JzZXJ2YXRpb24gZm9yIDQuMmIpICBJZiB3ZSB3YW50IHRvIGJlIGRvdWJsZS1wZXJm
ZWN0IHdlIHNob3VsZCBhY3R1YWxseSAgbm90IGV2ZW4gcmVzcG9uZCBpZiB3ZSBoYXZlIDIgQUNz
IG9uIHRoZSBzYW1lIEVTSSBvbiB0aGUgc2FtZSBQRSA7LSkgIA0KDQpTbWFsbCBvYnNlcnZhdGlv
biBmb3IgNC4zOiAgZW5hYmxpbmcgdGhlIHNlbmQtcmVmcmVzaCBvcHRpb24gaXMgYSBkYW5nZXJv
dXMgdGhpbmcuIEl0J3Mgbm90IG9ubHkgdGhleSBtYXkga2VlcCB1cCBhY3RpdmUgZW50cmllcyBp
biB0aGUgZmFzdC1wYXRoIGZvcmV2ZXIgKHNpbmNlIHRoZXJlIGlzIGFsd2F5cyBzb21lIHRyYWZm
aWMgZ2VuZXJhdGVkIGJ5IHRoZSAncHJvYmVzJyksIGl0J3MgYWxzbyB0aGF0IHRoZSBJUC9NQUMg
YmluZGluZyAna2VwdCB1cCcgZXZlbiBpZiB0aGVyZSBpcyBubyB0cmFmZmljIGlzIHNpdHRpbmcg
aW4gYWxsIFBFcyBhcyBSVC0yLiAgVGhlIHNlY3Rpb24gdGFsa3MgYWJvdXQgYWR2YW50YWdlcywg
bWF5YmUgaXQgZGVzZXJ2ZXMgYSBzZW50ZW5jZSB0byBwb2ludCBvdXQgdGhlIGRhbmdlcnMgb2Yg
c3VjaCBiZWhhdmlvci4gDQoNCj4gPj4gYilJdCBpcyB3b3J0aCBleHBsYWluaW5nIHdoYXQgaXMg
c3VnZ2VzdGVkIGJlaGF2aW9yIGlmIGVWUE4NCj4gPj4gYWR2ZXJ0aXNlcyB0aGUgc2FtZSBJUCB3
aXRoIG11bHRpcGxlIE1BQ3MgYW5kIHdoYXQgaGFwcGVucyB3aGVuIGUuZy4NCj4gPj4gdGhlIHNl
cnZlZCBNQUMgdmFuaXNoZXMNCj4gPj4NCj4gPkRvZXNuJ3QgdGhlIEVWUE4gUkZDIGFscmVhZHkg
c3RhdGluZyB0aGF0IHRoZSByb3V0ZXMgd291bGQgYmUgd2l0aGRyYXduDQo+ID5pbiB0aGF0IGNh
c2U/DQo+IA0KPiBbSk9SR0VdIEJhc2VkIG9uIG91ciBjdXJyZW50IHZlcnNpb24sIHRoZSBJUCBp
cyB1bmlxdWUgaW4gYSBQReKAmXMgcHJveHktQVJQIHRhYmxlLA0KPiBzbyBpZiBhIFBFIHJlY2Vp
dmVzIDIgUlQtMnMgZm9yIHRoZSBzYW1lIElQIGRpZmZlcmVudCBNQUMsIG9ubHkgb25lDQo+IElQ
LT5NQUMgYmluZGluZyB3aWxsIGJlIHBpY2tlZCB1cC4NCg0KW1Rvbnkgc2FpejpdICAgYXMgSSB3
cm90ZSBpbiBvdGhlciBlbWFpbCwgdGhlIHByb2JsZW0gaXMgaWYgdGhlIGZpcnN0IHJvdXRlIGdv
ZXMsIHRoZSBzZWNvbmQgc2hvdWxkIGJlIGhvbm9yZWQgSU1PLiBJdCBzZWVtcyB0byBtZSB3ZSBo
YXZlIHRoaXMgdmVyeSBjYXNlIEJUVyB3aXRoIHRoZSBhbGlhc2VkIGRlZmF1bHQgZ2F0ZXdheXM/
DQoNCj4gPj4NCj4gPj4gYylJdCBpcyB3b3J0aCBleHBsYWluaW5nIHdoYXQgdGhlIHN1Z2dlc3Rl
ZCBiZWhhdmlvciBzaG91bGQgYmUgd2hlbg0KPiA+PiDigJhsb2NhbC1iaXTigJkgTUFDcyBhcmUg
YmVpbmcgYWR2ZXJ0aXNlZCBmcm9tIHJlbW90ZXMgKGFuZCB3aGVuIHRob3NlDQo+ID4+IGNvbGxp
ZGUpDQo+ID4+DQo+ID5Eb2VzIEwyVlBOIGhhbmRsZSB0aGF0IGluIGFueSBpbnRlcmVzdGluZyB3
YXk/IEkgZG9uJ3QgdGhpbmsgRVZQTg0KPiA+aW50cm9kdWNlcyBhbnkgbmV3IGZhaWx1cmUgbW9k
ZXMgZm9yIGR1cGxpY2F0ZSBNQUMgYWRkcmVzc2VzLg0KPiANCj4gW0pPUkdFXSBpbmRlZWQsIHRo
aXMgZHJhZnQgZG9lcyBub3QgaW50cm9kdWNlIGFueSBuZXcgd2F5IHRvIGhhbmRsZSBNQUMNCj4g
YWRkcmVzc2VzIGluIHRoZSBNQUMtVlJGcy4gRVZQTiBoYXMgYSBNQUMgZHVwbGljYXRpb24gbWVj
aGFuaXNtLCB0aGVyZSBpcw0KPiBub3RoaW5nIGVsc2UgYWZhaWsuDQoNCiBbVG9ueSBzYWl6Ol0g
IGFncmVlZC4gDQoNCj4gPj4NCj4gPj4gZCllVlBOIGRyYWZ0IGRvZXMgbm90IGV4cGxhaW4gYW55
dGhpbmcgYWJvdXQgcG9zc2liaWxpdHkgdG8gc25vb3AgQA0KPiA+PiBESENQDQo+IA0KPiBbSk9S
R0VdIGRvIHlvdSBtZWFuIGluIHRoZSBwcm94eS1hcnAvbmQgZHJhZnQgb3IgaW4gdGhlIGJhc2Ug
c3BlYz8NCj4gSW4gdGhlIHByb3h5LWFycC9uZCBvbmx5IEFSUC9ORCBtZXNzYWdlcyBhcmUgdXNl
ZC4gREhDUCBpcyBub3QgYWx3YXlzIHRoZXJlLg0KPiBOb3QgdGhlcmUgaW4gdGhlIElYUCB1c2Ut
Y2FzZSBhbnl3YXkuDQoNCiBbVG9ueSBzYWl6Ol0gIGFncmVlZCwgIEkganVzdCByZWFkIHRoZSBk
cmFmdCBpbiB3aWRlciBzY29wZSB0aGFuIGp1c3QgSVhQIGFuZCB0aGF0J3Mgd2h5IEknbSBhc2tp
bmcuICANCg0KT25lIGNvdWxkIGV2ZW4gdGFsayBhYm91dCBpbml0aWFsbHkgc25vb3BpbmcgIElQ
IGZyYW1lcyBkaXJlY3RseSBpbiBjYXNlIGVWUE4gImF0dGFjaGVzIiB0byBhbiBhbHJlYWR5IHJ1
bm5pbmcgYnJpZGdlIGFuZCBzbm9vcHMgYWxyZWFkeSBhY3RpdmUgSVAgdHJhZmZpYyB0byBsZWFy
biBJUC9NQUMgYmluZGluZ3MgZnJvbSBicmlkZ2UgcG9ydHMuIFByb2JhYmx5IGh5cG90aGV0aWNh
bCBvbmx5ICA7LSkgDQoNCi0tIHRvbnkgDQoNCg==


