
From nick@inex.ie  Sat Dec  1 08:19:51 2012
Return-Path: <nick@inex.ie>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9E6F211E809C for <grow@ietfa.amsl.com>; Sat,  1 Dec 2012 08:19:51 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.6
X-Spam-Level: 
X-Spam-Status: No, score=-2.6 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, NO_RELAYS=-0.001]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id O+c82gVpoGUt for <grow@ietfa.amsl.com>; Sat,  1 Dec 2012 08:19:51 -0800 (PST)
Received: from mail.acquirer.com (mail.netability.ie [IPv6:2a03:8900:0:100::5]) by ietfa.amsl.com (Postfix) with ESMTP id E792011E8097 for <grow@ietf.org>; Sat,  1 Dec 2012 08:19:49 -0800 (PST)
X-Envelope-To: grow@ietf.org
Received: from cupcake.foobar.org ([IPv6:2001:4d68:2002:100:a4d3:c05a:5c65:3e44]) (authenticated bits=0) by mail.acquirer.com (8.14.4/8.14.4) with ESMTP id qB1GIING015084 (version=TLSv1/SSLv3 cipher=DHE-RSA-CAMELLIA256-SHA bits=256 verify=NO); Sat, 1 Dec 2012 16:18:23 GMT (envelope-from nick@inex.ie)
Message-ID: <50BA2E1B.5010409@inex.ie>
Date: Sat, 01 Dec 2012 16:19:39 +0000
From: Nick Hilliard <nick@inex.ie>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:17.0) Gecko/17.0 Thunderbird/17.0
MIME-Version: 1.0
To: Christopher Morrow <christopher.morrow@gmail.com>
References: <CAL9jLabt0S98HEygsQp=a98ARwiRxJ2x3F2=H5w-pDRbyqWE7A@mail.gmail.com>
In-Reply-To: <CAL9jLabt0S98HEygsQp=a98ARwiRxJ2x3F2=H5w-pDRbyqWE7A@mail.gmail.com>
X-Enigmail-Version: 1.4.6
X-Company-Info-1: Internet Neutral Exchange Association Limited. Registered in Ireland No. 253804
X-Company-Info-2: Registered Offices: 1-2, Marino Mart, Fairview, Dublin 3
X-Company-Info-3: Internet Neutral Exchange Association Limited is limited by guarantee
X-Company-Info-4: Offices: 4027 Kingswood Road, Citywest, Dublin 24.
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Cc: "grow@ietf.org grow@ietf.org" <grow@ietf.org>, draft-mcpherson-irr-routing-policy-considerations@tools.ietf.org, grow-ads@tools.ietf.org, grow-chairs@tools.ietf.org
Subject: Re: [GROW] WG Adoption call: draft-mcpherson-irr-routing-policy-considerations (END: 11/28/2012)
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 01 Dec 2012 16:19:51 -0000

oops, is it too late to add a hum in favour of this ID from here?

Nick

On 14/11/2012 22:35, Christopher Morrow wrote:
> Grow Folks,
> the authors of:  draft-mcpherson-irr-routing-policy-considerations
>   <http://tools.ietf.org/html/draft-mcpherson-irr-routing-policy-considerations-01>
> 
> have asked if their work is appropriate for the GROW-wg, could we take
> the next 14 days (ending Nov 28, 2012) to think/discuss this draft and
> it's efficacy to work in the GROW arena? The abstract of the draft is:
>   "The purpose of this document is to catalog past issues influencing
>    the efficacy of Internet Routing Registries (IRR) for inter-domain
>    routing policy specification and application in the global routing
>    system over the past two decades.  Additionally, it provides a
>    discussion regarding which of these issues are still problematic in
>    practice, and which are simply artifacts that are no longer
>    applicable but continue to stifle inter-provider policy-based
>    filtering adoption and IRR utility to this day."
> 
> Thanks!
> -Chris
> (co-chair)
> _______________________________________________
> GROW mailing list
> GROW@ietf.org
> https://www.ietf.org/mailman/listinfo/grow
> 


From dogwallah@gmail.com  Sat Dec  1 08:34:21 2012
Return-Path: <dogwallah@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6898721E803F for <grow@ietfa.amsl.com>; Sat,  1 Dec 2012 08:34:21 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.599
X-Spam-Level: 
X-Spam-Status: No, score=-3.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wd5slEsas-UB for <grow@ietfa.amsl.com>; Sat,  1 Dec 2012 08:34:11 -0800 (PST)
Received: from mail-vc0-f172.google.com (mail-vc0-f172.google.com [209.85.220.172]) by ietfa.amsl.com (Postfix) with ESMTP id E934321E8039 for <grow@ietf.org>; Sat,  1 Dec 2012 08:34:10 -0800 (PST)
Received: by mail-vc0-f172.google.com with SMTP id fw7so689519vcb.31 for <grow@ietf.org>; Sat, 01 Dec 2012 08:34:10 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=yKMcbTGzpXJoQ5dElYMBf53gnul/fL61phWBbcQeDpU=; b=rh5PQaSyNgF7qlyBAiF1zzfGcdxUZpn6Wn+OFUh08WC+0YbW33wy0lSF5JDAkNSPb8 684Lhw++tlVlizzhfWWnA2D0nY6wghDges47C1q5VywAF3BhxHOKxO9YXI13iPrAhUpd TFqjX7p6q7g+E18UPXz17DRDubJewQdv4gjmSOrCmlzfTIE4ShHLq/ZXudShy/rMNWu6 mNKKy2GWnVkZ90zDmbg/GNzclPRH2PGUtsGIrRZ1MOTS7Z3H9+mCbieKlc4vwEBt+rDg onmcWxGILC6lIHNJVuvwRpQFMnU3hc73HLYc0MQdZfQ/PhMg08P6LV2UrutW3l2juynn 9MoQ==
Received: by 10.220.240.141 with SMTP id la13mr4105779vcb.39.1354379650257; Sat, 01 Dec 2012 08:34:10 -0800 (PST)
MIME-Version: 1.0
Received: by 10.58.80.98 with HTTP; Sat, 1 Dec 2012 08:33:29 -0800 (PST)
In-Reply-To: <50BA2E1B.5010409@inex.ie>
References: <CAL9jLabt0S98HEygsQp=a98ARwiRxJ2x3F2=H5w-pDRbyqWE7A@mail.gmail.com> <50BA2E1B.5010409@inex.ie>
From: McTim <dogwallah@gmail.com>
Date: Sat, 1 Dec 2012 11:33:29 -0500
Message-ID: <CACAaNxhg6VrCDC1PYgWTbErKJStKZDYgzdKj0mtjVDBsHmWJeQ@mail.gmail.com>
To: Nick Hilliard <nick@inex.ie>
Content-Type: text/plain; charset=ISO-8859-1
Cc: "grow@ietf.org grow@ietf.org" <grow@ietf.org>, grow-chairs@tools.ietf.org
Subject: Re: [GROW] WG Adoption call: draft-mcpherson-irr-routing-policy-considerations (END: 11/28/2012)
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 01 Dec 2012 16:34:21 -0000

On Sat, Dec 1, 2012 at 11:19 AM, Nick Hilliard <nick@inex.ie> wrote:
> oops, is it too late to add a hum in favour of this ID from here?

if not too late, I can add mine, apologies, been behind on list mails.


-- 
Cheers,

McTim
"A name indicates what we seek. An address indicates where it is. A
route indicates how we get there."  Jon Postel

From aservin@lacnic.net  Sat Dec  1 12:56:45 2012
Return-Path: <aservin@lacnic.net>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AC27D11E80B8 for <grow@ietfa.amsl.com>; Sat,  1 Dec 2012 12:56:45 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.845
X-Spam-Level: 
X-Spam-Status: No, score=-0.845 tagged_above=-999 required=5 tests=[AWL=-1.755, BAYES_00=-2.599, FH_HOST_EQ_D_D_D_D=0.765, HOST_EQ_DIALUP=0.862, RCVD_IN_PBL=0.905, RCVD_IN_SORBS_DUL=0.877, RDNS_DYNAMIC=0.1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OUWJOUT2nUEJ for <grow@ietfa.amsl.com>; Sat,  1 Dec 2012 12:56:44 -0800 (PST)
Received: from mail.lacnic.net.uy (mail.lacnic.net.uy [IPv6:2001:13c7:7001:4000::3]) by ietfa.amsl.com (Postfix) with ESMTP id 6F7BC11E80A5 for <grow@ietf.org>; Sat,  1 Dec 2012 12:56:44 -0800 (PST)
Received: from [192.168.1.133] (r186-48-230-195.dialup.adsl.anteldata.net.uy [186.48.230.195]) by mail.lacnic.net.uy (Postfix) with ESMTP id 7D9F2308444 for <grow@ietf.org>; Sat,  1 Dec 2012 18:56:29 -0200 (UYST)
Message-ID: <50BA6EFB.4020103@lacnic.net>
Date: Sat, 01 Dec 2012 18:56:27 -0200
From: Arturo Servin <aservin@lacnic.net>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:17.0) Gecko/17.0 Thunderbird/17.0
MIME-Version: 1.0
To: grow@ietf.org
References: <CAL9jLabt0S98HEygsQp=a98ARwiRxJ2x3F2=H5w-pDRbyqWE7A@mail.gmail.com>
In-Reply-To: <CAL9jLabt0S98HEygsQp=a98ARwiRxJ2x3F2=H5w-pDRbyqWE7A@mail.gmail.com>
X-Enigmail-Version: 1.4.6
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-LACNIC.uy-MailScanner-Information: Please contact the ISP for more information
X-LACNIC.uy-MailScanner: Found to be clean
X-LACNIC.uy-MailScanner-SpamCheck: 
X-LACNIC.uy-MailScanner-From: aservin@lacnic.net
Subject: Re: [GROW] WG Adoption call: draft-mcpherson-irr-routing-policy-considerations (END: 11/28/2012)
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 01 Dec 2012 20:56:45 -0000

	I have this draft in my toread list. Sorry about the delay.

	I have read and I support to adopt it as wg item.

.as

On 14/11/2012 20:35, Christopher Morrow wrote:
> Grow Folks,
> the authors of:  draft-mcpherson-irr-routing-policy-considerations
>   <http://tools.ietf.org/html/draft-mcpherson-irr-routing-policy-considerations-01>
> 
> have asked if their work is appropriate for the GROW-wg, could we take
> the next 14 days (ending Nov 28, 2012) to think/discuss this draft and
> it's efficacy to work in the GROW arena? The abstract of the draft is:
>   "The purpose of this document is to catalog past issues influencing
>    the efficacy of Internet Routing Registries (IRR) for inter-domain
>    routing policy specification and application in the global routing
>    system over the past two decades.  Additionally, it provides a
>    discussion regarding which of these issues are still problematic in
>    practice, and which are simply artifacts that are no longer
>    applicable but continue to stifle inter-provider policy-based
>    filtering adoption and IRR utility to this day."
> 
> Thanks!
> -Chris
> (co-chair)
> _______________________________________________
> GROW mailing list
> GROW@ietf.org
> https://www.ietf.org/mailman/listinfo/grow
> 

From christopher.morrow@gmail.com  Mon Dec  3 19:54:25 2012
Return-Path: <christopher.morrow@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 28A441F0C59 for <grow@ietfa.amsl.com>; Mon,  3 Dec 2012 19:54:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.299
X-Spam-Level: 
X-Spam-Status: No, score=-103.299 tagged_above=-999 required=5 tests=[AWL=0.300, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qjFsq4YBG3n3 for <grow@ietfa.amsl.com>; Mon,  3 Dec 2012 19:54:24 -0800 (PST)
Received: from mail-ea0-f172.google.com (mail-ea0-f172.google.com [209.85.215.172]) by ietfa.amsl.com (Postfix) with ESMTP id 6049821F88E8 for <grow@ietf.org>; Mon,  3 Dec 2012 19:54:24 -0800 (PST)
Received: by mail-ea0-f172.google.com with SMTP id a1so1575621eaa.31 for <grow@ietf.org>; Mon, 03 Dec 2012 19:54:23 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; bh=+Ex+pD9D9EFugC2w5K9k7V2uviqQvXSW+JI17BNEi7E=; b=YIx1LDE3sZ2oitSLuz9jytEH/RUTIMcviJY9t4X1TcfrnalWZWCHV9tIiBU/0AlZbO m8LQTj85X0M6GioaqbhRu9gPDx3RmtI1tzUOuuUPJNxISnkUTgjcmRr1uI41dqrWzw0O r7edSzuINnnUCiOkNOblOh2pvhj4+w/Wi3Ch/0qtOh7nqwXOKIiRlotfCovyAFJ2YmgS /c6WYSd87jVS6Nchq0mVWsxLO6Bm29IULhOJQN7Itp1BMpPrUDqWZ5o12I8ci+fEVjsm GG5DdoP9ORM18D4JUbswXbbEIwFBzrX36/JPKeDLmPxQv/CNG5I7bzLjFs+7jvF+ku28 NFIg==
MIME-Version: 1.0
Received: by 10.14.184.131 with SMTP id s3mr43560804eem.38.1354593263273; Mon, 03 Dec 2012 19:54:23 -0800 (PST)
Received: by 10.223.177.5 with HTTP; Mon, 3 Dec 2012 19:54:23 -0800 (PST)
In-Reply-To: <CAL9jLabt0S98HEygsQp=a98ARwiRxJ2x3F2=H5w-pDRbyqWE7A@mail.gmail.com>
References: <CAL9jLabt0S98HEygsQp=a98ARwiRxJ2x3F2=H5w-pDRbyqWE7A@mail.gmail.com>
Date: Mon, 3 Dec 2012 22:54:23 -0500
Message-ID: <CAL9jLaYVQWabTzRKqLZ5dKT4t7NqxgFM-zutgiXe+MJDcHTz5w@mail.gmail.com>
From: Christopher Morrow <christopher.morrow@gmail.com>
To: grow-ads@tools.ietf.org, grow-chairs@tools.ietf.org,  "grow@ietf.org grow@ietf.org" <grow@ietf.org>,  draft-mcpherson-irr-routing-policy-considerations@tools.ietf.org
Content-Type: text/plain; charset=ISO-8859-1
Subject: Re: [GROW] WG Adoption call: draft-mcpherson-irr-routing-policy-considerations (END: 11/28/2012)
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 04 Dec 2012 03:54:25 -0000

howdy... a few days late, but in the end there was some support for
adopting it, and at least no non-support for it.