From nobody Mon Mar 30 07:06:11 2015
Return-Path: <stbryant@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6EE7A1ACEF4; Mon, 30 Mar 2015 07:06:08 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -11.81
X-Spam-Level: 
X-Spam-Status: No, score=-11.81 tagged_above=-999 required=5 tests=[BAYES_50=0.8, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZUiI6mAOSKML; Mon, 30 Mar 2015 07:06:05 -0700 (PDT)
Received: from aer-iport-3.cisco.com (aer-iport-3.cisco.com [173.38.203.53]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A66F31ACE22; Mon, 30 Mar 2015 07:06:04 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=13530; q=dns/txt; s=iport; t=1427724365; x=1428933965; h=message-id:date:from:reply-to:mime-version:to:cc:subject: references:in-reply-to; bh=yOfC3SHWQtBGB2RiegDlRX4xKxuZRYj+LELXlacNR7A=; b=hqzC88ypZY/nPUoeFDvIXl9PlGVVF3dcWE5nB34kCjPOGDWJOO038hGR TQwOjoHy022TEf9mszaxLdza/ODr9w4SZZQA1TqkXgsNoUmXlGKnhOics W6bHE120Q3GRTzKD5oG8OZcG5UO9IzHelm8/Ew3UhLMyS/xCnZvf53CbU E=;
X-IronPort-AV: E=Sophos;i="5.11,494,1422921600";  d="scan'208,217";a="404547874"
Received: from aer-iport-nat.cisco.com (HELO aer-core-1.cisco.com) ([173.38.203.22]) by aer-iport-3.cisco.com with ESMTP; 30 Mar 2015 14:06:02 +0000
Received: from [64.103.108.139] (dhcp-bdlk10-data-vlan301-64-103-108-139.cisco.com [64.103.108.139]) by aer-core-1.cisco.com (8.14.5/8.14.5) with ESMTP id t2UE62AG022673; Mon, 30 Mar 2015 14:06:02 GMT
Message-ID: <55195849.7090003@cisco.com>
Date: Mon, 30 Mar 2015 15:06:01 +0100
From: Stewart Bryant <stbryant@cisco.com>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: Jiangyuanlong <jiangyuanlong@huawei.com>, "draft-ietf-l2vpn-vpls-pe-etree@tools.ietf.org" <draft-ietf-l2vpn-vpls-pe-etree@tools.ietf.org>,  "pals@ietf.org" <pals@ietf.org>
References: <54EB4B87.2080202@cisco.com> <3B0A1BED22CAD649A1B3E97BE5DDD68B5A890926@szxema506-mbs.china.huawei.com>
In-Reply-To: <3B0A1BED22CAD649A1B3E97BE5DDD68B5A890926@szxema506-mbs.china.huawei.com>
Content-Type: multipart/alternative; boundary="------------040409010906070900010107"
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/yWtMkXy2rBikhXx7dJX1iraqXaU>
Cc: "l2vpn-chairs@tools.ietf.org" <l2vpn-chairs@tools.ietf.org>, "bess-chairs@tools.ietf.org" <bess-chairs@tools.ietf.org>, "pals-chairs@tools.ietf.org" <pals-chairs@tools.ietf.org>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] [Pals] Shepherd review of draft-ietf-l2vpn-vpls-pe-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: stbryant@cisco.com
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 30 Mar 2015 14:06:08 -0000

This is a multi-part message in MIME format.
--------------040409010906070900010107
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit

Yuanlong

The text has the following nits mostly this is just minor editing but
does need to be fixed.

RFC 6136 can be made an informational ref

=====================


tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt:
tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(20): Line is too long: the 
offending characters are '.'
tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(227): Line is too long: the 
offending characters are '.'
tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(415): Line is too long: the 
offending characters are ','
tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(579): Line is too long: the 
offending characters are 'e'
tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(582): Line is too long: the 
offending characters are 'N,'
tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(632): Line is too long: the 
offending characters are 'n'
tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(635): Line is too long: the 
offending characters are 'e'
tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(917): Line is too long: the 
offending characters are 'n'
tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(924): Line is too long: the 
offending characters are '.'
tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(1138): Line has weird spacing: 
'...) where  any t...'


   Checking boilerplate required by RFC 5378 and the IETF Trust (see
   http://trustee.ietf.org/license-info):
----------------------------------------------------------------------------

      No issues found here.

   Checking nits according to 
http://www.ietf.org/id-info/1id-guidelines.txt:
----------------------------------------------------------------------------

      No issues found here.

   Checking nits according to http://www.ietf.org/id-info/checklist :
----------------------------------------------------------------------------

   ** There are 9 instances of too long lines in the document, the 
longest one
      being 2 characters in excess of 72.


   Miscellaneous warnings:
----------------------------------------------------------------------------

   == Using lowercase 'not' together with uppercase 'MUST', 'SHALL', 
'SHOULD',
      or 'RECOMMENDED' is not an accepted usage according to RFC 2119.  
Please
      use uppercase 'NOT' together with RFC 2119 keywords (if that is 
what you
      mean).

      Found 'MUST not' in this paragraph:

      The Ethernet-Tree (E-Tree) service is defined in Metro Ethernet
      Forum (MEF) Technical Specification MEF 6.1 as a Rooted-Multipoint
      Ethernet Virtual Connection (EVC) service. It is a multipoint Ethernet
      service with special restrictions: the Ethernet frames from a root 
may be
      received by any other root or leaf, and the frames from a leaf may be
      received by any root, but MUST not be received by a leaf. Further, an
      E-Tree service may include multiple roots and multiple leaves. 
Although
      Virtual Private Multicast Service (VPMS) [VPMS] or Point-to-Multipoint
      (P2MP) multicast is a somewhat simplified version of this service, in
      fact, there is no exact corresponding terminology in IETF yet.

   -- The document date (February 26, 2015) is 32 days in the past.  Is this
      intentional?


   Checking references for intended status: Proposed Standard
----------------------------------------------------------------------------

      (See RFCs 3967 and 4897 for information about using normative 
references
      to lower-maturity documents in RFCs)

   == Unused Reference: 'MEF4' is defined on line 872, but no explicit
      reference was found in the text
      '[MEF4]    Metro Ethernet Forum, Metro Ethernet Network 
Architecture...'

   ** Downref: Normative reference to an Informational RFC: RFC 6136


      Summary: 2 errors (**), 0 flaws (~~), 2 warnings (==), 1 comment (--).


===============

Looking at this text



614	   1) if the root and leaf VLAN ID in the message match the local root
615	      and leaf VLAN ID, then continue to 3);

617	   2) else {

619	          if the bit V is cleared, then {

621	                if the PE is capable of VLAN mapping, then it MUST set
622	                VLAN-Mapping-Mode to TRUE;

624	                else {

626	                     A label release message with the error code "E-Tree
627	                     VLAN mapping not supported" is sent to the peer PE
628	                     and exit the process;

630	                     }

632	          }

634	          if the bit V is set, and the PE is capable of VLAN mapping,
635	          then the PE with the minimum IP address MUST set VLAN-Mapping-
636	          Mode to TRUE;

638	      }

640	   3) If the P bit is set, then:


It might be clearer to apply De Morgan:

Which I think makes it

1) if the root not equal local root or leaf VLAN ID not equal to leaf VLAN ID
    in the message match then


	          if the bit V is cleared, then {

	                if the PE is capable of VLAN mapping, then it MUST set
	                VLAN-Mapping-Mode to TRUE;

	                else {

	                     A label release message with the error code "E-Tree
	                     VLAN mapping not supported" is sent to the peer PE
	                     and exit the process;

	                     }

	          }

	          if the bit V is set, and the PE is capable of VLAN mapping,
	          then the PE with the minimum IP address MUST set VLAN-Mapping-
	          Mode to TRUE;

	      }

2) If the P bit is set, then:




--------------040409010906070900010107
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<html>
  <head>
    <meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix"><span
style="font-size:10.5pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D"
        lang="EN-US"><font color="#000000"><big>Yuanlong<br>
            <br>
            The text has the following nits mostly this is just minor
            editing but<br>
            does need to be fixed.<br>
          </big></font></span><br>
      RFC 6136 can be made an informational ref<br>
      <br>
      =====================<br>
      <br>
      <br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt:<br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(20): Line is too long:
      the offending characters are '.'<br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(227): Line is too long:
      the offending characters are '.'<br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(415): Line is too long:
      the offending characters are ','<br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(579): Line is too long:
      the offending characters are 'e'<br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(582): Line is too long:
      the offending characters are 'N,'<br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(632): Line is too long:
      the offending characters are 'n'<br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(635): Line is too long:
      the offending characters are 'e'<br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(917): Line is too long:
      the offending characters are 'n'<br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(924): Line is too long:
      the offending characters are '.'<br>
      tmp/draft-ietf-l2vpn-vpls-pe-etree-05.txt(1138): Line has weird
      spacing: '...) where  any t...'<br>
      <br>
      <br>
        Checking boilerplate required by RFC 5378 and the IETF Trust
      (see<br>
        <a class="moz-txt-link-freetext" href="http://trustee.ietf.org/license-info">http://trustee.ietf.org/license-info</a>):<br>
       
----------------------------------------------------------------------------<br>
      <br>
           No issues found here.<br>
      <br>
        Checking nits according to
      <a class="moz-txt-link-freetext" href="http://www.ietf.org/id-info/1id-guidelines.txt">http://www.ietf.org/id-info/1id-guidelines.txt</a>:<br>
       
----------------------------------------------------------------------------<br>
      <br>
           No issues found here.<br>
      <br>
        Checking nits according to <a class="moz-txt-link-freetext" href="http://www.ietf.org/id-info/checklist">http://www.ietf.org/id-info/checklist</a>
      :<br>
       
----------------------------------------------------------------------------<br>
      <br>
        ** There are 9 instances of too long lines in the document, the
      longest one<br>
           being 2 characters in excess of 72.<br>
      <br>
      <br>
        Miscellaneous warnings:<br>
       
----------------------------------------------------------------------------<br>
      <br>
        == Using lowercase 'not' together with uppercase 'MUST',
      'SHALL', 'SHOULD',<br>
           or 'RECOMMENDED' is not an accepted usage according to RFC
      2119.  Please<br>
           use uppercase 'NOT' together with RFC 2119 keywords (if that
      is what you<br>
           mean).<br>
           <br>
           Found 'MUST not' in this paragraph:<br>
           <br>
           The Ethernet-Tree (E-Tree) service is defined in Metro
      Ethernet<br>
           Forum (MEF) Technical Specification MEF 6.1 as a
      Rooted-Multipoint<br>
           Ethernet Virtual Connection (EVC) service. It is a multipoint
      Ethernet<br>
           service with special restrictions: the Ethernet frames from a
      root may be<br>
           received by any other root or leaf, and the frames from a
      leaf may be<br>
           received by any root, but MUST not be received by a leaf.
      Further, an<br>
           E-Tree service may include multiple roots and multiple
      leaves. Although<br>
           Virtual Private Multicast Service (VPMS) [VPMS] or
      Point-to-Multipoint<br>
           (P2MP) multicast is a somewhat simplified version of this
      service, in<br>
           fact, there is no exact corresponding terminology in IETF
      yet.<br>
      <br>
        -- The document date (February 26, 2015) is 32 days in the
      past.  Is this<br>
           intentional?<br>
      <br>
      <br>
        Checking references for intended status: Proposed Standard<br>
       
----------------------------------------------------------------------------<br>
      <br>
           (See RFCs 3967 and 4897 for information about using normative
      references<br>
           to lower-maturity documents in RFCs)<br>
      <br>
        == Unused Reference: 'MEF4' is defined on line 872, but no
      explicit<br>
           reference was found in the text<br>
           '[MEF4]    Metro Ethernet Forum, Metro Ethernet Network
      Architecture...'<br>
      <br>
        ** Downref: Normative reference to an Informational RFC: RFC
      6136<br>
      <br>
      <br>
           Summary: 2 errors (**), 0 flaws (~~), 2 warnings (==), 1
      comment (--).<br>
      <br>
      <br>
      ===============<br>
      <br>
      Looking at this text<br>
       <br>
      <br>
      <br>
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <pre>614	   1) if the root and leaf VLAN ID in the message match the local root
615	      and leaf VLAN ID, then continue to 3);

617	   2) else {

619	          if the bit V is cleared, then {

621	                if the PE is capable of VLAN mapping, then it MUST set
622	                VLAN-Mapping-Mode to TRUE;

624	                else {

626	                     A label release message with the error code "E-Tree
627	                     VLAN mapping not supported" is sent to the peer PE
628	                     and exit the process;

630	                     }

632	          }

634	          if the bit V is set, and the PE is capable of VLAN mapping,
635	          then the PE with the minimum IP address MUST set VLAN-Mapping-
636	          Mode to TRUE;

638	      }

640	   3) If the P bit is set, then:</pre>
      <br>
      It might be clearer to apply De Morgan: <br>
      <br>
      Which I think makes it<br>
      <br>
      <pre>1) if the root not equal local root or leaf VLAN ID not equal to leaf VLAN ID
   in the message match then


	          if the bit V is cleared, then {

	                if the PE is capable of VLAN mapping, then it MUST set
	                VLAN-Mapping-Mode to TRUE;

	                else {

	                     A label release message with the error code "E-Tree
	                     VLAN mapping not supported" is sent to the peer PE
	                     and exit the process;

	                     }

	          }

	          if the bit V is set, and the PE is capable of VLAN mapping,
	          then the PE with the minimum IP address MUST set VLAN-Mapping-
	          Mode to TRUE;

	      }

2) If the P bit is set, then:</pre>
      <br>
      <br>
    </div>
  </body>
</html>

--------------040409010906070900010107--


From nobody Mon Mar 30 21:56:54 2015
Return-Path: <Alexander.Vainshtein@ecitele.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BC2621B2A8F; Mon, 30 Mar 2015 21:56:50 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.901
X-Spam-Level: 
X-Spam-Status: No, score=-1.901 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EMnQAwvdz4cg; Mon, 30 Mar 2015 21:56:47 -0700 (PDT)
Received: from emea01-am1-obe.outbound.protection.outlook.com (mail-am1on0725.outbound.protection.outlook.com [IPv6:2a01:111:f400:fe00::725]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9F2951A924E; Mon, 30 Mar 2015 21:56:46 -0700 (PDT)
Received: from DB3PR03MB0780.eurprd03.prod.outlook.com (25.161.55.12) by DB3PR03MB0777.eurprd03.prod.outlook.com (25.161.54.27) with Microsoft SMTP Server (TLS) id 15.1.112.19; Tue, 31 Mar 2015 04:55:08 +0000
Received: from DB3PR03MB0780.eurprd03.prod.outlook.com ([25.161.55.12]) by DB3PR03MB0780.eurprd03.prod.outlook.com ([25.161.55.12]) with mapi id 15.01.0112.000; Tue, 31 Mar 2015 04:55:08 +0000
From: Alexander Vainshtein <Alexander.Vainshtein@ecitele.com>
To: Jiangyuanlong <jiangyuanlong@huawei.com>, "draft-ietf-l2vpn-vpls-pe-etree@tools.ietf.org" <draft-ietf-l2vpn-vpls-pe-etree@tools.ietf.org>, "pals@ietf.org" <pals@ietf.org>, "stbryant@cisco.com" <stbryant@cisco.com>
Thread-Topic: [Pals] Shepherd review of draft-ietf-l2vpn-vpls-pe-etree
Thread-Index: AQHQavK05uf6ETR9uE6Vq98UWEmj5Z02Bhvu
Date: Tue, 31 Mar 2015 04:55:08 +0000
Message-ID: <1427777706620.6234@ecitele.com>
References: <54EB4B87.2080202@cisco.com> <3B0A1BED22CAD649A1B3E97BE5DDD68B5A890926@szxema506-mbs.china.huawei.com>, <55195849.7090003@cisco.com>
In-Reply-To: <55195849.7090003@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [5.153.9.202]
authentication-results: tools.ietf.org; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:DB3PR03MB0777;
x-microsoft-antispam-prvs: <DB3PR03MB0777BFD0EF5AFE9461DCACFA9DF40@DB3PR03MB0777.eurprd03.prod.outlook.com>
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(377454003)(2501003)(19625215002)(106116001)(230783001)(16236675004)(54356999)(40100003)(117636001)(77156002)(2656002)(2201001)(19580405001)(2900100001)(76176999)(36756003)(16297215004)(62966003)(19580395003)(2950100001)(102836002)(50986999)(15975445007)(66066001)(86362001)(92566002)(71446004)(122556002)(19617315012)(46102003)(19627405001)(87936001); DIR:OUT; SFP:1102; SCL:1; SRVR:DB3PR03MB0777; H:DB3PR03MB0780.eurprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5005006)(5002010); SRVR:DB3PR03MB0777; BCL:0; PCL:0; RULEID:;  SRVR:DB3PR03MB0777; 
x-forefront-prvs: 0532BF6DC2
Content-Type: multipart/alternative; boundary="_000_14277777066206234ecitelecom_"
MIME-Version: 1.0
X-OriginatorOrg: ecitele.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 31 Mar 2015 04:55:08.1384 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 2c514a61-08de-4519-b4c0-921fef62c42a
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DB3PR03MB0777
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/lAeXY5iCfE8cld_ZbJlokcrl4Sc>
Cc: "l2vpn-chairs@tools.ietf.org" <l2vpn-chairs@tools.ietf.org>, "bess-chairs@tools.ietf.org" <bess-chairs@tools.ietf.org>, "pals-chairs@tools.ietf.org" <pals-chairs@tools.ietf.org>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] [Pals] Shepherd review of draft-ietf-l2vpn-vpls-pe-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Mar 2015 04:56:50 -0000

--_000_14277777066206234ecitelecom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_14277777066206234ecitelecom_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_14277777066206234ecitelecom_--


From nobody Mon Mar 30 23:21:28 2015
Return-Path: <satyamoh@cisco.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6E1731B2AC6 for <bess@ietfa.amsl.com>; Mon, 30 Mar 2015 23:21:27 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -9.562
X-Spam-Level: 
X-Spam-Status: No, score=-9.562 tagged_above=-999 required=5 tests=[BAYES_20=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, J_CHICKENPOX_56=0.6, MIME_CHARSET_FARAWAY=2.45, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WdPFOsQRRsis for <bess@ietfa.amsl.com>; Mon, 30 Mar 2015 23:21:21 -0700 (PDT)
Received: from rcdn-iport-3.cisco.com (rcdn-iport-3.cisco.com [173.37.86.74]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 40A6E1B2AC1 for <bess@ietf.org>; Mon, 30 Mar 2015 23:21:21 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=54818; q=dns/txt; s=iport; t=1427782883; x=1428992483; h=from:to:cc:subject:date:message-id:content-id: content-transfer-encoding:mime-version; bh=sTrFEGc78KLzlqTDEY5r/inxVFMbTa+hf5cIUCqkc6U=; b=WLFLgEfRrG0nEqA4MbTQNloCRV42qnD1+xk9hbYtPGcvpEv1KCFet/T2 +bocsVjFpqS2JlgzP9ifwrj5ivmE1ZCuDR2/GfvA9JZKDW0BufH4/5Up1 pfIN9wp0H1oYvT5/ilIf0mfq04xDGinrdLV7o/JMSvv58WzAvsxNN3G5j U=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A0ARBQAuPBpV/5tdJa1ZA4MGUlwEgw/CRQqFcwIcgSBMAQEBAQEBfYQUAQEBBAEBARcBDA0yCAsMAgQBBhMDAQEBAQQjBQQZDAsUCQoEDgUUB4gUDZYPnG4GgUyXVAEBAQEBAQEBAQEBAQEBAQEBAQEBARcEgReJD3+EDQgBEQEOHAsLCwUNC4JRgUsFhhqKQoNxhgOBHDqCeIJahB+FJoNIIoIBARyBUG8BgQMHFyJ/AQEB
X-IronPort-AV: E=Sophos;i="5.11,499,1422921600"; d="scan'208";a="407966624"
Received: from rcdn-core-4.cisco.com ([173.37.93.155]) by rcdn-iport-3.cisco.com with ESMTP; 31 Mar 2015 06:21:20 +0000
Received: from xhc-aln-x11.cisco.com (xhc-aln-x11.cisco.com [173.36.12.85]) by rcdn-core-4.cisco.com (8.14.5/8.14.5) with ESMTP id t2V6LIUt007568 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for <bess@ietf.org>; Tue, 31 Mar 2015 06:21:18 GMT
Received: from xmb-aln-x07.cisco.com ([169.254.2.91]) by xhc-aln-x11.cisco.com ([173.36.12.85]) with mapi id 14.03.0195.001; Tue, 31 Mar 2015 01:21:18 -0500
From: "Satya Mohanty (satyamoh)" <satyamoh@cisco.com>
To: "bess@ietf.org" <bess@ietf.org>
Thread-Topic: Request for WG adoption of draft-mohanty-bess-evpn-df-election
Thread-Index: AQHQa3rlLEUMkzGn3E2wY+x+uaBVDA==
Date: Tue, 31 Mar 2015 06:21:17 +0000
Message-ID: <D13F8823.2CE8D%satyamoh@cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.65.74.238]
Content-Type: text/plain; charset="euc-kr"
Content-ID: <6D90797E600A9548BEC72A86F39B2D77@emea.cisco.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/UVWtxkYkk9LTP1-vXbLj3uy3V8I>
Cc: "Jakob Heitz \(jheitz\)" <jheitz@cisco.com>, "Tamas Mondal \(tmondal\)" <tmondal@cisco.com>
Subject: [bess] Request for WG adoption of draft-mohanty-bess-evpn-df-election
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Mar 2015 06:21:27 -0000
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From nobody Mon Mar 30 23:46:28 2015
Return-Path: <jiangyuanlong@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 507E61B2AE4; Mon, 30 Mar 2015 23:46:25 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.21
X-Spam-Level: 
X-Spam-Status: No, score=-4.21 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id DgCizJDcoks6; Mon, 30 Mar 2015 23:46:21 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A306A1B2AE3; Mon, 30 Mar 2015 23:46:19 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml402-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BUI09420; Tue, 31 Mar 2015 06:46:18 +0000 (GMT)
Received: from SZXEMA414-HUB.china.huawei.com (10.82.72.73) by lhreml402-hub.china.huawei.com (10.201.5.241) with Microsoft SMTP Server (TLS) id 14.3.158.1; Tue, 31 Mar 2015 07:46:14 +0100
Received: from SZXEMA506-MBS.china.huawei.com ([169.254.4.68]) by SZXEMA414-HUB.china.huawei.com ([10.82.72.73]) with mapi id 14.03.0158.001; Tue, 31 Mar 2015 14:46:05 +0800
From: Jiangyuanlong <jiangyuanlong@huawei.com>
To: Alexander Vainshtein <Alexander.Vainshtein@ecitele.com>, "draft-ietf-l2vpn-vpls-pe-etree@tools.ietf.org" <draft-ietf-l2vpn-vpls-pe-etree@tools.ietf.org>, "pals@ietf.org" <pals@ietf.org>, "stbryant@cisco.com" <stbryant@cisco.com>
Thread-Topic: [Pals] Shepherd review of draft-ietf-l2vpn-vpls-pe-etree
Thread-Index: AQHQT4AUGI+hr/i+r0Gj9wcHU/COq502BhvugABRP5A=
Date: Tue, 31 Mar 2015 06:46:05 +0000
Message-ID: <3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C55DA@szxema506-mbs.china.huawei.com>
References: <54EB4B87.2080202@cisco.com> <3B0A1BED22CAD649A1B3E97BE5DDD68B5A890926@szxema506-mbs.china.huawei.com>, <55195849.7090003@cisco.com> <1427777706620.6234@ecitele.com>
In-Reply-To: <1427777706620.6234@ecitele.com>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.66.76.118]
Content-Type: multipart/alternative; boundary="_000_3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C55DAszxema506mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/gDlBJtw3RFXY1UUDbJbfX9X3oFk>
Cc: "l2vpn-chairs@tools.ietf.org" <l2vpn-chairs@tools.ietf.org>, "bess-chairs@tools.ietf.org" <bess-chairs@tools.ietf.org>, "pals-chairs@tools.ietf.org" <pals-chairs@tools.ietf.org>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] [Pals] Shepherd review of draft-ietf-l2vpn-vpls-pe-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Mar 2015 06:46:25 -0000