Could the authors spin a re-titled draft and hopefully they are
incorporating comments/etc as well?

thanks!
-chris

On Wed, Nov 14, 2012 at 5:35 PM, Christopher Morrow
<christopher.morrow@gmail.com> wrote:
> Grow Folks,
> the authors of:  draft-mcpherson-irr-routing-policy-considerations
>   <http://tools.ietf.org/html/draft-mcpherson-irr-routing-policy-considerations-01>
>
> have asked if their work is appropriate for the GROW-wg, could we take
> the next 14 days (ending Nov 28, 2012) to think/discuss this draft and
> it's efficacy to work in the GROW arena? The abstract of the draft is:
>   "The purpose of this document is to catalog past issues influencing
>    the efficacy of Internet Routing Registries (IRR) for inter-domain
>    routing policy specification and application in the global routing
>    system over the past two decades.  Additionally, it provides a
>    discussion regarding which of these issues are still problematic in
>    practice, and which are simply artifacts that are no longer
>    applicable but continue to stifle inter-provider policy-based
>    filtering adoption and IRR utility to this day."
>
> Thanks!
> -Chris
> (co-chair)

From christopher.morrow@gmail.com  Mon Dec  3 20:04:15 2012
Return-Path: <christopher.morrow@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 04C0D1F0C67 for <grow@ietfa.amsl.com>; Mon,  3 Dec 2012 20:04:15 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.449
X-Spam-Level: 
X-Spam-Status: No, score=-103.449 tagged_above=-999 required=5 tests=[AWL=0.150, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id lemTvpQtLGZM for <grow@ietfa.amsl.com>; Mon,  3 Dec 2012 20:04:14 -0800 (PST)
Received: from mail-ea0-f172.google.com (mail-ea0-f172.google.com [209.85.215.172]) by ietfa.amsl.com (Postfix) with ESMTP id 0A87D1F0C61 for <grow@ietf.org>; Mon,  3 Dec 2012 20:04:13 -0800 (PST)
Received: by mail-ea0-f172.google.com with SMTP id a1so1577701eaa.31 for <grow@ietf.org>; Mon, 03 Dec 2012 20:04:13 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; bh=G2WsfMujoRWmU0JFsf7eQe8fc2r8KSdi4gzCTBEWFaE=; b=B28ZD+IcmK1vjRwfH4qUYTs4Vu5hBIBou0e4pwRCf0KmXABjudohTTWccDEasSRFDI 78zRk8e1lbDqtdBX4deXgmA+f9M0xNBhdBVx/sx88lixOwnzh6KkamvPGH4B5Ne1+Q+o TLVOv5b0p2GBwuy76Z3vX0WEE3KxDcwqLAxg1l2zJJIXxgXGb8mQvvjcNOtcSxxHTC9n 4Ey7873QcNqMUO+vxlPmTFydvq9LG5WBJSehcOcisYZJaUMjSRZOnRB+3fkQbNwjkAbm Qs814pfi/O1bSpEaP5lI0OqmeXWYwb4QeTpPzFgpePg72k51hAbRbQkPba+utShBGIhW NBeg==
MIME-Version: 1.0
Received: by 10.14.184.131 with SMTP id s3mr43640688eem.38.1354593853178; Mon, 03 Dec 2012 20:04:13 -0800 (PST)
Received: by 10.223.177.5 with HTTP; Mon, 3 Dec 2012 20:04:13 -0800 (PST)
In-Reply-To: <CAL9jLaagY5bBe_8eRST8noEp2-o00Zkm2Qc59Ys7pak++EPGNw@mail.gmail.com>
References: <CAL9jLaagY5bBe_8eRST8noEp2-o00Zkm2Qc59Ys7pak++EPGNw@mail.gmail.com>
Date: Mon, 3 Dec 2012 23:04:13 -0500
Message-ID: <CAL9jLaYjOLUJnuxsdysr7WNnJ+GCvA_e2jXuifgBbEV1g_Ldhw@mail.gmail.com>
From: Christopher Morrow <christopher.morrow@gmail.com>
To: grow-chairs@tools.ietf.org, grow-ads@tools.ietf.org,  "grow@ietf.org grow@ietf.org" <grow@ietf.org>,  draft-foo-sidr-simple-leak-attack-bgpsec-no-help@tools.ietf.org
Content-Type: text/plain; charset=ISO-8859-1
Subject: Re: [GROW] RouteLeaks - problem or not?
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 04 Dec 2012 04:04:15 -0000

ok, so after some considerable discussion (and correction of my
non-optimally-phrased questions) it seems there's some energy in
discussing this in GROW...

It seems that the draft: draft-foo-sidr-simple-leak-attack-bgpsec-no-help

looks like a good starting point for this discussion, could we re-spin
this as a GROW draft (re-title and submit) and perhaps send along
updates according to the comments received (if any?).

Once that appears it'd be grand if the list folks could discuss it a
bit more so we can see where the discussion leads.

-chris

On Wed, Nov 14, 2012 at 5:18 PM, Christopher Morrow
<christopher.morrow@gmail.com> wrote:
> GROW Folks,
> The SIDR working group is working on security for origination and path
> data related to BGP routes. There has been a note (a few) about SIDR's
> effect(s) or not on 'route leaks'. There have even been a few notes on
> 'what is a route leak'. To date there is a draft which discusses route
> leaks:
>   <http://tools.ietf.org/html/draft-foo-sidr-simple-leak-attack-bgpsec-no-help-02>
>
> where the authors have attempted to describe one (or many possible)
> situations which are called 'route leaks'. They also attempt to
> outline security issues which are follow-on effects of the situation
> described.
>
> SIDR attempted to look at route-leaks and came up a bit stymied, they
> asked IDR for some assistance with the issue, IDR pushed back to GROW
> to decide:
>   1) What is a 'route leak' (perhaps the above draft identifies one
> examplar to be used in that definition)
>   2) Are 'route leaks' a problem that Operations folks care about
>   3) Should IDR (or the IETF proper) address 'route leaks' with some
> form(s) of fix action.
>
> The end result of the above 3 steps is to push back into IDR one of
> two action requests:
>   1) "Yes, route leaks are a problem, please fix them."
>      or
>   2) "No, route leaks are not a problem, take no action."
>
> If #1 above is the answer, and IDR decides that changes to the BGP
> protocol are warranted (or are a possible solution to the problem)
> then SIDR has agreed to do what they can to 'secure' the bits
> added/changed/used in that endeavor.
>
> Could we have some discussion on-list about this problem, and some
> discussion about whether or not the draft referenced above fits the
> definition we would like to use for 'route leak'? I would also like
> the authors of the draft to decide where they would like to take their
> draft:
>    1) SIDR
>    2) IDR
>    3) GROW
>    4) other
>
> Thanks!
> -Chris
> (co-chair 1:2 of grow, and 1:3 in sidr)

From arturo.servin@gmail.com  Tue Dec  4 05:35:21 2012
Return-Path: <arturo.servin@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id ECF7421F86EE for <grow@ietfa.amsl.com>; Tue,  4 Dec 2012 05:35:21 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.599
X-Spam-Level: 
X-Spam-Status: No, score=-3.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4Qw3Af+0Ic6Y for <grow@ietfa.amsl.com>; Tue,  4 Dec 2012 05:35:21 -0800 (PST)
Received: from mail-gh0-f172.google.com (mail-gh0-f172.google.com [209.85.160.172]) by ietfa.amsl.com (Postfix) with ESMTP id F0D7921F8629 for <grow@ietf.org>; Tue,  4 Dec 2012 05:35:20 -0800 (PST)
Received: by mail-gh0-f172.google.com with SMTP id z22so615423ghb.31 for <grow@ietf.org>; Tue, 04 Dec 2012 05:35:18 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=message-id:date:from:user-agent:mime-version:to:cc:subject :references:in-reply-to:content-type:content-transfer-encoding; bh=SjNM/MUtE5V7OsLa06JEVZFeVB9r0fqlv6z5Cub7DPw=; b=Ebbn6Ka/UyB338tGSlyq6fdoeh7pAHkJruS3euVII7U2n5qlGjRe2woS+KeF748273 qFKZo0V3Afp+pPkJxufeKdy6nJOuthX/pDZZo7VirCCeUKPg0uIerIIGi4JF4vtW0CTX KeRqkAnoewUAc0WasSSVBRFUF36KlaXC7tLQXHh8sMPcISGkzjAAqN50AsvTnCWEqABc x7D/fC5elIQh8Ma3AUbMgA+S0pV9sqvy8lNbVl6ONIFoy4mWA1eBeBHP+hTiax5k/yv0 5X+OFKRHRJTEyIpUAtpAfPL83lVUP12reBbIvdGP4MSJjhyEooG7k3sDjbIWqmuvtDZM mvIw==
Received: by 10.236.128.141 with SMTP id f13mr14917044yhi.52.1354628118349; Tue, 04 Dec 2012 05:35:18 -0800 (PST)
Received: from 85-7-200.lacnic.net.uy ([200.7.85.160]) by mx.google.com with ESMTPS id v4sm809512ank.9.2012.12.04.05.35.15 (version=SSLv3 cipher=OTHER); Tue, 04 Dec 2012 05:35:17 -0800 (PST)
Message-ID: <50BDFC11.4060506@gmail.com>
Date: Tue, 04 Dec 2012 11:35:13 -0200
From: Arturo Servin <arturo.servin@gmail.com>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:17.0) Gecko/17.0 Thunderbird/17.0
MIME-Version: 1.0
To: Christopher Morrow <christopher.morrow@gmail.com>
References: <CAL9jLaagY5bBe_8eRST8noEp2-o00Zkm2Qc59Ys7pak++EPGNw@mail.gmail.com> <CAL9jLaYjOLUJnuxsdysr7WNnJ+GCvA_e2jXuifgBbEV1g_Ldhw@mail.gmail.com>
In-Reply-To: <CAL9jLaYjOLUJnuxsdysr7WNnJ+GCvA_e2jXuifgBbEV1g_Ldhw@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Cc: "grow@ietf.org grow@ietf.org" <grow@ietf.org>, draft-foo-sidr-simple-leak-attack-bgpsec-no-help@tools.ietf.org, grow-ads@tools.ietf.org, grow-chairs@tools.ietf.org
Subject: Re: [GROW] RouteLeaks - problem or not?
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 04 Dec 2012 13:35:22 -0000

	draft-foo-sidr-simple-leak-attack-bgpsec-no-help seems as a good start.

	I would suggest to make it more about leaks in general and not just
about security attacks (considering that many of the incidents with
leaks are mistakes and no targeted attacks).

my 20 cents,
as

	

On 04/12/2012 02:04, Christopher Morrow wrote:
> ok, so after some considerable discussion (and correction of my
> non-optimally-phrased questions) it seems there's some energy in
> discussing this in GROW...
> 
> It seems that the draft: draft-foo-sidr-simple-leak-attack-bgpsec-no-help
> 
> looks like a good starting point for this discussion, could we re-spin
> this as a GROW draft (re-title and submit) and perhaps send along
> updates according to the comments received (if any?).
> 
> Once that appears it'd be grand if the list folks could discuss it a
> bit more so we can see where the discussion leads.
> 
> -chris
> 
> On Wed, Nov 14, 2012 at 5:18 PM, Christopher Morrow
> <christopher.morrow@gmail.com> wrote:
>> GROW Folks,
>> The SIDR working group is working on security for origination and path
>> data related to BGP routes. There has been a note (a few) about SIDR's
>> effect(s) or not on 'route leaks'. There have even been a few notes on
>> 'what is a route leak'. To date there is a draft which discusses route
>> leaks:
>>   <http://tools.ietf.org/html/draft-foo-sidr-simple-leak-attack-bgpsec-no-help-02>
>>
>> where the authors have attempted to describe one (or many possible)
>> situations which are called 'route leaks'. They also attempt to
>> outline security issues which are follow-on effects of the situation
>> described.
>>
>> SIDR attempted to look at route-leaks and came up a bit stymied, they
>> asked IDR for some assistance with the issue, IDR pushed back to GROW
>> to decide:
>>   1) What is a 'route leak' (perhaps the above draft identifies one
>> examplar to be used in that definition)
>>   2) Are 'route leaks' a problem that Operations folks care about
>>   3) Should IDR (or the IETF proper) address 'route leaks' with some
>> form(s) of fix action.
>>
>> The end result of the above 3 steps is to push back into IDR one of
>> two action requests:
>>   1) "Yes, route leaks are a problem, please fix them."
>>      or
>>   2) "No, route leaks are not a problem, take no action."
>>
>> If #1 above is the answer, and IDR decides that changes to the BGP
>> protocol are warranted (or are a possible solution to the problem)
>> then SIDR has agreed to do what they can to 'secure' the bits
>> added/changed/used in that endeavor.
>>
>> Could we have some discussion on-list about this problem, and some
>> discussion about whether or not the draft referenced above fits the
>> definition we would like to use for 'route leak'? I would also like
>> the authors of the draft to decide where they would like to take their
>> draft:
>>    1) SIDR
>>    2) IDR
>>    3) GROW
>>    4) other
>>
>> Thanks!
>> -Chris
>> (co-chair 1:2 of grow, and 1:3 in sidr)
> _______________________________________________
> GROW mailing list
> GROW@ietf.org
> https://www.ietf.org/mailman/listinfo/grow
> 

From christopher.morrow@gmail.com  Tue Dec  4 06:34:44 2012
Return-Path: <christopher.morrow@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3F1D221F8AC1 for <grow@ietfa.amsl.com>; Tue,  4 Dec 2012 06:34:44 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.499
X-Spam-Level: 
X-Spam-Status: No, score=-103.499 tagged_above=-999 required=5 tests=[AWL=0.100, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QdoxTa8tGxP2 for <grow@ietfa.amsl.com>; Tue,  4 Dec 2012 06:34:43 -0800 (PST)
Received: from mail-ea0-f172.google.com (mail-ea0-f172.google.com [209.85.215.172]) by ietfa.amsl.com (Postfix) with ESMTP id 69B8D21F89DD for <grow@ietf.org>; Tue,  4 Dec 2012 06:34:43 -0800 (PST)
Received: by mail-ea0-f172.google.com with SMTP id a1so1806622eaa.31 for <grow@ietf.org>; Tue, 04 Dec 2012 06:34:42 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=lyWoK0W9w7o+31ly+dYMh/jCBpfQjSg+sX+XynegUno=; b=DeF8/6SKoOznFHKZ9xb2BbQy4F/FDvBaW3v+6RRGATmwZ4KJxmnHgADG/ARMJampLS D/FkbMA8KjEGkzRNEJtzcPGOs54eWVhdC8UKBb+UW8WnSMGhD66q5hiThnO1qdGzfiyq nA1t+JhELWhqqkWPGaDbzehvPozpd8MGICTnoVdDaXjM7oagoiuN1a1/TFu1lR4tPcUp rERVXHJQfsHaz4PGhkBfDOkpdS4OoE33bX1OQjHWF3olNarhHLAMkbf0p1DDlDg1i2fl JwTMDlOK90YtwFODJWZxW6eoS5bm4Nzw2yLmfdwLord32y+4ooH2R7QwqdqPVNjyUndJ 6z5A==
MIME-Version: 1.0
Received: by 10.14.202.3 with SMTP id c3mr18657921eeo.4.1354631682508; Tue, 04 Dec 2012 06:34:42 -0800 (PST)
Received: by 10.223.177.5 with HTTP; Tue, 4 Dec 2012 06:34:42 -0800 (PST)
In-Reply-To: <50BDFC11.4060506@gmail.com>
References: <CAL9jLaagY5bBe_8eRST8noEp2-o00Zkm2Qc59Ys7pak++EPGNw@mail.gmail.com> <CAL9jLaYjOLUJnuxsdysr7WNnJ+GCvA_e2jXuifgBbEV1g_Ldhw@mail.gmail.com> <50BDFC11.4060506@gmail.com>
Date: Tue, 4 Dec 2012 09:34:42 -0500
Message-ID: <CAL9jLaYzpk+fqigtGfVFvvm2pepAG3zy60UAxC4_Fpv8X-8PgA@mail.gmail.com>
From: Christopher Morrow <christopher.morrow@gmail.com>
To: Arturo Servin <arturo.servin@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Cc: "grow@ietf.org grow@ietf.org" <grow@ietf.org>, draft-foo-sidr-simple-leak-attack-bgpsec-no-help@tools.ietf.org, grow-ads@tools.ietf.org, grow-chairs@tools.ietf.org
Subject: Re: [GROW] RouteLeaks - problem or not?
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 04 Dec 2012 14:34:44 -0000

On Tue, Dec 4, 2012 at 8:35 AM, Arturo Servin <arturo.servin@gmail.com> wrote:
>
>
>         draft-foo-sidr-simple-leak-attack-bgpsec-no-help seems as a good start.
>
>         I would suggest to make it more about leaks in general and not just
> about security attacks (considering that many of the incidents with
> leaks are mistakes and no targeted attacks).
>

that was (one) of my comments, yes. (to the authors)

> my 20 cents,
> as
>
>
>
> On 04/12/2012 02:04, Christopher Morrow wrote:
>> ok, so after some considerable discussion (and correction of my
>> non-optimally-phrased questions) it seems there's some energy in
>> discussing this in GROW...
>>
>> It seems that the draft: draft-foo-sidr-simple-leak-attack-bgpsec-no-help
>>
>> looks like a good starting point for this discussion, could we re-spin
>> this as a GROW draft (re-title and submit) and perhaps send along
>> updates according to the comments received (if any?).
>>
>> Once that appears it'd be grand if the list folks could discuss it a
>> bit more so we can see where the discussion leads.
>>
>> -chris
>>
>> On Wed, Nov 14, 2012 at 5:18 PM, Christopher Morrow
>> <christopher.morrow@gmail.com> wrote:
>>> GROW Folks,
>>> The SIDR working group is working on security for origination and path
>>> data related to BGP routes. There has been a note (a few) about SIDR's
>>> effect(s) or not on 'route leaks'. There have even been a few notes on
>>> 'what is a route leak'. To date there is a draft which discusses route
>>> leaks:
>>>   <http://tools.ietf.org/html/draft-foo-sidr-simple-leak-attack-bgpsec-no-help-02>
>>>
>>> where the authors have attempted to describe one (or many possible)
>>> situations which are called 'route leaks'. They also attempt to
>>> outline security issues which are follow-on effects of the situation
>>> described.
>>>
>>> SIDR attempted to look at route-leaks and came up a bit stymied, they
>>> asked IDR for some assistance with the issue, IDR pushed back to GROW
>>> to decide:
>>>   1) What is a 'route leak' (perhaps the above draft identifies one
>>> examplar to be used in that definition)
>>>   2) Are 'route leaks' a problem that Operations folks care about
>>>   3) Should IDR (or the IETF proper) address 'route leaks' with some
>>> form(s) of fix action.
>>>
>>> The end result of the above 3 steps is to push back into IDR one of
>>> two action requests:
>>>   1) "Yes, route leaks are a problem, please fix them."
>>>      or
>>>   2) "No, route leaks are not a problem, take no action."
>>>
>>> If #1 above is the answer, and IDR decides that changes to the BGP
>>> protocol are warranted (or are a possible solution to the problem)
>>> then SIDR has agreed to do what they can to 'secure' the bits
>>> added/changed/used in that endeavor.
>>>
>>> Could we have some discussion on-list about this problem, and some
>>> discussion about whether or not the draft referenced above fits the
>>> definition we would like to use for 'route leak'? I would also like
>>> the authors of the draft to decide where they would like to take their
>>> draft:
>>>    1) SIDR
>>>    2) IDR
>>>    3) GROW
>>>    4) other
>>>
>>> Thanks!
>>> -Chris
>>> (co-chair 1:2 of grow, and 1:3 in sidr)
>> _______________________________________________
>> GROW mailing list
>> GROW@ietf.org
>> https://www.ietf.org/mailman/listinfo/grow
>>

From Donald.Smith@CenturyLink.com  Tue Dec  4 09:20:05 2012
Return-Path: <Donald.Smith@CenturyLink.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D337321F8C70 for <grow@ietfa.amsl.com>; Tue,  4 Dec 2012 09:20:04 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id MX4EtthxHNti for <grow@ietfa.amsl.com>; Tue,  4 Dec 2012 09:20:04 -0800 (PST)
Received: from sudnp799.qwest.com (sudnp799.qwest.com [155.70.32.99]) by ietfa.amsl.com (Postfix) with ESMTP id 0C83F21F8C6D for <grow@ietf.org>; Tue,  4 Dec 2012 09:20:03 -0800 (PST)
Received: from lxomavmpc030.qintra.com (lxomavmpc030.qintra.com [151.117.207.30]) by sudnp799.qwest.com (8.14.4/8.14.4) with ESMTP id qB4HK08X025245 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Tue, 4 Dec 2012 10:20:02 -0700 (MST)
Received: from lxomavmpc030.qintra.com (unknown [127.0.0.1]) by IMSA (Postfix) with ESMTP id 7333E1E006C; Tue,  4 Dec 2012 11:19:55 -0600 (CST)
Received: from sudnp796.qintra.com (unknown [10.6.10.61]) by lxomavmpc030.qintra.com (Postfix) with ESMTP id 30B851E0049; Tue,  4 Dec 2012 11:19:55 -0600 (CST)
Received: from sudnp796.qintra.com (localhost [127.0.0.1]) by sudnp796.qintra.com (8.14.4/8.14.4) with ESMTP id qB4HJsrW026730; Tue, 4 Dec 2012 10:19:54 -0700 (MST)
Received: from vddcwhubex502.ctl.intranet (vddcwhubex502.qintra.com [151.119.128.29]) by sudnp796.qintra.com (8.14.4/8.14.4) with ESMTP id qB4HJrYj026690 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Tue, 4 Dec 2012 10:19:54 -0700 (MST)
Received: from PDDCWMBXEX503.ctl.intranet ([fe80::9033:ef22:df02:32a9]) by vddcwhubex502.ctl.intranet ([2002:9777:801d::9777:801d]) with mapi id 14.02.0318.001; Tue, 4 Dec 2012 10:19:53 -0700
From: "Smith, Donald" <Donald.Smith@CenturyLink.com>
To: "'Christopher Morrow'" <christopher.morrow@gmail.com>, "'Arturo Servin'" <arturo.servin@gmail.com>
Thread-Topic: [GROW] RouteLeaks - problem or not?
Thread-Index: AQHN0dRxU3f5iK0xX0SNZFicJoYHjJgJGg+AgAAQnwD//7V00A==
Date: Tue, 4 Dec 2012 17:19:52 +0000
Message-ID: <68EFACB32CF4464298EA2779B058889D0A27331C@PDDCWMBXEX503.ctl.intranet>
References: <CAL9jLaagY5bBe_8eRST8noEp2-o00Zkm2Qc59Ys7pak++EPGNw@mail.gmail.com> <CAL9jLaYjOLUJnuxsdysr7WNnJ+GCvA_e2jXuifgBbEV1g_Ldhw@mail.gmail.com> <50BDFC11.4060506@gmail.com> <CAL9jLaYzpk+fqigtGfVFvvm2pepAG3zy60UAxC4_Fpv8X-8PgA@mail.gmail.com>
In-Reply-To: <CAL9jLaYzpk+fqigtGfVFvvm2pepAG3zy60UAxC4_Fpv8X-8PgA@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [151.119.128.7]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "'grow@ietf.org grow@ietf.org'" <grow@ietf.org>, "'grow-ads@tools.ietf.org'" <grow-ads@tools.ietf.org>, "'draft-foo-sidr-simple-leak-attack-bgpsec-no-help@tools.ietf.org'" <draft-foo-sidr-simple-leak-attack-bgpsec-no-help@tools.ietf.org>, "'grow-chairs@tools.ietf.org'" <grow-chairs@tools.ietf.org>
Subject: Re: [GROW] RouteLeaks - problem or not?
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 04 Dec 2012 17:20:05 -0000

"Pampers use multiple layers of protection to prevent leakage. Rommel used =
defense in depth to defend European fortresses." (A.White) Donald.Smith@Cen=
turyLink.com


>-----Original Message-----
>From: grow-bounces@ietf.org [mailto:grow-bounces@ietf.org] On Behalf Of
>Christopher Morrow
>Sent: Tuesday, December 04, 2012 7:35 AM
>To: Arturo Servin
>Cc: grow@ietf.org grow@ietf.org; draft-foo-sidr-simple-leak-attack-
>bgpsec-no-help@tools.ietf.org; grow-ads@tools.ietf.org; grow-
>chairs@tools.ietf.org
>Subject: Re: [GROW] RouteLeaks - problem or not?
>
>On Tue, Dec 4, 2012 at 8:35 AM, Arturo Servin <arturo.servin@gmail.com>
>wrote:
>>
>>
>>         draft-foo-sidr-simple-leak-attack-bgpsec-no-help seems as a
>good start.
>>
>>         I would suggest to make it more about leaks in general and not
>just
>> about security attacks (considering that many of the incidents with
>> leaks are mistakes and no targeted attacks).

I would say that even accidental announcements affect the A (availability) =
in CIA so it is security.
It may not be an attack and this might just be a nit.

Additionally I wanted to add my support as an operator for 1.
"Yes, route leaks are a problem, please fix them."

>>
>
>that was (one) of my comments, yes. (to the authors)
>
>> my 20 cents,
>> as
>>
>>
>>
>> On 04/12/2012 02:04, Christopher Morrow wrote:
>>> ok, so after some considerable discussion (and correction of my
>>> non-optimally-phrased questions) it seems there's some energy in
>>> discussing this in GROW...
>>>
>>> It seems that the draft: draft-foo-sidr-simple-leak-attack-bgpsec-no-
>help
>>>
>>> looks like a good starting point for this discussion, could we re-
>spin
>>> this as a GROW draft (re-title and submit) and perhaps send along
>>> updates according to the comments received (if any?).
>>>
>>> Once that appears it'd be grand if the list folks could discuss it a
>>> bit more so we can see where the discussion leads.
>>>
>>> -chris
>>>
>>> On Wed, Nov 14, 2012 at 5:18 PM, Christopher Morrow
>>> <christopher.morrow@gmail.com> wrote:
>>>> GROW Folks,
>>>> The SIDR working group is working on security for origination and
>path
>>>> data related to BGP routes. There has been a note (a few) about
>SIDR's
>>>> effect(s) or not on 'route leaks'. There have even been a few notes
>on
>>>> 'what is a route leak'. To date there is a draft which discusses
>route
>>>> leaks:
>>>>   <http://tools.ietf.org/html/draft-foo-sidr-simple-leak-attack-
>bgpsec-no-help-02>
>>>>
>>>> where the authors have attempted to describe one (or many possible)
>>>> situations which are called 'route leaks'. They also attempt to
>>>> outline security issues which are follow-on effects of the situation
>>>> described.
>>>>
>>>> SIDR attempted to look at route-leaks and came up a bit stymied,
>they
>>>> asked IDR for some assistance with the issue, IDR pushed back to
>GROW
>>>> to decide:
>>>>   1) What is a 'route leak' (perhaps the above draft identifies one
>>>> examplar to be used in that definition)
>>>>   2) Are 'route leaks' a problem that Operations folks care about
>>>>   3) Should IDR (or the IETF proper) address 'route leaks' with some
>>>> form(s) of fix action.
>>>>
>>>> The end result of the above 3 steps is to push back into IDR one of
>>>> two action requests:
>>>>   1) "Yes, route leaks are a problem, please fix them."
>>>>      or
>>>>   2) "No, route leaks are not a problem, take no action."
>>>>
>>>> If #1 above is the answer, and IDR decides that changes to the BGP
>>>> protocol are warranted (or are a possible solution to the problem)
>>>> then SIDR has agreed to do what they can to 'secure' the bits
>>>> added/changed/used in that endeavor.
>>>>
>>>> Could we have some discussion on-list about this problem, and some
>>>> discussion about whether or not the draft referenced above fits the
>>>> definition we would like to use for 'route leak'? I would also like
>>>> the authors of the draft to decide where they would like to take
>their
>>>> draft:
>>>>    1) SIDR
>>>>    2) IDR
>>>>    3) GROW
>>>>    4) other
>>>>
>>>> Thanks!
>>>> -Chris
>>>> (co-chair 1:2 of grow, and 1:3 in sidr)
>>> _______________________________________________
>>> GROW mailing list
>>> GROW@ietf.org
>>> https://www.ietf.org/mailman/listinfo/grow
>>>
>_______________________________________________
>GROW mailing list
>GROW@ietf.org
>https://www.ietf.org/mailman/listinfo/grow