--_000_3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C55DAszxema506mbschi_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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=

--_000_3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C55DAszxema506mbschi_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C55DAszxema506mbschi_--


From nobody Tue Mar 31 00:01:46 2015
Return-Path: <Alexander.Vainshtein@ecitele.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EE5D71B2AF6; Tue, 31 Mar 2015 00:01:44 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.901
X-Spam-Level: 
X-Spam-Status: No, score=-1.901 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hom5NGYOYwPt; Tue, 31 Mar 2015 00:01:41 -0700 (PDT)
Received: from emea01-am1-obe.outbound.protection.outlook.com (mail-am1on0798.outbound.protection.outlook.com [IPv6:2a01:111:f400:fe00::798]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 70B731B2AF3; Tue, 31 Mar 2015 00:01:40 -0700 (PDT)
Received: from DB3PR03MB0780.eurprd03.prod.outlook.com (25.161.55.12) by DB3PR03MB0780.eurprd03.prod.outlook.com (25.161.55.12) with Microsoft SMTP Server (TLS) id 15.1.112.16; Tue, 31 Mar 2015 07:01:17 +0000
Received: from DB3PR03MB0780.eurprd03.prod.outlook.com ([25.161.55.12]) by DB3PR03MB0780.eurprd03.prod.outlook.com ([25.161.55.12]) with mapi id 15.01.0112.000; Tue, 31 Mar 2015 07:01:17 +0000
From: Alexander Vainshtein <Alexander.Vainshtein@ecitele.com>
To: Jiangyuanlong <jiangyuanlong@huawei.com>
Thread-Topic: [Pals] Shepherd review of draft-ietf-l2vpn-vpls-pe-etree
Thread-Index: AQHQavK05uf6ETR9uE6Vq98UWEmj5Z02BhvugAAgt4CAAANFwA==
Date: Tue, 31 Mar 2015 07:01:17 +0000
Message-ID: <DB3PR03MB0780C4AC3E2EB773B21D697B9DF40@DB3PR03MB0780.eurprd03.prod.outlook.com>
References: <54EB4B87.2080202@cisco.com> <3B0A1BED22CAD649A1B3E97BE5DDD68B5A890926@szxema506-mbs.china.huawei.com>, <55195849.7090003@cisco.com> <1427777706620.6234@ecitele.com> <3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C55DA@szxema506-mbs.china.huawei.com>
In-Reply-To: <3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C55DA@szxema506-mbs.china.huawei.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [147.234.56.21]
authentication-results: huawei.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:DB3PR03MB0780;
x-microsoft-antispam-prvs: <DB3PR03MB0780D60BED088D44B72E590B9DF40@DB3PR03MB0780.eurprd03.prod.outlook.com>
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(252514010)(377454003)(164054003)(2900100001)(50986999)(76176999)(102836002)(19625215002)(15975445007)(92566002)(19580405001)(19617315012)(19300405004)(54356999)(2950100001)(16236675004)(46102003)(62966003)(19609705001)(77156002)(230783001)(87936001)(86362001)(40100003)(19580395003)(122556002)(76576001)(66066001)(106116001)(2656002)(110136001)(93886004)(33656002); DIR:OUT; SFP:1102; SCL:1; SRVR:DB3PR03MB0780; H:DB3PR03MB0780.eurprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5005006)(5002010); SRVR:DB3PR03MB0780; BCL:0; PCL:0; RULEID:;  SRVR:DB3PR03MB0780; 
x-forefront-prvs: 0532BF6DC2
Content-Type: multipart/alternative; boundary="_000_DB3PR03MB0780C4AC3E2EB773B21D697B9DF40DB3PR03MB0780eurp_"
MIME-Version: 1.0
X-OriginatorOrg: ecitele.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 31 Mar 2015 07:01:17.3863 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 2c514a61-08de-4519-b4c0-921fef62c42a
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DB3PR03MB0780
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/ut0vlXtSuOlz03Cm48ruoQ8hJMc>
Cc: "l2vpn-chairs@tools.ietf.org" <l2vpn-chairs@tools.ietf.org>, "draft-ietf-l2vpn-vpls-pe-etree@tools.ietf.org" <draft-ietf-l2vpn-vpls-pe-etree@tools.ietf.org>, "pals-chairs@tools.ietf.org" <pals-chairs@tools.ietf.org>, "bess-chairs@tools.ietf.org" <bess-chairs@tools.ietf.org>, "pals@ietf.org" <pals@ietf.org>, "bess@ietf.org" <bess@ietf.org>, "stbryant@cisco.com" <stbryant@cisco.com>
Subject: Re: [bess] [Pals] Shepherd review of draft-ietf-l2vpn-vpls-pe-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Mar 2015 07:01:45 -0000

--_000_DB3PR03MB0780C4AC3E2EB773B21D697B9DF40DB3PR03MB0780eurp_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_DB3PR03MB0780C4AC3E2EB773B21D697B9DF40DB3PR03MB0780eurp_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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=

--_000_DB3PR03MB0780C4AC3E2EB773B21D697B9DF40DB3PR03MB0780eurp_--


From nobody Tue Mar 31 00:49:23 2015
Return-Path: <jiangyuanlong@huawei.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9C7BD1B2B21; Tue, 31 Mar 2015 00:49:21 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.21
X-Spam-Level: 
X-Spam-Status: No, score=-4.21 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id SwJ0CktwUdol; Tue, 31 Mar 2015 00:49:18 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 975B51A9042; Tue, 31 Mar 2015 00:49:16 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml406-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BQW83396; Tue, 31 Mar 2015 07:49:14 +0000 (GMT)
Received: from SZXEMA414-HUB.china.huawei.com (10.82.72.73) by lhreml406-hub.china.huawei.com (10.201.5.243) with Microsoft SMTP Server (TLS) id 14.3.158.1; Tue, 31 Mar 2015 08:49:13 +0100
Received: from SZXEMA506-MBS.china.huawei.com ([169.254.4.68]) by SZXEMA414-HUB.china.huawei.com ([10.82.72.73]) with mapi id 14.03.0158.001; Tue, 31 Mar 2015 15:49:02 +0800
From: Jiangyuanlong <jiangyuanlong@huawei.com>
To: "stbryant@cisco.com" <stbryant@cisco.com>, "draft-ietf-l2vpn-vpls-pe-etree@tools.ietf.org" <draft-ietf-l2vpn-vpls-pe-etree@tools.ietf.org>, "pals@ietf.org" <pals@ietf.org>
Thread-Topic: [Pals] Shepherd review of draft-ietf-l2vpn-vpls-pe-etree
Thread-Index: AQHQT4AUGI+hr/i+r0Gj9wcHU/COq50CjlQggDIx3YCAAZnXAA==
Date: Tue, 31 Mar 2015 07:49:01 +0000
Message-ID: <3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C5632@szxema506-mbs.china.huawei.com>
References: <54EB4B87.2080202@cisco.com> <3B0A1BED22CAD649A1B3E97BE5DDD68B5A890926@szxema506-mbs.china.huawei.com> <55195849.7090003@cisco.com>
In-Reply-To: <55195849.7090003@cisco.com>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.66.76.118]
Content-Type: multipart/alternative; boundary="_000_3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C5632szxema506mbschi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/GYf_nMwpX8atRs2z7YB3bXRTh2A>
Cc: "l2vpn-chairs@tools.ietf.org" <l2vpn-chairs@tools.ietf.org>, "bess-chairs@tools.ietf.org" <bess-chairs@tools.ietf.org>, "pals-chairs@tools.ietf.org" <pals-chairs@tools.ietf.org>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] [Pals] Shepherd review of draft-ietf-l2vpn-vpls-pe-etree
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Mar 2015 07:49:21 -0000

--_000_3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C5632szxema506mbschi_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C5632szxema506mbschi_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_3B0A1BED22CAD649A1B3E97BE5DDD68B5A8C5632szxema506mbschi_--