From christopher.morrow@gmail.com  Fri Dec  7 09:52:55 2012
Return-Path: <christopher.morrow@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 231E321F86B9 for <grow@ietfa.amsl.com>; Fri,  7 Dec 2012 09:52:55 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.299
X-Spam-Level: 
X-Spam-Status: No, score=-103.299 tagged_above=-999 required=5 tests=[AWL=-0.300, BAYES_00=-2.599, J_CHICKENPOX_32=0.6, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eem6PJHh2-hN for <grow@ietfa.amsl.com>; Fri,  7 Dec 2012 09:52:54 -0800 (PST)
Received: from mail-ee0-f44.google.com (mail-ee0-f44.google.com [74.125.83.44]) by ietfa.amsl.com (Postfix) with ESMTP id F3DD321F8703 for <grow@ietf.org>; Fri,  7 Dec 2012 09:52:53 -0800 (PST)
Received: by mail-ee0-f44.google.com with SMTP id b47so490354eek.31 for <grow@ietf.org>; Fri, 07 Dec 2012 09:52:51 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=uGqsZoPqsUufIyHc9CJUX/to0xlUteFZNx1baC4Xef0=; b=tbwof6QL4DLDl6RKqlMGvNMqDEP3g0txyEBqkxfeJN2vb4+F7oh+bcaKZOtKfycrdJ JlEMOzVpFfH0p0YRl5jAOasenHSEWRyX80hCnEYG+LpyXP8NDnx6/su8bTwSoFJqZfn2 XAywuyd7nJatMrkBu/psRvD8+ECzCD/0sBFuLWKlXNjOkm9/EhdC75/NQE4PXWXDBd0S x7FaElPIa60RSWEkpi8YLL9B7AS4DNrnNwUi5ehVP87FUYmwqT1ThrHPrXYGKvRytB50 cVEq3ekiBN4StFdDeltn/vjfT06O4xvUr25v45j+JR40SelNXpgzrdchkgDAS6gs/qmR koCA==
MIME-Version: 1.0
Received: by 10.14.2.196 with SMTP id 44mr19093321eef.25.1354902771879; Fri, 07 Dec 2012 09:52:51 -0800 (PST)
Received: by 10.223.177.5 with HTTP; Fri, 7 Dec 2012 09:52:51 -0800 (PST)
In-Reply-To: <50B7162D.5060406@gmail.com>
References: <CA+b+ERk0A1Kkex-f2goRv5Wa+GOXENMgXqA52myHzxO23bJfvA@mail.gmail.com> <50B4EA44.7090502@merit.edu> <CA+b+ERm0DKs5Pzk5c0e2PJaQeMHvK1Y6Yqwvg0_G88ibjhoUTw@mail.gmail.com> <50B4EDB0.6040102@gmail.com> <CAL9jLaagOxmx6na-TTyuZB_ochQ5LOLOCr3RJz=CGFKvC_WRqQ@mail.gmail.com> <329D9664-957F-4682-85FC-E8028CC37EE1@puck.nether.net> <CAL9jLaYj+WOWVH92j-XOLXzoSKyTLVuc43rR2NdaPw_pG7k8qQ@mail.gmail.com> <50B5FD06.2060008@lacnic.net> <50B7162D.5060406@gmail.com>
Date: Fri, 7 Dec 2012 12:52:51 -0500
Message-ID: <CAL9jLab2mcoUCV0-CxdqMqibN6fx=v8=QU8ex=AMuRmkDMfkRA@mail.gmail.com>
From: Christopher Morrow <christopher.morrow@gmail.com>
To: Mattia Rossi <mattia.rossi.mailinglists@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Cc: grow@ietf.org
Subject: Re: [GROW] MRT
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 07 Dec 2012 17:52:55 -0000

On Thu, Nov 29, 2012 at 3:00 AM, Mattia Rossi
<mattia.rossi.mailinglists@gmail.com> wrote:
> Am 28.11.2012 13:01, schrieb Arturo Servin:
>
>> Chris,
>>
>>         The problem that I had is that it pybgpdum requires dpkt and it
>> did not
>> support 4 byte ASN and you need a hack.
>>
>>         Not sure if that is was problem as well. If it were, let me know
>> and I
>> will try to remember what I did to make it work but I think this is the
>> place with the answer:
>>
>> http://caia.swin.edu.au/urp/bgp/tools.html
>>
> Yes, the 4 Byte AS patch is part of the MDFMT. I would have hoped that by
> now somebody would have included it in dpkt, but it seems it didn't happen.
> The patch includes support for TABLE_DUMP_V2 as well, which is also not in
> dpkt by default.
> Dpkt still has an issue with certain RIPE MRT dumps, as some of the BGP4MP
> messages in there contain more entries than a Python String can handle, and
> thus dpkt will crash. (Dpkt will try to read the whole BGP4MP message in one
> go into a Python String).
>

ah, ok... so this patch set does work (for me in some brief testing),
did it get submitted to jon or was somethign waiting on that for a
reason not obvious to me?

> Cheers,
>
> Mat
>
> _______________________________________________
> GROW mailing list
> GROW@ietf.org
> https://www.ietf.org/mailman/listinfo/grow

From christopher.morrow@gmail.com  Fri Dec  7 09:53:19 2012
Return-Path: <christopher.morrow@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2281421F88C1 for <grow@ietfa.amsl.com>; Fri,  7 Dec 2012 09:53:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.224
X-Spam-Level: 
X-Spam-Status: No, score=-103.224 tagged_above=-999 required=5 tests=[AWL=-0.225, BAYES_00=-2.599, J_CHICKENPOX_32=0.6, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id VYLoDEb8LirE for <grow@ietfa.amsl.com>; Fri,  7 Dec 2012 09:53:18 -0800 (PST)
Received: from mail-ea0-f172.google.com (mail-ea0-f172.google.com [209.85.215.172]) by ietfa.amsl.com (Postfix) with ESMTP id 3D85C21F860F for <grow@ietf.org>; Fri,  7 Dec 2012 09:53:18 -0800 (PST)
Received: by mail-ea0-f172.google.com with SMTP id a1so301123eaa.31 for <grow@ietf.org>; Fri, 07 Dec 2012 09:53:17 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=MYLZSr2DrLNQ/H6YbVS1RM843ixnDPelPdHt5mtZEFo=; b=uoslsC3Yhf2iO4OfU1vEWSWBImAWM9GQMWLemSfPSk6/cJVYWbTFjmgcyKN2EdUCQa 4ig1jTlbc4BFpgPjourEQCQ/+2yDGl1fIKHTUStrVODeB13Sh8X4MlQBw617PTAXaM/w cEKnh05+Vt5sW9yiwIbbIuKGX5rdHX5WrQJQEiXo71Wfp9mBQQ7o5EMUqqBQyajIgKO2 kWI7S37adNF0VDOs0NAbY1ePoNi8/6nyd8hkRf55XLWPj/vlzMYKkJiREL2jUzkRThpg qyB3TxGeRrOJvUOwtP/W/bza4/nacdH5G35IDpFHOMIiQZNl2xCE+AglkcML0+6u1ZdX 5mDw==
MIME-Version: 1.0
Received: by 10.14.202.3 with SMTP id c3mr19333068eeo.4.1354902797408; Fri, 07 Dec 2012 09:53:17 -0800 (PST)
Received: by 10.223.177.5 with HTTP; Fri, 7 Dec 2012 09:53:17 -0800 (PST)
In-Reply-To: <CAL9jLab2mcoUCV0-CxdqMqibN6fx=v8=QU8ex=AMuRmkDMfkRA@mail.gmail.com>
References: <CA+b+ERk0A1Kkex-f2goRv5Wa+GOXENMgXqA52myHzxO23bJfvA@mail.gmail.com> <50B4EA44.7090502@merit.edu> <CA+b+ERm0DKs5Pzk5c0e2PJaQeMHvK1Y6Yqwvg0_G88ibjhoUTw@mail.gmail.com> <50B4EDB0.6040102@gmail.com> <CAL9jLaagOxmx6na-TTyuZB_ochQ5LOLOCr3RJz=CGFKvC_WRqQ@mail.gmail.com> <329D9664-957F-4682-85FC-E8028CC37EE1@puck.nether.net> <CAL9jLaYj+WOWVH92j-XOLXzoSKyTLVuc43rR2NdaPw_pG7k8qQ@mail.gmail.com> <50B5FD06.2060008@lacnic.net> <50B7162D.5060406@gmail.com> <CAL9jLab2mcoUCV0-CxdqMqibN6fx=v8=QU8ex=AMuRmkDMfkRA@mail.gmail.com>
Date: Fri, 7 Dec 2012 12:53:17 -0500
Message-ID: <CAL9jLaYd87yS7xO6puNp1nSOTJ3BiavZL+q7ZN7DAho+R5Dopw@mail.gmail.com>
From: Christopher Morrow <christopher.morrow@gmail.com>
To: Mattia Rossi <mattia.rossi.mailinglists@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Cc: grow@ietf.org
Subject: Re: [GROW] MRT
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 07 Dec 2012 17:53:19 -0000

On Fri, Dec 7, 2012 at 12:52 PM, Christopher Morrow
<christopher.morrow@gmail.com> wrote:
> On Thu, Nov 29, 2012 at 3:00 AM, Mattia Rossi
> <mattia.rossi.mailinglists@gmail.com> wrote:
>> Am 28.11.2012 13:01, schrieb Arturo Servin:
>>
>>> Chris,
>>>
>>>         The problem that I had is that it pybgpdum requires dpkt and it
>>> did not
>>> support 4 byte ASN and you need a hack.
>>>
>>>         Not sure if that is was problem as well. If it were, let me know
>>> and I
>>> will try to remember what I did to make it work but I think this is the
>>> place with the answer:
>>>
>>> http://caia.swin.edu.au/urp/bgp/tools.html
>>>
>> Yes, the 4 Byte AS patch is part of the MDFMT. I would have hoped that by
>> now somebody would have included it in dpkt, but it seems it didn't happen.
>> The patch includes support for TABLE_DUMP_V2 as well, which is also not in
>> dpkt by default.
>> Dpkt still has an issue with certain RIPE MRT dumps, as some of the BGP4MP
>> messages in there contain more entries than a Python String can handle, and
>> thus dpkt will crash. (Dpkt will try to read the whole BGP4MP message in one
>> go into a Python String).
>>
>
> ah, ok... so this patch set does work (for me in some brief testing),
> did it get submitted to jon or was somethign waiting on that for a
> reason not obvious to me?

oh, unittests requests, yes....

>
>> Cheers,
>>
>> Mat
>>
>> _______________________________________________
>> GROW mailing list
>> GROW@ietf.org
>> https://www.ietf.org/mailman/listinfo/grow

From danny@tcb.net  Fri Dec  7 15:35:48 2012
Return-Path: <danny@tcb.net>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7616921F8797 for <grow@ietfa.amsl.com>; Fri,  7 Dec 2012 15:35:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -100.437
X-Spam-Level: 
X-Spam-Status: No, score=-100.437 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FH_RELAY_NODNS=1.451, HELO_MISMATCH_ORG=0.611,  RDNS_NONE=0.1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id TeuiKR2nMggv for <grow@ietfa.amsl.com>; Fri,  7 Dec 2012 15:35:47 -0800 (PST)
Received: from mail.friendswithtools.org (unknown [64.78.239.70]) by ietfa.amsl.com (Postfix) with ESMTP id D2A2221F86E8 for <grow@ietf.org>; Fri,  7 Dec 2012 15:35:47 -0800 (PST)
Received: from dspam (unknown [127.0.0.1]) by mail.friendswithtools.org (Postfix) with SMTP id 8FA4F300092 for <grow@ietf.org>; Fri,  7 Dec 2012 23:35:47 +0000 (UTC)
Received: from dul1dmcphers-m2.home (pool-71-171-124-149.clppva.fios.verizon.net [71.171.124.149]) (using TLSv1 with cipher AES128-SHA (128/128 bits)) (No client certificate requested) by mail.friendswithtools.org (Postfix) with ESMTPSA id 20393300091; Fri,  7 Dec 2012 16:35:46 -0700 (MST)
From: Danny McPherson <danny@tcb.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable
Date: Fri, 7 Dec 2012 18:35:46 -0500
References: <FC2FA0E2-6920-43D4-AC5C-D6576C522A20@tcb.net>
To: "grow@ietf.org grow@ietf.org" <grow@ietf.org>, idr wg <idr@ietf.org>
Message-Id: <09E34042-DFEB-4360-9D7B-6C7986952B80@tcb.net>
Mime-Version: 1.0 (Apple Message framework v1283)
X-Mailer: Apple Mail (2.1283)
X-DSPAM-Result: Innocent
X-DSPAM-Processed: Fri Dec  7 16:35:47 2012
X-DSPAM-Confidence: 0.9898
X-DSPAM-Improbability: 1 in 9709 chance of being spam
X-DSPAM-Probability: 0.0000
X-DSPAM-Signature: 50c27d53199631760617152
X-DSPAM-Factors: 27, 2012+at, 0.01000, at+#+#+PM, 0.01000, the+#+of, 0.01000, the+#+of, 0.01000, in+the, 0.01000, in+the, 0.01000, 2012+#+#+#+PM, 0.01000, 2012+#+#+#+PM, 0.01000, Url*//tools, 0.01000, Url*http, 0.01000, Url*http, 0.01000, of+the, 0.01000, of+the, 0.01000, of+#+#+#+to, 0.01000, On+#+#+#+at, 0.01000, Url*ietf, 0.01000, Url*ietf, 0.01000, On+#+#+2012, 0.01000, 2012+#+#+38, 0.40000, playing+with, 0.40000, 38+#+Murphy, 0.40000, Subject+#+sidr, 0.40000, name+#+#+on, 0.40000, Time+was, 0.40000, Time+was, 0.40000, proposals+#+#+effectively, 0.40000, change+was, 0.40000
Subject: [GROW] Fwd: [sidr] about "beaconing" and the bgspec-protoocol
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 07 Dec 2012 23:35:48 -0000

[apologies in advance for cross-post]=20

An example of why some of the work in SIDR is making me uneasy, =
particularly given they're currently aiming for standards track at the =
moment..

I'll just note that this is *your* BGP system they're talking about..