From nobody Tue Mar 31 13:10:59 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 63C9C1ACE98 for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 13:10:57 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -5.71
X-Spam-Level: 
X-Spam-Status: No, score=-5.71 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_12=0.6, J_CHICKENPOX_14=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id MmLEwgtim7lj for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 13:10:55 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 05CD61ACE8B for <bess@ietf.org>; Tue, 31 Mar 2015 13:10:55 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id DBE26CC02EA4E; Tue, 31 Mar 2015 20:10:49 +0000 (GMT)
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2VKArKM014208 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Tue, 31 Mar 2015 22:10:53 +0200
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR711WXCHHUB01.zeu.alcatel-lucent.com ([135.239.2.111]) with mapi id 14.03.0195.001; Tue, 31 Mar 2015 22:10:53 +0200
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Antoni Przygienda <antoni.przygienda@ericsson.com>, Erik Nordmark <nordmark@acm.org>, "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAPgqSA///mFgCABtbMgIAB8pOA
Date: Tue, 31 Mar 2015 20:10:52 +0000
Message-ID: <D1404756.6D045%jorge.rabadan@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <D138D9BF.6C4CE%jorge.rabadan@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828B24C@eusaamb103.ericsson.se>
In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F1828B24C@eusaamb103.ericsson.se>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.40]
Content-Type: text/plain; charset="utf-8"
Content-ID: <B2478CEA96A67D4EBD81A6C124B82E56@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/cHzb3EpZE2tp4UTZIYCjqUbEsgk>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Mar 2015 20:10:57 -0000
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From nobody Tue Mar 31 13:11:01 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CB1451ACE8B for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 13:10:57 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.309
X-Spam-Level: 
X-Spam-Status: No, score=-6.309 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tscN3gTZwAip for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 13:10:54 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A9B9A1ACE63 for <bess@ietf.org>; Tue, 31 Mar 2015 13:10:52 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id 06673E76BE6F3; Tue, 31 Mar 2015 20:10:47 +0000 (GMT)
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t2VKAo0X014106 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Tue, 31 Mar 2015 22:10:50 +0200
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR712WXCHHUB03.zeu.alcatel-lucent.com ([135.239.2.74]) with mapi id 14.03.0195.001; Tue, 31 Mar 2015 22:10:50 +0200
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Antoni Przygienda <antoni.przygienda@ericsson.com>, Robert Raszuk <robert@raszuk.net>, "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAZ/NmAAAeuvZAAEFu+gAAIR4HQAKZAb4AAAO64gAAB4cEAAAEeZgAAAElRAAAB2icAAD3RCQA=
Date: Tue, 31 Mar 2015 20:10:49 +0000
Message-ID: <D140442C.6D02F%jorge.rabadan@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se> <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se> <61BFC96E-350C-4F77-84E6-EFA6794B5EC0@alcatel-lucent.com> <CA+b+ERmVVaz4jm3+CRc7WMfrQn1TN5=sEiK-Qq+nngtw3yw--Q@mail.gmail.com> <A606D412-1B8B-4DAB-B4B0-1B19F31FF50B@alcatel-lucent.com> <15A9EAE0-D02F-4476-88FC-2E3BD66FA33D@alcatel-lucent.com> <CA+b+ER=RsuRBAsB=i+CgyTXOt_D9_J1QtSMWOYzn7moK1CV4DQ@mail.gmail.com> <2E4BB27CAB87BF43B4207C0E55860F1828B205@eusaamb103.ericsson.se>
In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F1828B205@eusaamb103.ericsson.se>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.39]
Content-Type: multipart/alternative; boundary="_000_D140442C6D02Fjorgerabadanalcatellucentcom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/TvEfhJvo5jVmKzuSRjzGWJmEsfQ>
Cc: Erik Nordmark <nordmark@acm.org>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Mar 2015 20:10:58 -0000

--_000_D140442C6D02Fjorgerabadanalcatellucentcom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_D140442C6D02Fjorgerabadanalcatellucentcom_
Content-Type: text/html; charset="utf-8"
Content-ID: <497C09B1955E114F895AB12126725846@exchange.lucent.com>
Content-Transfer-Encoding: base64
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--_000_D140442C6D02Fjorgerabadanalcatellucentcom_--


From nobody Tue Mar 31 16:51:10 2015
Return-Path: <antoni.przygienda@ericsson.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D8C7C1A8729 for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 16:51:08 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.2
X-Spam-Level: 
X-Spam-Status: No, score=-4.2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id VKodTAXEchrf for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 16:51:05 -0700 (PDT)
Received: from usevmg21.ericsson.net (usevmg21.ericsson.net [198.24.6.65]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B77821A6F01 for <bess@ietf.org>; Tue, 31 Mar 2015 16:51:04 -0700 (PDT)
X-AuditID: c6180641-f790b6d000004359-fd-551ad0a259fa
Received: from EUSAAHC005.ericsson.se (Unknown_Domain [147.117.188.87]) by usevmg21.ericsson.net (Symantec Mail Security) with SMTP id DE.0B.17241.2A0DA155; Tue, 31 Mar 2015 18:51:46 +0200 (CEST)
Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC005.ericsson.se ([147.117.188.87]) with mapi id 14.03.0210.002; Tue, 31 Mar 2015 19:51:02 -0400
From: Antoni Przygienda <antoni.przygienda@ericsson.com>
To: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, Robert Raszuk <robert@raszuk.net>, "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAZ/NmAAAeuvZAAEFu+gAAIR4HQALLTFoAAAO63gAAB4cEAAAEeZwAAAElQAAAHRjGAAEgP8YAAANpwQA==
Date: Tue, 31 Mar 2015 23:51:03 +0000
Message-ID: <2E4BB27CAB87BF43B4207C0E55860F1828C971@eusaamb103.ericsson.se>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <2E4BB27CAB87BF43B4207C0E55860F18288682@eusaamb103.ericsson.se> <D51715DF-AD05-4863-A0F3-34B93B7C5945@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828AE5A@eusaamb103.ericsson.se> <61BFC96E-350C-4F77-84E6-EFA6794B5EC0@alcatel-lucent.com> <CA+b+ERmVVaz4jm3+CRc7WMfrQn1TN5=sEiK-Qq+nngtw3yw--Q@mail.gmail.com> <A606D412-1B8B-4DAB-B4B0-1B19F31FF50B@alcatel-lucent.com> <15A9EAE0-D02F-4476-88FC-2E3BD66FA33D@alcatel-lucent.com> <CA+b+ER=RsuRBAsB=i+CgyTXOt_D9_J1QtSMWOYzn7moK1CV4DQ@mail.gmail.com> <2E4BB27CAB87BF43B4207C0E55860F1828B205@eusaamb103.ericsson.se> <D140442C.6D02F%jorge.rabadan@alcatel-lucent.com>
In-Reply-To: <D140442C.6D02F%jorge.rabadan@alcatel-lucent.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [147.117.188.9]
Content-Type: multipart/alternative; boundary="_000_2E4BB27CAB87BF43B4207C0E55860F1828C971eusaamb103ericsso_"
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFprAIsWRmVeSWpSXmKPExsUyuXRPuO6iC1KhBm/mqFmsOD6T2eLbpSUs Fv1TdzJZNC1sYraY/fgHmwOrx+Ur3h6tz/ayeixZ8pPJY/fGBUwBLFFcNimpOZllqUX6dglc Gb2Pm5gLpthV7Djt1cC4xaaLkZNDQsBEYvvU94wQtpjEhXvr2boYuTiEBI4ySlx985sVwlnO KLFw5TawKjYBC4nL354yg9giAosZJd6fUwaxmQVcJeYsfs8GYgsLKEi0d3yFqlGUmP98HSOE XSdxbt1ysDiLgKrE6VebwOp5Bbwldk7+zQSx7AGrxJnFr5hAEpwCdhK7Fx4AK2IEOu/7qTVM EMvEJW49mc8EcbaAxJI955khbFGJl4//sULYihL7+qezQ9TnS0zvmMoOsUxQ4uTMJywTGEVn IRk1C0nZLCRlsxg5gOKaEut36UOUKEpM6X7IDmFrSLTOmcuOLL6AkX0VI0dpcWpZbrqR4SZG YBQek2Bz3MG44JPlIUYBDkYlHl6FfslQIdbEsuLK3EOM0hwsSuK8ZVcOhggJpCeWpGanphak FsUXleakFh9iZOLglGpgNFXQ0JLfr/f51Y2Mpz/vGjz/Y8Yvs/P+Zi33au4XeRfqdI5P89qW LdO4TUbi1r0vZ+KXNgb9Tvznee3nrm1P4j/8uJyps2jhM9PwTNHHk17GT5u5IbzeJGWB3zZD Rt6fR9Jynl6fveflrdS+asM3MRPevm5cerJjgr/80sUmH1dbKVjIKWvMFVViKc5INNRiLipO BADN1KUrowIAAA==
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/nYTw4XhROvP1S1UEAlwPFA6z7P8>
Cc: Erik Nordmark <nordmark@acm.org>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Mar 2015 23:51:09 -0000