-danny


Begin forwarded message:

> From: Danny McPherson <danny@tcb.net>
> Subject: Re: [sidr] about "beaconing" and the bgspec-protoocol
> Date: December 7, 2012 6:24:00 PM EST
> To: "Murphy, Sandra" <Sandra.Murphy@sparta.com>
> Cc: "sidr@ietf.org" <sidr@ietf.org>
>=20
>=20
> On Dec 7, 2012, at 2:38 PM, Murphy, Sandra wrote:
>=20
>> =46rom comments made at the mike in the last IETG sidr session after =
the discussion of key rollover techniques, I think there might be a bit =
of confusion about beaconing.
>>=20
>> An Expire Time was a feature of the bgpsec protocol in versions =
00-01.  The purpose of the Expire Time  was to prevent replay and ensure =
freshness.  The effect of this feature was to require periodic =
readvertisements of all prefixes, hence the name "beaconing".
>>=20
>> Based on wg discussions, "beaconing" was removed from the bgpsec =
protocol in versions 02 (Mar 12) forward.
>>=20
>> Protection against human time scale replay, e.g., from neighbor =
relationships that change, was suggested to be possible through the use =
of key rollover.
>=20
>=20
> I'm pretty sure the captured proceedings for "Discussion of Key =
Rollover Mechanisms for Replay-Attack Protection" [!=3Dbeacon?] mentions =
"beacon" at least eleven times.  Additionally, "BGPSEC router key =
rollover" was about periodic updates (aka "beacons") as well.
>=20
> Folks can name them periodic updates, or beacons, or even stretch and =
call them triggered updates if you want.  Adding new attributes to =
standards-track BGP that require a "refresh" function through *some* new =
attribute by the origin AND independently by each transit AS (and =
perhaps each BGPSEC-enabled BGP router therein) under the auspices of =
"key rollover" rather than "freshness" is, well, still pretty much the =
same thing functionally - they're still periodic updates in a soft state =
protocol that don't exist in BGP today.
>=20
> I still contend that doing anything to introduce periodic updates in =
the global inter-domain routing system is a terrible idea and is =
something we should avoid altogether.  Even [ripv1-rfc1058] provides =
some hints as to why this is a bad idea.
>=20
> I know we're focusing on hyper-deduced stuff here, but the =
combinatorial effects of orders of magnitude larger updates in BGP =
through BGPSEC proposals, breaking and effectively disabling the ability =
for BGP update packing anywhere, considerable added churn through =
origination and transit node-triggered "beacons", and added processing =
overhead for cryptographic functions for signing and validation, all in =
today's BGP, sure makes me real concern about our goals and if the =
reward outweighs the risk here.
>=20
> I highly recommend that these documents be experimental and people =
playing with this stuff in BGP do so in closed environments, not on =
routers attached to the global routing system; it's fragile enough as =
is.
>=20
>=20
> -danny
>=20
> [!=3Dbeacon?] =
http://www.ietf.org/proceedings/85/slides/slides-85-sidr-4.pdf
> [ripv1-1058] http://tools.ietf.org/html/rfc1058
>=20
> _______________________________________________
> sidr mailing list
> sidr@ietf.org
> https://www.ietf.org/mailman/listinfo/sidr
>=20



From shares@ndzh.com  Fri Dec 21 14:38:13 2012
Return-Path: <shares@ndzh.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9BF2821F8583 for <grow@ietfa.amsl.com>; Fri, 21 Dec 2012 14:38:13 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.139
X-Spam-Level: **
X-Spam-Status: No, score=2.139 tagged_above=-999 required=5 tests=[AWL=-0.226,  BAYES_20=-0.74, DOS_OUTLOOK_TO_MX=1, FH_RELAY_NODNS=1.451, HELO_MISMATCH_COM=0.553, HTML_MESSAGE=0.001, RDNS_NONE=0.1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id TkQjX8TY8IYA for <grow@ietfa.amsl.com>; Fri, 21 Dec 2012 14:38:12 -0800 (PST)
Received: from hickoryhill-consulting.com (unknown [64.9.205.143]) by ietfa.amsl.com (Postfix) with ESMTP id 2FF5021F88B0 for <grow@ietf.org>; Fri, 21 Dec 2012 14:38:11 -0800 (PST)
X-Default-Received-SPF: pass (skip=forwardok (res=PASS)) x-ip-name=64.112.195.202; 
From: "Susan Hares" <shares@ndzh.com>
To: <grow@ietf.org>
Date: Fri, 21 Dec 2012 17:37:53 -0500
Message-ID: <008801cddfcb$d11fcdd0$735f6970$@ndzh.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_NextPart_000_0089_01CDDFA1.E84B9A90"
X-Mailer: Microsoft Outlook 14.0
Thread-Index: Ac3fy8dOBkTw7JkWQmefZqEEiEfLDw==
Content-Language: en-us
X-Authenticated-User: skh@ndzh.com 
X-IsFriend: <shares@ndzh.com>
Cc: 'Susan Hares' <skh@ndzh.com>, "Jon Mitchell \(GNS\)" <Jon.Mitchell@microsoft.com>, "'John G. Scudder'" <jgs@bgp.nu>, stbryant@cisco.com
Subject: [GROW] Requesting Grow's review of draft-ietf-idr-as-private-reservation-02.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 21 Dec 2012 22:38:13 -0000

This is a multipart message in MIME format.

------=_NextPart_000_0089_01CDDFA1.E84B9A90
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit

Peter and Chris: 

 

We request that the GROW WG review
draft-ietf-idr-as-private-reservation-02.txt?
(http://datatracker.ietf.org/doc/draft-ietf-idr-as-private-reservation/)

 

Since it modifies RFC 1930, a BCP, it is likely to be a BCP.  The draft
deals with the creation of Private USE ASN space in 4 byte octets. 

 

Two additional questions we'd love input on:  

 

1.     Do they think the range is too big, too small, or just right"  (We'll
call this the 3 bears question (smile))

2.     Would they prefer to see the range reserved with some portion of the
range allocated? 

 

For example, reserve all the space but allocate only 50% of the space. 

 

This draft has had substantial IDR debate which is (of course) online, and
has been through IDR WG LC. 

 

We'd appreciate hearing about any operational issues GROW finds with this
draft. 

 

Thank you, 

 

John and Sue 

 


------=_NextPart_000_0089_01CDDFA1.E84B9A90
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;}
/* 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:1256402928;
	mso-list-type:hybrid;
	mso-list-template-ids:-2011262508 67698703 67698713 67698715 67698703 =
67698713 67698715 67698703 67698713 67698715;}
@list l0:level1
	{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;}
@list l1
	{mso-list-id:1518419333;
	mso-list-type:hybrid;
	mso-list-template-ids:1705380002 67698703 67698713 67698715 67698703 =
67698713 67698715 67698703 67698713 67698715;}
@list l1:level1
	{mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l1:level2
	{mso-level-number-format:alpha-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l1:level3
	{mso-level-number-format:roman-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:right;
	text-indent:-9.0pt;}
@list l1:level4
	{mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l1:level5
	{mso-level-number-format:alpha-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l1:level6
	{mso-level-number-format:roman-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:right;
	text-indent:-9.0pt;}
@list l1:level7
	{mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l1:level8
	{mso-level-number-format:alpha-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l1:level9
	{mso-level-number-format:roman-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:right;
	text-indent:-9.0pt;}
ol
	{margin-bottom:0in;}
ul
	{margin-bottom:0in;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue =
vlink=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Peter and =
Chris: <o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>We request =
that the GROW WG review =
&nbsp;draft-ietf-idr-as-private-reservation-02.txt? &nbsp;&nbsp;(<a =
href=3D"http://datatracker.ietf.org/doc/draft-ietf-idr-as-private-reserva=
tion/">http://datatracker.ietf.org/doc/draft-ietf-idr-as-private-reservat=
ion/</a>)<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Since it =
modifies RFC 1930, a BCP, it is likely to be a BCP. &nbsp;The draft =
deals with the creation of Private USE ASN space in 4 byte octets. =
<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Two =
additional questions we&#8217;d love input on: =
&nbsp;<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoListParagraph =
style=3D'text-indent:-.25in;mso-list:l0 level1 lfo2'><![if =
!supportLists]><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><span =
style=3D'mso-list:Ignore'>1.<span style=3D'font:7.0pt "Times New =
Roman"'>&nbsp;&nbsp;&nbsp;&nbsp; </span></span></span><![endif]><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Do they =
think the range is too big, too small, or just right&#8221;&nbsp; =
(We&#8217;ll call this the 3 bears question =
(smile))<o:p></o:p></span></p><p class=3DMsoListParagraph =
style=3D'text-indent:-.25in;mso-list:l0 level1 lfo2'><![if =
!supportLists]><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><span =
style=3D'mso-list:Ignore'>2.<span style=3D'font:7.0pt "Times New =
Roman"'>&nbsp;&nbsp;&nbsp;&nbsp; </span></span></span><![endif]><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Would they =
prefer to see the range reserved with some portion of the range =
allocated? <o:p></o:p></span></p><p class=3DMsoNormal =
style=3D'margin-left:.25in'><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal style=3D'margin-left:.5in'><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>For example, =
reserve all the space but allocate only 50% of the space. =
<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>This draft =
has had substantial IDR debate which is (of course) online, and has been =
through IDR WG LC. <o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>We&#8217;d =
appreciate hearing about any operational issues GROW finds with this =
draft. <o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Thank you, =
<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>John and Sue =
<o:p></o:p></span></p><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div></body></html>
------=_NextPart_000_0089_01CDDFA1.E84B9A90--


From warren@kumari.net  Fri Dec 21 15:22:34 2012
Return-Path: <warren@kumari.net>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1B93421F8A41 for <grow@ietfa.amsl.com>; Fri, 21 Dec 2012 15:22:34 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.599
X-Spam-Level: 
X-Spam-Status: No, score=-102.599 tagged_above=-999 required=5 tests=[AWL=0.000, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bl3A0ygzMkiO for <grow@ietfa.amsl.com>; Fri, 21 Dec 2012 15:22:33 -0800 (PST)
Received: from vimes.kumari.net (smtp1.kumari.net [204.194.22.1]) by ietfa.amsl.com (Postfix) with ESMTP id 96E0721F8A40 for <grow@ietf.org>; Fri, 21 Dec 2012 15:22:33 -0800 (PST)
Received: from [192.168.1.38] (unknown [66.84.81.126]) by vimes.kumari.net (Postfix) with ESMTPSA id 5516E1B403EA; Fri, 21 Dec 2012 18:22:32 -0500 (EST)
Content-Type: text/plain; charset=windows-1252
Mime-Version: 1.0 (Mac OS X Mail 6.2 \(1499\))
From: Warren Kumari <warren@kumari.net>
In-Reply-To: <008801cddfcb$d11fcdd0$735f6970$@ndzh.com>
Date: Fri, 21 Dec 2012 18:22:31 -0500
Content-Transfer-Encoding: quoted-printable
Message-Id: <B9BC66C4-3FE8-4F01-8D52-2E046F664985@kumari.net>
References: <008801cddfcb$d11fcdd0$735f6970$@ndzh.com>
To: "Susan Hares" <shares@ndzh.com>
X-Mailer: Apple Mail (2.1499)
Cc: 'Susan Hares' <skh@ndzh.com>, "Jon Mitchell \(GNS\)" <Jon.Mitchell@microsoft.com>, grow@ietf.org, "'John G. Scudder'" <jgs@bgp.nu>, Warren Kumari <warren@kumari.net>, stbryant@cisco.com
Subject: Re: [GROW] Requesting Grow's review of draft-ietf-idr-as-private-reservation-02.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 21 Dec 2012 23:22:34 -0000

On Dec 21, 2012, at 5:37 PM, "Susan Hares" <shares@ndzh.com> wrote:

> Peter and Chris:
> =20
> We request that the GROW WG review  =
draft-ietf-idr-as-private-reservation-02.txt?   =
(http://datatracker.ietf.org/doc/draft-ietf-idr-as-private-reservation/)
> =20
> Since it modifies RFC 1930, a BCP, it is likely to be a BCP.  The =
draft deals with the creation of Private USE ASN space in 4 byte octets.
> =20
> Two additional questions we=92d love input on: =20
> =20
> 1.     Do they think the range is too big, too small, or just right=94 =
 (We=92ll call this the 3 bears question (smile))
> 2.     Would they prefer to see the range reserved with some portion =
of the range allocated?
> =20
> For example, reserve all the space but allocate only 50% of the space.
> =20
> This draft has had substantial IDR debate which is (of course) online, =
and has been through IDR WG LC.
> =20
> We=92d appreciate hearing about any operational issues GROW finds with =
this draft.

As someone who reads both GROW and IDR I'd just like to more that there =
has already been some operators providing input over on the IDR list...
This is not intended to discourage GROWers from commenting, but rather a =
suggestion that before erupting in indignation (or giggles!) you have a =
quick look @ the IDR threads on this=85
They may:
a: address your concerns or
b: provide you with more ammunition for making your case :-P

W

> =20
> Thank you,
> =20
> John and Sue
> =20
> _______________________________________________
> GROW mailing list
> GROW@ietf.org
> https://www.ietf.org/mailman/listinfo/grow

--
For every complex problem, there is a solution that is simple, neat, and =
wrong.
                -- H. L. Mencken





From rraszuk@gmail.com  Fri Dec 21 15:34:18 2012
Return-Path: <rraszuk@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8A36D21F8891 for <grow@ietfa.amsl.com>; Fri, 21 Dec 2012 15:34:18 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.943
X-Spam-Level: 
X-Spam-Status: No, score=-2.943 tagged_above=-999 required=5 tests=[AWL=0.034,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WcSwRw-bVMPQ for <grow@ietfa.amsl.com>; Fri, 21 Dec 2012 15:34:18 -0800 (PST)
Received: from mail-ie0-f175.google.com (mail-ie0-f175.google.com [209.85.223.175]) by ietfa.amsl.com (Postfix) with ESMTP id 190B721F87DF for <grow@ietf.org>; Fri, 21 Dec 2012 15:34:18 -0800 (PST)
Received: by mail-ie0-f175.google.com with SMTP id qd14so6905972ieb.6 for <grow@ietf.org>; Fri, 21 Dec 2012 15:34:17 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=lnlifm2zaU7bAQCRzI/IQhi5HOfwVmUq6Zcz3U4CgE0=; b=r88fBcnWHseejiPA9s008kT+3iII7PQpmuZ4Ablt/O/hpXj2/wfSxGqZqYgqxjLpMj PofRgcjWGWpGiINdcAEV33IEVvVw7xsb5ePHty3bPeh0mydotwnq3vc6xFGD6Ln7HEwU jtQjH5lFO6MgR/IW9cDEkz1LQKbCUFidp4cBKBxD6Qk56f0AxBMzosWpI9ErCQ/QA8oQ 5QL83aKTw7efsNoOgsvIgY5bOJpca4jwQzTtEQtm/vXR5qSWRuzr/dCsuV/M1TuFBxdJ HfwKGiM7jjcqIsrXxfbEbGhOzzwxKi8jOVbWwYOmH1UK/k0FLu968TEYKw6chfCeVxwE 2Acw==
MIME-Version: 1.0
Received: by 10.50.202.73 with SMTP id kg9mr15171333igc.51.1356132857697; Fri, 21 Dec 2012 15:34:17 -0800 (PST)
Sender: rraszuk@gmail.com
Received: by 10.64.167.204 with HTTP; Fri, 21 Dec 2012 15:34:17 -0800 (PST)
In-Reply-To: <B9BC66C4-3FE8-4F01-8D52-2E046F664985@kumari.net>
References: <008801cddfcb$d11fcdd0$735f6970$@ndzh.com> <B9BC66C4-3FE8-4F01-8D52-2E046F664985@kumari.net>
Date: Sat, 22 Dec 2012 00:34:17 +0100
X-Google-Sender-Auth: j0iuZVlzxjG-mxwPlhG5Nr0KaCY
Message-ID: <CA+b+ERn5UxbWJECKu3GXL8wekiE3wDR_eJNkB05Vi456Q3L3xA@mail.gmail.com>
From: Robert Raszuk <robert@raszuk.net>
To: Warren Kumari <warren@kumari.net>
Content-Type: text/plain; charset=ISO-8859-1
Cc: Susan Hares <skh@ndzh.com>, "Jon Mitchell \(GNS\)" <Jon.Mitchell@microsoft.com>, grow@ietf.org, "John G. Scudder" <jgs@bgp.nu>, stbryant@cisco.com
Subject: Re: [GROW] Requesting Grow's review of draft-ietf-idr-as-private-reservation-02.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 21 Dec 2012 23:34:18 -0000

> As someone who reads both GROW and IDR

On this cross post ...