--_000_2E4BB27CAB87BF43B4207C0E55860F1828C971eusaamb103ericsso_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64

VGhlIGNoYWxsZW5nZSBmb3IgSVB2NCBpcyB0aGF0IEkgZG9u4oCZdCBzZWUgYW4gZWFzeSB3YXkg
dG8gbGVhcm4gZHluYW1pY2FsbHkgZnJvbSBhY2Nlc3MgYXR0YWNobWVudCBjaXJjdWl0cyB0aGF0
IGEgZ2l2ZW4gaXB2NCBpcyBhbnljYXN0LiBFdmVuIGZvciBkZWZhdWx0IGdhdGV3YXlzLCBpZiB0
aGV5IGFyZSBpbnRlZ3JhdGVkIGluIHRoZSBFVlBOIFBFLCB3ZSBhcmUgZ29vZCwgYnV0IGlmIHRo
ZXkgYXJlIGV4dGVybmFsIGFuZCBjb25uZWN0ZWQgdG8gYSBNQUMtVlJGLCBpdCBpcyBub3Qgc28g
Y2xlYXIgaG93IHRvIGxlYXJuIHRoYXQgKHVubGVzcyB5b3UgbGVhcm4gdGhvc2UgZW50cmllcyBm
cm9tIHRoZSBtYW5hZ2VtZW50IGludGVyZmFjZSkuDQpbVG9ueSBzYWl6Ol0gYWdyZWVkDQoNCk9u
ZSBvZiB0aGUgcmVhc29ucyB3aHkgd2UgaGF2ZSBsb3RzIG9mIOKAnFNIT1VMRHPigJ0gaW4gdGhl
IGRyYWZ0IGFuZCBub3Qg4oCcTVVTVOKAnSBpcyBiZWNhdXNlIHRoZSBpbXBsZW1lbnRhdGlvbiBo
YXMgdG8gYmUgZmxleGlibGUgZW5vdWdoIHRvIGJlIGNvbmZpZ3VyZWQgaW4gYSBkaWZmZXJlbnQg
d2F5IGRlcGVuZGluZyBvbiB0aGUgdXNlLWNhc2UsIHdoaWNoIGlzIG9uZSBvZiB0aGUgcG9pbnRz
IHRoYXQgVG9ueSBtZW50aW9ucyBiZWxvdy4gSW4gdGhlIHVzZS1jYXNlIGRlc2NyaWJlZCBhdCB0
aGUgbW9tZW50IHRoZXJlIGlzIG5vIGFueWNhc3QgYW5kIGR1cGxpY2F0ZSBJUCBkZXRlY3Rpb24g
aXMgdmVyeSBpbXBvcnRhbnQuIFdlIHdpbGwgYWRkIHRoZSBEQyB1c2UgY2FzZSBpbiB0aGUgbmV4
dCByZXYgYXMgc3VnZ2VzdGVkIGJ5IFJvYmVydCBhbmQgb3RoZXJzLg0KDQpbVG9ueSBzYWl6Ol0g
YWdyZWVkLg0KDQotLS0gdG9ueQ0K

--_000_2E4BB27CAB87BF43B4207C0E55860F1828C971eusaamb103ericsso_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_2E4BB27CAB87BF43B4207C0E55860F1828C971eusaamb103ericsso_--


From nobody Tue Mar 31 16:53:31 2015
Return-Path: <antoni.przygienda@ericsson.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D1B081A8728 for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 16:53:29 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.601
X-Spam-Level: 
X-Spam-Status: No, score=-3.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_14=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XJczuRAZ1Vrn for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 16:53:28 -0700 (PDT)
Received: from usevmg21.ericsson.net (usevmg21.ericsson.net [198.24.6.65]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 58B6D1A7005 for <bess@ietf.org>; Tue, 31 Mar 2015 16:53:27 -0700 (PDT)
X-AuditID: c6180641-f790b6d000004359-81-551ad131fffd
Received: from EUSAAHC006.ericsson.se (Unknown_Domain [147.117.188.90]) by usevmg21.ericsson.net (Symantec Mail Security) with SMTP id 10.4B.17241.131DA155; Tue, 31 Mar 2015 18:54:09 +0200 (CEST)
Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC006.ericsson.se ([147.117.188.90]) with mapi id 14.03.0210.002; Tue, 31 Mar 2015 19:53:26 -0400
From: Antoni Przygienda <antoni.przygienda@ericsson.com>
To: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>, Erik Nordmark <nordmark@acm.org>, "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAZ/NmAAAuXkQAAFTe+gAEF5IcAAADNZlA=
Date: Tue, 31 Mar 2015 23:53:25 +0000
Message-ID: <2E4BB27CAB87BF43B4207C0E55860F1828C982@eusaamb103.ericsson.se>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <D138D9BF.6C4CE%jorge.rabadan@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828B24C@eusaamb103.ericsson.se> <D1404756.6D045%jorge.rabadan@alcatel-lucent.com>
In-Reply-To: <D1404756.6D045%jorge.rabadan@alcatel-lucent.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [147.117.188.9]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFnrLLMWRmVeSWpSXmKPExsUyuXRPlK7hRalQg2nzNCxWHJ/JbPHt0hIW i/6pO5ksZj/+webA4nH5irdH67O9rB5LlvxkCmCO4rJJSc3JLEst0rdL4MrYuvkcY8ES5Yq5 LS3sDYwblLoYOTkkBEwktq9/yQJhi0lcuLeeDcQWEjjKKNF7RKuLkQvIXs4ocfbrTLAEm4CF xOVvT5lBEiICixgllnc0gSWYBZQlDs8/zw5iCwsoSLR3fGUGsUUEFCXmP1/HCGH7SaxfN5+p i5GDg0VAVeJYexFImFfAW2L65rlMEMsmMEnMvHGQFSTBKWAn8fzXIiYQmxHouu+n1jBB7BKX uPVkPhPE1QISS/acZ4awRSVePv7HCmErSuzrn84OsotZQFNi/S59iFZFiSndD9kh9gpKnJz5 hGUCo9gsJFNnIXTMQtIxC0nHAkaWVYwcpcWpZbnpRoabGIHxc0yCzXEH44JPlocYBTgYlXh4 FfolQ4VYE8uKK3MPMUpzsCiJ85ZdORgiJJCeWJKanZpakFoUX1Sak1p8iJGJg1OqgbFAhOf0 6j0CZRUnDq0qOaR09YPMqXeXPDYlHprynGt68JPrs7RuWPpqXLhbt6igWL0tJVPodfhZkc+T z30LC5Y1fJP6pS98h756RdVLiexvz0pbdHwjl34Pan+/xivn3bI9gtuvbPrDdNTB99jrqHap 9S2GB5dzMnCFc3SqMQcqLQ99fynPyVSJpTgj0VCLuag4EQDFPMbwgAIAAA==
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/SltmkDATZ-sgMg241vSLTAq1UOo>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Mar 2015 23:53:30 -0000
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==


From nobody Tue Mar 31 17:23:55 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2852F1AD376 for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 17:23:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.31
X-Spam-Level: 
X-Spam-Status: No, score=-6.31 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_14=0.6, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gZaaFJi4Qh9c for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 17:23:52 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpgre-esg-01.alcatel-lucent.com [135.245.210.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2D8451A0016 for <bess@ietf.org>; Tue, 31 Mar 2015 17:23:51 -0700 (PDT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (unknown [135.239.2.122]) by Websense Email Security Gateway with ESMTPS id 07AB75A601E1F; Wed,  1 Apr 2015 00:23:46 +0000 (GMT)
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id t310Nnk6028638 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 1 Apr 2015 02:23:49 +0200
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR711WXCHHUB01.zeu.alcatel-lucent.com ([135.239.2.111]) with mapi id 14.03.0195.001; Wed, 1 Apr 2015 02:23:49 +0200
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: Antoni Przygienda <antoni.przygienda@ericsson.com>, Erik Nordmark <nordmark@acm.org>, "Henderickx, Wim (Wim)" <wim.henderickx@alcatel-lucent.com>
Thread-Topic: [bess] ARP ND draft
Thread-Index: AdBnDY3J9A+gzIbxRfejJpOtp2cILgAPgqSA///mFgCABtbMgIAB8pOAgACzlID//5MjgA==
Date: Wed, 1 Apr 2015 00:23:49 +0000
Message-ID: <D1408815.6D137%jorge.rabadan@alcatel-lucent.com>
References: <2E4BB27CAB87BF43B4207C0E55860F18287F74@eusaamb103.ericsson.se> <55134595.7060209@acm.org> <D138D9BF.6C4CE%jorge.rabadan@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828B24C@eusaamb103.ericsson.se> <D1404756.6D045%jorge.rabadan@alcatel-lucent.com> <2E4BB27CAB87BF43B4207C0E55860F1828C982@eusaamb103.ericsson.se>
In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F1828C982@eusaamb103.ericsson.se>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.38]
Content-Type: text/plain; charset="utf-8"
Content-ID: <37DBE037F0F9D54BB0C2259F2174E9A6@exchange.lucent.com>
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/G3FgahD5iN0XMD3hNJg_qTRkSLY>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: Re: [bess] ARP ND draft
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Apr 2015 00:23:54 -0000
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==


From nobody Tue Mar 31 19:28:15 2015
Return-Path: <jorge.rabadan@alcatel-lucent.com>
X-Original-To: bess@ietfa.amsl.com
Delivered-To: bess@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E4F5F1A0A85 for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 19:28:12 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.909
X-Spam-Level: 
X-Spam-Status: No, score=-6.909 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 42NpiGh6yjun for <bess@ietfa.amsl.com>; Tue, 31 Mar 2015 19:28:09 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D822A1A065C for <bess@ietf.org>; Tue, 31 Mar 2015 19:28:08 -0700 (PDT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (unknown [135.239.2.42]) by Websense Email Security Gateway with ESMTPS id AC3CDFFD86124; Wed,  1 Apr 2015 02:28:06 +0000 (GMT)
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id t312S783031669 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 1 Apr 2015 04:28:07 +0200
Received: from FR711WXCHMBA03.zeu.alcatel-lucent.com ([169.254.3.123]) by FR711WXCHHUB01.zeu.alcatel-lucent.com ([135.239.2.111]) with mapi id 14.03.0195.001; Wed, 1 Apr 2015 04:28:07 +0200
From: "Rabadan, Jorge (Jorge)" <jorge.rabadan@alcatel-lucent.com>
To: "draft-ietf-l2vpn-pbb-evpn@tools.ietf.org" <draft-ietf-l2vpn-pbb-evpn@tools.ietf.org>
Thread-Topic: PBB-EVPN draft - use of sticky bit in mac mobility ext. com.
Thread-Index: AQHQbCN84MFbGAANckOZK5Xiu2V3gA==
Date: Wed, 1 Apr 2015 02:28:07 +0000
Message-ID: <D140A5C4.6D1DF%jorge.rabadan@alcatel-lucent.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.4.8.150116
x-originating-ip: [135.239.27.39]
Content-Type: multipart/alternative; boundary="_000_D140A5C46D1DFjorgerabadanalcatellucentcom_"
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/bess/ZgGuVX6dd4EdAJ1kBp_njBlENv0>
Cc: "bess@ietf.org" <bess@ietf.org>
Subject: [bess] PBB-EVPN draft - use of sticky bit in mac mobility ext. com.
X-BeenThere: bess@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: BGP-Enabled ServiceS working group discussion list <bess.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bess>, <mailto:bess-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/bess/>
List-Post: <mailto:bess@ietf.org>
List-Help: <mailto:bess-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bess>, <mailto:bess-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 01 Apr 2015 02:28:13 -0000

--_000_D140A5C46D1DFjorgerabadanalcatellucentcom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64

RGVhciBhdXRob3JzLA0KDQpJbiBQQkItRVZQTiB3ZSBhZHZlcnRpc2UgUEUgc2hhcmVkIEJNQUNz
IG9yIGRlZGljYXRlZCAocGVyLUVTKSBCTUFDcy4gRWl0aGVyIHdheSwgSSBiZWxpZXZlIGl0IGlz
IGEgZ29vZCBwcmFjdGljZSB0byBhZHZlcnRpc2UgdGhlbSBhcyDigJlzdGF0aWPigJkgaS5lLiBh
bG9uZyB3aXRoIHRoZSDigJlzdGlja3nigJkgYml0LiBUaGF0IHByb3ZpZGVzIGEgbmF0dXJhbCBw
cm90ZWN0aW9uIGFnYWluc3QgQk1BQ3MgdGhhdCBtaWdodCBiZSBsZWFybnQgbG9jYWxseSBhbmQg
YXJlIG5vdCDigJhtYW5hZ2Vk4oCZLg0KDQpJbiBSRkM3NDMyLCB3aGVuIHRoZSBzdGlja3kgYml0
IGlzIHNldCwgdGhlIHNlcXVlbmNlIG51bWJlciBpcyB6ZXJvLg0KRm9yIFBCQi1FVlBOLCB0aGF0
IG1lYW5zIHRoYXQgd2hlbiBzaGFyZWQgQk1BQ3MgYXJlIHVzZWQgYW5kIHBlci1JU0lEIGxvYWQg
YmFsYW5jaW5nIG11bHRpaG9taW5nIGlzIGluIHBsYWNlLCB0aGUgc2hhcmVkIEJNQUNzIGNhbm5v
dCBiZSBhZHZlcnRpc2VkIGFzIHN0YXRpYyAoc2luY2UgdGhlIHNlcXVlbmNlIG51bWJlciBpcyB1
c2VkIGFzIGEgQ01BQyBmbHVzaCBub3RpZmljYXRpb24pLg0KDQpTaW5jZSB0aGUgUEUgQk1BQ3Mg
YXJlIG5vdCBzdWJqZWN0IHRvIG1vYmlsaXR5IHByb2NlZHVyZXMgYW5kIGFyZSBieSBuYXR1cmUg
4oCYc3RhdGlj4oCZIGFuZCBtYW5hZ2VkLCB3b3VsZCBpdCBiZSBwb3NzaWJsZSB0byBleHBsaWNp
dGx5IGFsbG93IGluIHRoZSBwYmItZXZwbiBkcmFmdCB0aGUgYWR2ZXJ0aXNlbWVudCBvZiB0aGUg
c3RpY2t5IGJpdCBhbG9uZyB3aXRoIGEgc2VxdWVuY2UgbnVtYmVyLCB3aGVuIHRoZSBtYWMtbW9i
aWxpdHkgZXh0ZW5kZWQgY29tbXVuaXR5IGlzIHVzZWQgZm9yIENNQUMgZmx1c2ggbm90aWZpY2F0
aW9uPw0KDQpUaGlzIGNhbiBiZSBvcHRpb25hbCBhbmQgd291bGQgYWxsb3cgYW4gZXh0cmEgbGV2
ZWwgb2Ygc2VjdXJpdHkgaW4gYSBQQkItRVZQTiBuZXR3b3JrLg0KSWYgeW91IGFncmVlIHdpdGgg
dGhhdCwgSSBjYW4gcHJvdmlkZSBhIHRleHQgaWYgbmVlZGVkLg0KDQpMb29raW5nIGZvcndhcmQg
dG8geW91ciBmZWVkYmFjay4NClRoYW5rIHlvdS4NCkpvcmdlDQo=

--_000_D140A5C46D1DFjorgerabadanalcatellucentcom_
Content-Type: text/html; charset="utf-8"
Content-ID: <C5B8161C977B234BA4F91C27581B4B85@exchange.lucent.com>
Content-Transfer-Encoding: base64
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--_000_D140A5C46D1DFjorgerabadanalcatellucentcom_--