I really do not know who is subscribed to GROW and would not be
subscribed  to IDR ... Could corresponding chairs run a mailer diff
and at least check how many folks is not on both lists ?

But formally this is good. At least the IETF last call will go smooth(er) ;)

I think Sue forgot to set some deadline for comments - not sure if
everyone just implicitly assumes two weeks.

Best wishes,
R.

From shares@ndzh.com  Fri Dec 21 18:43:29 2012
Return-Path: <shares@ndzh.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3049C21F8781 for <grow@ietfa.amsl.com>; Fri, 21 Dec 2012 18:43:29 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 1.221
X-Spam-Level: *
X-Spam-Status: No, score=1.221 tagged_above=-999 required=5 tests=[AWL=0.715,  BAYES_00=-2.599, DOS_OUTLOOK_TO_MX=1, FH_RELAY_NODNS=1.451, HELO_MISMATCH_COM=0.553, HTML_MESSAGE=0.001, RDNS_NONE=0.1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id C+40LiFZMsM5 for <grow@ietfa.amsl.com>; Fri, 21 Dec 2012 18:43:28 -0800 (PST)
Received: from hickoryhill-consulting.com (unknown [64.9.205.143]) by ietfa.amsl.com (Postfix) with ESMTP id E6B3721F86FA for <grow@ietf.org>; Fri, 21 Dec 2012 18:43:27 -0800 (PST)
X-Default-Received-SPF: pass (skip=forwardok (res=PASS)) x-ip-name=64.112.195.202; 
From: "Susan Hares" <shares@ndzh.com>
To: <grow@ietf.org>
References: 
In-Reply-To: 
Date: Fri, 21 Dec 2012 21:43:05 -0500
Message-ID: <000c01cddfee$12706a20$37513e60$@ndzh.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_NextPart_000_000D_01CDDFC4.29A389E0"
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQKydZMDSpLGaBauxKW0zASzOek48JZa8L1Q
Content-Language: en-us
X-Authenticated-User: skh@ndzh.com 
X-IsFriend: <shares@ndzh.com>
Cc: 'Susan Hares' <skh@ndzh.com>, "Jon Mitchell \(GNS\)" <Jon.Mitchell@microsoft.com>, "'John G. Scudder'" <jgs@bgp.nu>, stbryant@cisco.com
Subject: Re: [GROW] Requesting Grow's review of draft-ietf-idr-as-private-reservation-02.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 22 Dec 2012 02:43:29 -0000

This is a multipart message in MIME format.

------=_NextPart_000_000D_01CDDFC4.29A389E0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit

(Reposting with time of 2 weeks) 

 

Peter and Chris: 

 

We request that the GROW WG review
draft-ietf-idr-as-private-reservation-02.txt?
(http://datatracker.ietf.org/doc/draft-ietf-idr-as-private-reservation/)

 

Since it modifies RFC 1930, a BCP, it is likely to be a BCP.  The draft
deals with the creation of Private USE ASN space in 4 byte octets. 

 

Two additional questions we'd love input on:  

 

1.     Do they think the range is too big, too small, or just right"  (We'll
call this the 3 bears question (smile))

2.     Would they prefer to see the range reserved with some portion of the
range allocated? 

 

For example, reserve all the space but allocate only 50% of the space. 

 

This draft has had substantial IDR debate which is (of course) online, and
has been through IDR WG LC. 

 

We'd appreciate hearing about any operational issues GROW finds with this
draft by 1/4/2013.  (2 weeks) 

 

Thank you, 

 

John and Sue 

 


------=_NextPart_000_000D_01CDDFC4.29A389E0
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;}
/* 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.EmailStyle18
	{mso-style-type:personal;
	font-family:"Calibri","sans-serif";
	color:windowtext;}
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:1256402928;
	mso-list-type:hybrid;
	mso-list-template-ids:-2011262508 67698703 67698713 67698715 67698703 =
67698713 67698715 67698703 67698713 67698715;}
@list l0:level1
	{mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level2
	{mso-level-number-format:alpha-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level3
	{mso-level-number-format:roman-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:right;
	text-indent:-9.0pt;}
@list l0:level4
	{mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level5
	{mso-level-number-format:alpha-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level6
	{mso-level-number-format:roman-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:right;
	text-indent:-9.0pt;}
@list l0:level7
	{mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level8
	{mso-level-number-format:alpha-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:left;
	text-indent:-.25in;}
@list l0:level9
	{mso-level-number-format:roman-lower;
	mso-level-tab-stop:none;
	mso-level-number-position:right;
	text-indent:-9.0pt;}
ol
	{margin-bottom:0in;}
ul
	{margin-bottom:0in;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue =
vlink=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><b><span =
style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif";color:#1F497D=
'>(Reposting with time of 2 weeks) <o:p></o:p></span></b></p><p =
class=3DMsoNormal><b><span =
style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif";color:#1F497D=
'><o:p>&nbsp;</o:p></span></b></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Peter and =
Chris: <o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>We request =
that the GROW WG review =
&nbsp;draft-ietf-idr-as-private-reservation-02.txt? &nbsp;&nbsp;(<a =
href=3D"http://datatracker.ietf.org/doc/draft-ietf-idr-as-private-reserva=
tion/">http://datatracker.ietf.org/doc/draft-ietf-idr-as-private-reservat=
ion/</a>)<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Since it =
modifies RFC 1930, a BCP, it is likely to be a BCP. &nbsp;The draft =
deals with the creation of Private USE ASN space in 4 byte octets. =
<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Two =
additional questions we&#8217;d love input on: =
&nbsp;<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoListParagraph =
style=3D'text-indent:-.25in;mso-list:l0 level1 lfo2'><![if =
!supportLists]><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><span =
style=3D'mso-list:Ignore'>1.<span style=3D'font:7.0pt "Times New =
Roman"'>&nbsp;&nbsp;&nbsp;&nbsp; </span></span></span><![endif]><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Do they =
think the range is too big, too small, or just right&#8221;&nbsp; =
(We&#8217;ll call this the 3 bears question =
(smile))<o:p></o:p></span></p><p class=3DMsoListParagraph =
style=3D'text-indent:-.25in;mso-list:l0 level1 lfo2'><![if =
!supportLists]><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><span =
style=3D'mso-list:Ignore'>2.<span style=3D'font:7.0pt "Times New =
Roman"'>&nbsp;&nbsp;&nbsp;&nbsp; </span></span></span><![endif]><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Would they =
prefer to see the range reserved with some portion of the range =
allocated? <o:p></o:p></span></p><p class=3DMsoNormal =
style=3D'margin-left:.25in'><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal style=3D'margin-left:.5in'><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>For example, =
reserve all the space but allocate only 50% of the space. =
<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>This draft =
has had substantial IDR debate which is (of course) online, and has been =
through IDR WG LC. <o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>We&#8217;d =
appreciate hearing about any operational issues GROW finds with this =
draft<span style=3D'color:#1F497D'> by 1/4/2013.&nbsp; (2 weeks) =
</span><o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>Thank you, =
<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'><o:p>&nbsp;</=
o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif"'>John and Sue =
<o:p></o:p></span></p><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div></body></html>
------=_NextPart_000_000D_01CDDFC4.29A389E0--


From shares@ndzh.com  Fri Dec 21 18:45:54 2012
Return-Path: <shares@ndzh.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DFC1F21F8929 for <grow@ietfa.amsl.com>; Fri, 21 Dec 2012 18:45:46 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 1.189
X-Spam-Level: *
X-Spam-Status: No, score=1.189 tagged_above=-999 required=5 tests=[AWL=0.684,  BAYES_00=-2.599, DOS_OUTLOOK_TO_MX=1, FH_RELAY_NODNS=1.451, HELO_MISMATCH_COM=0.553, RDNS_NONE=0.1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6AEkVd8AyFdn for <grow@ietfa.amsl.com>; Fri, 21 Dec 2012 18:45:45 -0800 (PST)
Received: from hickoryhill-consulting.com (unknown [64.9.205.143]) by ietfa.amsl.com (Postfix) with ESMTP id EF44921F8422 for <grow@ietf.org>; Fri, 21 Dec 2012 18:45:38 -0800 (PST)
X-Default-Received-SPF: pass (skip=forwardok (res=PASS)) x-ip-name=64.112.195.202; 
From: "Susan Hares" <shares@ndzh.com>
To: "'Warren Kumari'" <warren@kumari.net>
References: <008801cddfcb$d11fcdd0$735f6970$@ndzh.com> <B9BC66C4-3FE8-4F01-8D52-2E046F664985@kumari.net>
In-Reply-To: <B9BC66C4-3FE8-4F01-8D52-2E046F664985@kumari.net>
Date: Fri, 21 Dec 2012 21:45:19 -0500
Message-ID: <002801cddfee$62033400$26099c00$@ndzh.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQGqxnSzwsk3WhZLIGQY9swB9jDMxQJ5+lF8mFZ/xFA=
Content-Language: en-us
X-Authenticated-User: skh@ndzh.com 
X-IsFriend: <shares@ndzh.com>
Cc: grow@ietf.org, 'Susan Hares' <skh@ndzh.com>, "'John G. Scudder'" <jgs@bgp.nu>, "'Jon Mitchell \(GNS\)'" <Jon.Mitchell@microsoft.com>, stbryant@cisco.com
Subject: Re: [GROW] Requesting Grow's review of draft-ietf-idr-as-private-reservation-02.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 22 Dec 2012 02:45:55 -0000

Warren:

Thank you for the good comments.  To help the discussion not repeat itself,
please do see the idr list.  

The time period to comment is 2 weeks with comment period ending on
1/4/2013. 

Sue 

-----Original Message-----
From: Warren Kumari [mailto:warren@kumari.net] 
Sent: Friday, December 21, 2012 6:23 PM
To: Susan Hares
Cc: Warren Kumari; grow@ietf.org; 'Susan Hares'; Jon Mitchell (GNS); 'John
G. Scudder'; stbryant@cisco.com
Subject: Re: [GROW] Requesting Grow's review of
draft-ietf-idr-as-private-reservation-02.txt


On Dec 21, 2012, at 5:37 PM, "Susan Hares" <shares@ndzh.com> wrote:

> Peter and Chris:
>  
> We request that the GROW WG review
draft-ietf-idr-as-private-reservation-02.txt?
(http://datatracker.ietf.org/doc/draft-ietf-idr-as-private-reservation/)
>  
> Since it modifies RFC 1930, a BCP, it is likely to be a BCP.  The draft
deals with the creation of Private USE ASN space in 4 byte octets.
>  
> Two additional questions we'd love input on:  
>  
> 1.     Do they think the range is too big, too small, or just right"
(We'll call this the 3 bears question (smile))
> 2.     Would they prefer to see the range reserved with some portion of
the range allocated?
>  
> For example, reserve all the space but allocate only 50% of the space.
>  
> This draft has had substantial IDR debate which is (of course) online, and
has been through IDR WG LC.
>  
> We'd appreciate hearing about any operational issues GROW finds with this
draft.

As someone who reads both GROW and IDR I'd just like to more that there has
already been some operators providing input over on the IDR list...
This is not intended to discourage GROWers from commenting, but rather a
suggestion that before erupting in indignation (or giggles!) you have a
quick look @ the IDR threads on this. They may:
a: address your concerns or
b: provide you with more ammunition for making your case :-P

W

>  
> Thank you,
>  
> John and Sue
>  
> _______________________________________________
> GROW mailing list
> GROW@ietf.org
> https://www.ietf.org/mailman/listinfo/grow

--
For every complex problem, there is a solution that is simple, neat, and
wrong.
                -- H. L. Mencken






From internet-drafts@ietf.org  Thu Dec 27 10:35:05 2012
Return-Path: <internet-drafts@ietf.org>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 97A9E21F8D78; Thu, 27 Dec 2012 10:35:05 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.411
X-Spam-Level: 
X-Spam-Status: No, score=-102.411 tagged_above=-999 required=5 tests=[AWL=-0.039, BAYES_00=-2.599, SARE_SUB_OBFU_Q1=0.227, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id W1MpgiUGdUNR; Thu, 27 Dec 2012 10:35:05 -0800 (PST)
Received: from ietfa.amsl.com (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 222EE21F83EF; Thu, 27 Dec 2012 10:35:05 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From: internet-drafts@ietf.org
To: i-d-announce@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 4.37
Message-ID: <20121227183505.30845.59212.idtracker@ietfa.amsl.com>
Date: Thu, 27 Dec 2012 10:35:05 -0800
Cc: grow@ietf.org
Subject: [GROW] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 27 Dec 2012 18:35:05 -0000

A New Internet-Draft is available from the on-line Internet-Drafts director=
ies.
 This draft is a work item of the Global Routing Operations Working Group o=
f the IETF.

	Title           : Operational Requirements for Enhanced Error Handling Beh=
aviour in BGP-4
	Author(s)       : Rob Shakir
	Filename        : draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
	Pages           : 19
	Date            : 2012-12-27

Abstract:
   BGP is utilised as a key intra- and inter-autonomous system routing
   protocol in modern IP networks.  The failure modes, as defined by the
   original protocol standards, are based on a number of assumptions
   around the impact of session failure.  Numerous incidents both in the
   global Internet routing table and within service provider networks
   have been caused by strict handling of a single invalid UPDATE
   message causing large-scale failures in one or more autonomous
   systems.

   This memo describes the current use of BGP within service provider
   networks, and outlines a set of requirements for further work to
   enhance the mechanisms available to a BGP implementation when
   erroneous data is detected.  Whilst this document does not provide
   specification of any standard, it is intended as an overview of a set
   of enhancements to BGP to improve the protocol's robustness to suit
   its current deployment.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-grow-ops-reqs-for-bgp-error-han=
dling

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-grow-ops-reqs-for-bgp-error-handling-=
06

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-grow-ops-reqs-for-bgp-error-h=
andling-06


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


From rob.shakir@bt.com  Thu Dec 27 10:42:02 2012
Return-Path: <rob.shakir@bt.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B2CD621F8D6B; Thu, 27 Dec 2012 10:42:02 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.372
X-Spam-Level: 
X-Spam-Status: No, score=-3.372 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Xn3A+dPDF84b; Thu, 27 Dec 2012 10:42:02 -0800 (PST)
Received: from smtpe1.intersmtp.com (smtp63.intersmtp.com [62.239.224.236]) by ietfa.amsl.com (Postfix) with ESMTP id F13F321F8D67; Thu, 27 Dec 2012 10:42:01 -0800 (PST)
Received: from EVMHT68-UKRD.domain1.systemhost.net (10.36.3.105) by RDW083A007ED63.smtp-e3.hygiene.service (10.187.98.12) with Microsoft SMTP Server (TLS) id 8.3.279.1; Thu, 27 Dec 2012 18:41:55 +0000
Received: from E07HT04-UKBR.domain1.systemhost.net (193.113.197.162) by EVMHT68-UKRD.domain1.systemhost.net (10.36.3.105) with Microsoft SMTP Server (TLS) id 8.3.279.1; Thu, 27 Dec 2012 18:41:53 +0000
Received: from EMV02-UKBR.domain1.systemhost.net ([169.254.2.250]) by E07HT04-UKBR.domain1.systemhost.net ([193.113.197.162]) with mapi; Thu, 27 Dec 2012 18:41:53 +0000
From: <rob.shakir@bt.com>
To: <internet-drafts@ietf.org>, <i-d-announce@ietf.org>
Date: Thu, 27 Dec 2012 18:41:50 +0000
Thread-Topic: [GROW] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
Thread-Index: Ac3kYdbcZ/K2BcQERw28QPY8Oe+ftg==
Message-ID: <CD024583.EA71%rob.shakir@bt.com>
In-Reply-To: <20121227183505.30845.59212.idtracker@ietfa.amsl.com>
Accept-Language: en-US, en-GB
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/14.2.5.121010
acceptlanguage: en-US, en-GB
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: grow@ietf.org
Subject: Re: [GROW] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 27 Dec 2012 18:42:02 -0000

On 27/12/2012 18:35, "internet-drafts@ietf.org" <internet-drafts@ietf.org>
wrote:

>
>A New Internet-Draft is available from the on-line Internet-Drafts
>directories.
> This draft is a work item of the Global Routing Operations Working Group
>of the IETF.
>
>	Title           : Operational Requirements for Enhanced Error Handling
>Behaviour in BGP-4
>	Author(s)       : Rob Shakir
>	Filename        : draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
>	Pages           : 19
>	Date            : 2012-12-27

Hi GROW!

This update is a fairly major re-spin of the BGP Error Handling
requirements draft. The technical content should be as per the previous
revisions however, following the ietf/RtgDir last call comments, I have
made the following changes:

* Made the amendments that were discussed and there was no disagreement
with from our meeting in Atlanta -- this is essentially renaming the
Critical/Semantic error types to Critical/Non-Critical.

* Significant de-duplication within the text including merging the
operational monitoring/toolset discussions into the error handling
sections.

* Adoption of rfc2119 language throughout to clarify the requirements.

* Removal of some of the discussion around more detailed justifications
for why particular decisions were made. I think this was useful through
the discussion phase of this draft, but it seems like GROW/IDR have
converged on a relatively stable set of requirements, so I have trimmed
back some of this discussion.

I'd really welcome any further comments on this before we re-submit for
publication. To eke these out - Peter/Chris - can you kick off a WGLC for
this draft please? :-)

Seasons greetings!
r.


From rjs@rob.sh  Fri Dec 28 05:10:08 2012
Return-Path: <rjs@rob.sh>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 206AF21F8609; Fri, 28 Dec 2012 05:10:08 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.152
X-Spam-Level: 
X-Spam-Status: No, score=0.152 tagged_above=-999 required=5 tests=[AWL=-2.476,  BAYES_00=-2.599, GB_SUMOF=5, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RGiQ4Myuwey2; Fri, 28 Dec 2012 05:10:07 -0800 (PST)
Received: from cappuccino.rob.sh (cappuccino.rob.sh [IPv6:2001:b98:201:101::10:cafe]) by ietfa.amsl.com (Postfix) with ESMTP id 4110121F88F1; Fri, 28 Dec 2012 05:10:06 -0800 (PST)
Received: from [46.65.174.227] (helo=latte.munster) by cappuccino.rob.sh with esmtpa (Exim 4.72) (envelope-from <rjs@rob.sh>) id 1ToZeK-00072O-BL; Fri, 28 Dec 2012 13:06:56 +0000
Mime-Version: 1.0 (Apple Message framework v1283)
Content-Type: text/plain; charset=windows-1252
From: Rob Shakir <rjs@rob.sh>
In-Reply-To: <025801cde4f6$d7783700$8668a500$@highwayman.com>
Date: Fri, 28 Dec 2012 13:10:04 +0000
Content-Transfer-Encoding: quoted-printable
Message-Id: <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com>
To: "Chris Hall" <chris.hall@highwayman.com>
X-Mailer: Apple Mail (2.1283)
Cc: idr@ietf.org, grow@ietf.org
Subject: Re: [GROW] [Idr] Fwd: I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 28 Dec 2012 13:10:08 -0000

Hi Chris,

(re: CCing IDR & GROW)

On 28 Dec 2012, at 12:28, Chris Hall wrote:

> Rob Shakir wrote (on Thu 27-Dec-2012 at 18:44):
>>=20
>> Any comments very welcome (to me or grow@).
>=20
> I'm afraid I still don't get it :-(  What am I missing ?
>=20
> UPDATE Message Length errors are Critical because they (1) "result in
> cases whereby the NLRI attribute cannot be correctly extracted".
>=20
> The implication is that a failure to extract all NLRI is Critical.  Is
> that a requirement ?

If the NLRI cannot be determined, then this is a Critical error, yes. I =
left the wording relatively open on whether this is *all* NLRI, as I am =
not sure that in the requirements draft we should specify direct =
solutions to specific issues, to e.g., say how to handle cases where =
MP_REACH_NLRI and MP_UNREACH_NLRI are in the same message [this is a =
case that I do not believe is forbidden by rfc2858 - if the working =
group could clarify whether this is something that we feel the draft =
needs to handle or can explicitly be omitted, then that would be =
appreciated].

>=20
> Later:=20
>=20
>  (2) "All errors whereby the contained NLRI can be
>       extracted are referred to as Non-Critical".=20
>=20
> And that includes:
>=20
>  (3) "where the length of all path attributes contained
>       within the UPDATE does not correspond to the
>       total path attribute length."
>=20
> That is, at least, more explicit than
> draft-ietf-idr-error-handling-03, which glosses over (3).
>=20
> But if (3) is non-critical then there is some chance that some NLRI
> will not be extracted, which appears to violate (1) and (2).

Disclaimer: As I am sure that my comments previously have made clear, I =
do not maintain a code base for a BGP daemon/implementation - so please =
feel free to correct my logic below.

I do not believe that (3) implies that the NLRI cannot be correctly =
found. If the sum of total length is incorrect, then we can still =
extract the individual attributes - we just find that there is not =
enough data to fill the overall length we were told and/or we have too =
much attribute data compared to the total attribute length. In the case =
where the NLRI attribute itself has a length error, then this is a =
critical error (based on the "Errors parsing the NLRI attribute of an =
UPDATE message" definition of Critical error), and a similar Critical =
error occurs in the latter case, where the Total Path Attributes + =
Withdrawn Routes are not equal to total UPDATE message length.

Either way -- again, I would say that this is something that we need to =
put text together for draft-ietf-idr-error-handling rather than the =
requirements document (this sounds like a solution, and the requirement =
does not have a SHOULD or MUST here, it is an "it is expected that=85" =
comment).

> Then (4) "In order to maximise the number of cases whereby the NLRI
> attributes [plural, now, BTW] can be reliably extracted from a
> received message...".  Ah.  So it is not a Critical Error if "the NLRI
> attribute cannot be correctly extracted".

No - it is a Critical error if we cannot extract the NLRI. This =
recommendation is to give an increased chance that the NLRI can be =
extracted as per the IDR error handling draft. This then (by virtue of =
resulting in the NLRI being extracted) minimises the number of cases =
that result in a Critical error. The plural here is to reflect that the =
existence of >1 type of NLRI attribute.

> For me the requirement remains "conflicted".  On the one hand it seems
> to say that it is a Critical Error if the NLRI cannot be extracted and
> parsed.  On the other it seems to say it's OK if you cannot extract
> some NLRI.

If you'll forgive me for removing a significant proportion of your =
message, I think that we need to take another step back here. It seems =
to me that the key question that you are highlighting is "What level of =
confidence do we need to have before we declare that the NLRI cannot be =
extracted?" -- do you agree?

=46rom an operator perspective, I would like to compromise *certainty* =
for *robustness*. You are right, we are compromising correctness here, =
we might end up withdrawing an incorrect NLRI and impacting service =
operation for that prefix - however, it is somewhat preferable to me to =
withdraw a a subset of the NLRI incorrectly, rather than impact all NLRI =
in one single action. We clearly need to provide some bounds on how much =
we compromise the certainty (and live within the realms of possibility, =
such that we are not just taking a shot in the dark). This is what the =
definitions of Critical and Non-Critical within the document are =
intended to provide. Once again I will refer to the requirement that =
there is a balance between correctness and robustness - rather than a =
locally risk averse approach that results in harmful wider behaviour.

Is it acceptable that we leave this as guidance within the requirements? =
If not, please could you suggest how the definitions of =
Critical/Non-Critical could be altered to address your concerns? I would =
also appreciate further input from IDR as to whether this is sufficient =
requirement from GROW to allow a solution document to be written?

Many thanks,
r.




From brian.peter.dickson@gmail.com  Fri Dec 28 11:02:26 2012
Return-Path: <brian.peter.dickson@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D97D321F8E08; Fri, 28 Dec 2012 11:02:26 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.446
X-Spam-Level: 
X-Spam-Status: No, score=-0.446 tagged_above=-999 required=5 tests=[AWL=-2.675, BAYES_00=-2.599, GB_SUMOF=5, HTML_MESSAGE=0.001, J_CHICKENPOX_23=0.6, RCVD_IN_DNSWL_LOW=-1, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KdzDJeWDE6Oy; Fri, 28 Dec 2012 11:02:25 -0800 (PST)
Received: from mail-ee0-f45.google.com (mail-ee0-f45.google.com [74.125.83.45]) by ietfa.amsl.com (Postfix) with ESMTP id 609D321F8E07; Fri, 28 Dec 2012 11:02:24 -0800 (PST)
Received: by mail-ee0-f45.google.com with SMTP id d49so5251499eek.18 for <multiple recipients>; Fri, 28 Dec 2012 11:02:23 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=QO66L9RP0qeN8is3o5XU+MOtDvQRZzGqO7F5d6q24Co=; b=m+fJvs3K2/I/63jqKXNsPYs5syHvDLu1j6Vry74TPeX2yCov5OdY781ApN9YOPaVW1 zg4e/Tvboj6DLRWFI756XDYg+nOyPrD448Y0f+K1OogzAlyvTg1Dbpy1Dv5rHfwte+h5 6/yyw5ctb14slwPH9Q55/1LQ85ZPcP/o5bYXiyHZlh0ulEZb02cojVIZQMjxNy51ClIw Rpkdnyindt4cJSaiqVfnkzNkRaeZpaFyL9NaB/ImFEgqDWuu67tIxCF5876n2dINTi3m QHS3nFgO0JECcxVvSFJCjBnQbITJqHnSRUlTnHs/x2KIQqXp0Pe5RypeYbYQruD7k4BT o9Qw==
MIME-Version: 1.0
Received: by 10.14.225.194 with SMTP id z42mr88628745eep.22.1356721343446; Fri, 28 Dec 2012 11:02:23 -0800 (PST)
Received: by 10.223.173.199 with HTTP; Fri, 28 Dec 2012 11:02:23 -0800 (PST)
In-Reply-To: <028701cde525$bf262070$3d726150$@highwayman.com>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com>
Date: Fri, 28 Dec 2012 14:02:23 -0500
Message-ID: <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com>
From: Brian Dickson <brian.peter.dickson@gmail.com>
To: Chris Hall <chris.hall@highwayman.com>
Content-Type: multipart/alternative; boundary=047d7b66f24bbc3cc104d1ee4d5f
Cc: idr@ietf.org, grow@ietf.org
Subject: Re: [GROW] [Idr] Fwd: I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 28 Dec 2012 19:02:27 -0000

--047d7b66f24bbc3cc104d1ee4d5f
Content-Type: text/plain; charset=ISO-8859-1

Here's a quick question, about possible mechanisms for determining whether
or not we need to tear down a session.

If there is a chance that some NLRI weren't properly decoded:

- What about requesting (presuming the option was negotiated) a
route-refresh, or a "confirm per-AFI-SAFI prefix list"?

It's an expensive "parity check" but is one way of ensuring that we haven't
missed a withdrawn prefix.
If it isn't in the subsequently received list of prefixes, we missed it and
should withdraw it.
Everything else is a no-op - maybe we missed a new prefix (with new
attribute?), but that is what "treat as withdraw" gets you.

Just trying to make sure the cure isn't worse than the disease, in all
situations.
(Where "cure" is fail to withdraw because we "lost" the withdrawal because
of malformed update, or "cure" is tear down the session.)

Brian

On Fri, Dec 28, 2012 at 1:04 PM, Chris Hall <chris.hall@highwayman.com>wrote:

> Rob Shakir wrote (on Fri 28-Dec-2012 at 13:10 +0000):
> >
> > (re: CCing IDR & GROW)
> >
> > On 28 Dec 2012, at 12:28, Chris Hall wrote:
> >
> > > Rob Shakir wrote (on Thu 27-Dec-2012 at 18:44):
> > >>
> > >> Any comments very welcome (to me or grow@).
>
> > > I'm afraid I still don't get it :-(  What am I missing ?
> > >
> > > UPDATE Message Length errors are Critical because they:
> > >
> > > (1) "result in cases whereby the NLRI attribute cannot
> > >      be correctly extracted".
> > >
> > > The implication is that a failure to extract all NLRI is Critical.
> > > Is that a requirement ?
>
> > If the NLRI cannot be determined, then this is a Critical error,
> > yes. I left the wording relatively open on whether this is *all*
> > NLRI, ...
>
> OK.  So, if the message is broken (in any way):
>
>   * if no NLRI can be found, that's a Critical Error.
>
>     That would include the case where some broken
>     attribute upstream of any MP_XXX obscures that
>     MP_XXX.
>
>   * if some NLRI can be found, that's a Non-Critical error.
>
>     Such NLRI as can be found are treated-as-withdraw.
>
>     AND any NLRI that may or may not have been lost, are
>     ignored.
>
>     AND whatever dangers lost NLRI may pose are covered
>     by the caveats in 4.1.
>
> Yes ?
>
> > ... as I am not sure that in the requirements draft we should
> > specify direct solutions to specific issues, to e.g., say how to
> > handle cases where MP_REACH_NLRI and MP_UNREACH_NLRI are in the same
> > message [this is a case that I do not believe is forbidden by
> > rfc2858 - if the working group could clarify whether this is
> > something that we feel the draft needs to handle or can explicitly
> > be omitted, then that would be appreciated].
>
> The RFCs also allow any mix of IPv4 Unicast Reachable/Unreachable in
> the body of the message, with or without MP_REACH_NLRI and/or
> MP_UNREACH_NLRI.  If we call each of these a "collection of NLRI",
> then there may be between 1 and 4 collections of NLRI in a message
> (ignoring the End-of-RIB pseudo-UPDATE).  It is an error to have more
> than one MP_REACH_NLRI attribute or more than one MP_UNREACH_NLRI
> attribute.  It is not clear to me whether those should be Critical
> Errors -- but while we are relaxing all error checking, I don't see
> why they should be.
>
> It is possible that most implementations actually only send one
> collection per message... so, as you suggest, on a you-know-and-I-know
> basis, we'all could ignore the "theoretical" issues of more than one
> collection in a message.
>
> But, this does not matter if the requirements allow lost NLRI to
> simply be ignored.  If the sender sends any MP_XXX first, then there
> is little chance that NLRI will be lost, and things work better.  If
> the sender follows the old rules, and particularly if it sends more
> than one collection at a time, then there is more chance that NLRI
> will be lost.
>
> > > Later:
> > >
> > >  (2) "All errors whereby the contained NLRI can be
> > >       extracted are referred to as Non-Critical".
> > >
> > > And that includes:
> > >
> > >  (3) "where the length of all path attributes contained
> > >       within the UPDATE does not correspond to the
> > >       total path attribute length."
> > >
> > > That is, at least, more explicit than
> > > draft-ietf-idr-error-handling-03, which glosses over (3).
> > >
> > > But if (3) is non-critical then there is some chance that
> > > some NLRI will not be extracted, which appears to violate
> > > (1) and (2).
>
> > Disclaimer: As I am sure that my comments previously have made
> > clear, I do not maintain a code base for a BGP daemon/implementation
> > - so please feel free to correct my logic below.
> >
> > I do not believe that (3) implies that the NLRI cannot be correctly
> > found.
>
> What (3) implies to me is that somewhere amongst the attributes
> something is broken.  As Jakob Heitz succinctly puts it: "Once you
> have a malformed update, NOTHING is certain."  In particular, you
> cannot be certain that all NLRI referred to by the sender can be found
> by the receiver.
>
> > If the sum of total length is incorrect, then we can still
> > extract the individual attributes - we just find that there is not
> > enough data to fill the overall length we were told and/or we have
> > too much attribute data compared to the total attribute length.
>
> This is precisely where I think the requirements come unstuck.  There
> is almost no redundancy in the attribute encoding.  And almost all the
> redundancy there is is discarded by the new-form-error-handling.  So,
> if the sum of the attribute lengths is incorrect, then (inter alia)
> the receiver CANNOT know whether the attributes it has extracted are
> the attributes the sender intended to send.  In particular, the
> receiver cannot know it has extracted all NLRI (except in the,
> probably obscure, case of having extracted both MP_REACH_NLRI and
> MP_UNREACH_NLRI).  "Once you have a malformed update, NOTHING is
> certain."
>
> Conversely, if the sum of attribute lengths is correct, then there is
> a fighting chance that the attributes received are the attributes
> sent.  But, if one of those attributes is (say) nominally an
> ATOMIC_AGGREGATE which is apparently 200 octets long, that might be a
> worry.
>
> ...
> > > Then (4) "In order to maximise the number of cases whereby the
> > > NLRI attributes [plural, now, BTW] can be reliably extracted
> > > from a received message...".  Ah.  So it is not a Critical
> > > Error if "the NLRI attribute cannot be correctly extracted".
>
> > No - it is a Critical error if we cannot extract the NLRI. This
> > recommendation is to give an increased chance that the NLRI can be
> > extracted as per the IDR error handling draft. This then (by virtue
> > of resulting in the NLRI being extracted) minimises the number of
> > cases that result in a Critical error.
>
> By this do you mean "all" NLRI or only "some" NLRI ?  As above.
>
> > The plural here is to reflect
> > that the existence of >1 type of NLRI attribute.
>
> That would imply that it should be "NLRI attributes" throughout ?
> Though in most places where the draft speaks of "NLRI attribute" I
> think it means one, some, any or all of the "collections of NLRI" (as
> defined above).
>
> > > For me the requirement remains "conflicted".  On the one hand it
> > > seems to say that it is a Critical Error if the NLRI cannot be
> > > extracted and parsed.  On the other it seems to say it's OK if
> > > you cannot extract some NLRI.
>
> > If you'll forgive me for removing a significant proportion of your
> > message, I think that we need to take another step back here. It
> > seems to me that the key question that you are highlighting is "What
> > level of confidence do we need to have before we declare that the
> > NLRI cannot be extracted?" -- do you agree?
>
> Absolutely.
>
> > From an operator perspective, I would like to compromise *certainty*
> > for *robustness*. You are right, we are compromising correctness
> > here, we might end up withdrawing an incorrect NLRI and impacting
> > service operation for that prefix - however, it is somewhat
> > preferable to me to withdraw a a subset of the NLRI incorrectly,
> > rather than impact all NLRI in one single action. We clearly need to
> > provide some bounds on how much we compromise the certainty (and
> > live within the realms of possibility, such that we are not just
> > taking a shot in the dark). This is what the definitions of Critical
> > and Non-Critical within the document are intended to provide. Once
> > again I will refer to the requirement that there is a balance
> > between correctness and robustness - rather than a locally risk
> > averse approach that results in harmful wider behaviour.
>
> Where one can extract the NLRI from a broken message, then
> treat-as-withdraw is AFAICS no worse for the NLRI in question than
> session-reset, and hugely better for all the other NLRI received from
> the peer.
>
> Where not *all* the NLRI in a message have been extracted, then we
> take a step beyond withdrawing something which should not have been
> withdrawn.  For each of those NLRI the receiver will (unknowingly) do
> one of:
>
>   (a) continue using a route which the sender has
>       withdrawn;
>
>   (b) continue to use an out of date version of a route
>       which the sender has changed in some way, possibly
>       materially;
>
>   (c) to fail to use a new route the sender has now made
>       available.
>
> Of these (a) looks serious and (b) could be, but certainly the effect
> is different from session-reset; while (c) is, essentially,
> treat-as-withdraw.  If some risk "lost routes" is taken, then the
> impact of that clearly must be weighed against the impact of
> session-reset.  I have not found a discussion of the impact of "lost
> routes" in the draft, so I have no idea what the trade-off is, here.
>
> For completeness, if all forms and degrees of attribute broken-ness
> are required to be acceptable, then there are cases where the receiver
> cannot know whether *all* NLRI have been extracted, or not.
>
> > Is it acceptable that we leave this as guidance within the
> > requirements? If not, please could you suggest how the definitions
> > of Critical/Non-Critical could be altered to address your concerns?
>
> Without a change at the sender end, the problem is that: "Once you
> have a malformed update, NOTHING is certain."
>
> So, as I said earlier, the fundamental question is:
>
>   (F) is it OK to continue with a session after processing
>       a message which may have contained some NLRI which
>       could not be extracted ?
>
> As you say, the question hinges on the "which may have contained".  As
> above, it seems to me that the draft skates over the issue of "lost
> routes".
>
> It also hinges on the operational impact of such "lost routes", on
> which I am unqualified to pronounce.
>
> It seems to me there is a range of possibilities:
>
>   1) at one end of the spectrum, if "lost routes" are to
>      be avoided at all costs, then the rules for parsing
>      attributes need to be tight and without a change at
>      the sender end, what is achievable is constrained.
>
>   2) if "lost routes" are acceptable in the "theoretical"
>      cases, but you-know-and-I-know that most of the
>      time the case will not arise...
>
>      ...or "lost routes" are acceptable provided some
>      defined steps are taken to identify as much NLRI
>      as is reasonably possible...
>
>      ...then the requirements need to (a) make the case
>      and (b) describe the trade-off (so that designs
>      made to follow the requirements can be judged in
>      those terms).
>
>   3) at the other end of the spectrum, if "lost routes"
>      are always preferable to session-reset, then this
>      opens up an even looser approach to UPDATE message
>      handling, in which practically nothing need cause
>      a session-reset -- ie, practically nothing is a
>      Critical Error.
>
> And, as I have suggested in previous discussions, it is possible to
> consider case (2) in two parts:
>
>   a) where the sender is unchanged.
>
>      In which case the issue is the degree of attribute
>      broken-ness vs the likelihood of "lost routes".
>
>      If this is viewed as an intermediate step, then
>      perhaps the requirements could err on the side of
>      safety ?
>
>      Perhaps this can be addressed by a "knob" allowing
>      more or less strictness in the parsing of
>      attributes ?
>
>   b) where the sender is changed.
>
>      In which case the issue ought to be moot, because
>      the receiver will then be able to know that there
>      are no "lost routes".
>
> > I would also appreciate further input from IDR as to whether this is
> > sufficient requirement from GROW to allow a solution document to be
> > written?
>
> The new requirements specify only Critical (session-reset) and
> Non-Critical (treat-as-withdraw) Errors.  This appears to rule out the
> "attribute discard" option in draft-ietf-idr-error-handling-03.  Is
> that intended ?
>
> Chris
>
> _______________________________________________
> Idr mailing list
> Idr@ietf.org
> https://www.ietf.org/mailman/listinfo/idr
>

--047d7b66f24bbc3cc104d1ee4d5f
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Here&#39;s a quick question, about possible mechanisms for determining whet=
her or not we need to tear down a session.<div><br></div><div>If there is a=
 chance that some NLRI weren&#39;t properly decoded:</div><div><br></div>
<div>- What about requesting (presuming the option was negotiated) a route-=
refresh, or a &quot;confirm per-AFI-SAFI prefix list&quot;?</div><div><br><=
/div><div>It&#39;s an expensive &quot;parity check&quot; but is one way of =
ensuring that we haven&#39;t missed a withdrawn prefix.</div>
<div>If it isn&#39;t in the subsequently received list of prefixes, we miss=
ed it and should withdraw it.</div><div>Everything else is a no-op - maybe =
we missed a new prefix (with new attribute?), but that is what &quot;treat =
as withdraw&quot; gets you.</div>
<div><br></div><div>Just trying to make sure the cure isn&#39;t worse than =
the disease, in all situations.</div><div>(Where &quot;cure&quot; is fail t=
o withdraw because we &quot;lost&quot; the withdrawal because of malformed =
update, or &quot;cure&quot; is tear down the session.)</div>
<div><br></div><div>Brian<br><br><div class=3D"gmail_quote">On Fri, Dec 28,=
 2012 at 1:04 PM, Chris Hall <span dir=3D"ltr">&lt;<a href=3D"mailto:chris.=
hall@highwayman.com" target=3D"_blank">chris.hall@highwayman.com</a>&gt;</s=
pan> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">Rob Shakir wrote (on Fri 28-Dec-2012 at 13:1=
0 +0000):<br>
<div class=3D"im">&gt;<br>
&gt; (re: CCing IDR &amp; GROW)<br>
&gt;<br>
&gt; On 28 Dec 2012, at 12:28, Chris Hall wrote:<br>
&gt;<br>
&gt; &gt; Rob Shakir wrote (on Thu 27-Dec-2012 at 18:44):<br>
&gt; &gt;&gt;<br>
&gt; &gt;&gt; Any comments very welcome (to me or grow@).<br>
<br>
&gt; &gt; I&#39;m afraid I still don&#39;t get it :-( =A0What am I missing =
?<br>
&gt; &gt;<br>
&gt; &gt; UPDATE Message Length errors are Critical because they:<br>
&gt; &gt;<br>
&gt; &gt; (1) &quot;result in cases whereby the NLRI attribute cannot<br>
&gt; &gt; =A0 =A0 =A0be correctly extracted&quot;.<br>
&gt; &gt;<br>
&gt; &gt; The implication is that a failure to extract all NLRI is Critical=
.<br>
&gt; &gt; Is that a requirement ?<br>
<br>
&gt; If the NLRI cannot be determined, then this is a Critical error,<br>
&gt; yes. I left the wording relatively open on whether this is *all*<br>
</div>&gt; NLRI, ...<br>
<br>
OK. =A0So, if the message is broken (in any way):<br>
<br>
=A0 * if no NLRI can be found, that&#39;s a Critical Error.<br>
<br>
=A0 =A0 That would include the case where some broken<br>
=A0 =A0 attribute upstream of any MP_XXX obscures that<br>
=A0 =A0 MP_XXX.<br>
<br>
=A0 * if some NLRI can be found, that&#39;s a Non-Critical error.<br>
<br>
=A0 =A0 Such NLRI as can be found are treated-as-withdraw.<br>
<br>
=A0 =A0 AND any NLRI that may or may not have been lost, are<br>
=A0 =A0 ignored.<br>
<br>
=A0 =A0 AND whatever dangers lost NLRI may pose are covered<br>
=A0 =A0 by the caveats in 4.1.<br>
<br>
Yes ?<br>
<br>
&gt; ... as I am not sure that in the requirements draft we should<br>
<div class=3D"im">&gt; specify direct solutions to specific issues, to e.g.=
, say how to<br>
&gt; handle cases where MP_REACH_NLRI and MP_UNREACH_NLRI are in the same<b=
r>
&gt; message [this is a case that I do not believe is forbidden by<br>
&gt; rfc2858 - if the working group could clarify whether this is<br>
&gt; something that we feel the draft needs to handle or can explicitly<br>
&gt; be omitted, then that would be appreciated].<br>
<br>
</div>The RFCs also allow any mix of IPv4 Unicast Reachable/Unreachable in<=
br>
the body of the message, with or without MP_REACH_NLRI and/or<br>
MP_UNREACH_NLRI. =A0If we call each of these a &quot;collection of NLRI&quo=
t;,<br>
then there may be between 1 and 4 collections of NLRI in a message<br>
(ignoring the End-of-RIB pseudo-UPDATE). =A0It is an error to have more<br>
than one MP_REACH_NLRI attribute or more than one MP_UNREACH_NLRI<br>
attribute. =A0It is not clear to me whether those should be Critical<br>
Errors -- but while we are relaxing all error checking, I don&#39;t see<br>
why they should be.<br>
<br>
It is possible that most implementations actually only send one<br>
collection per message... so, as you suggest, on a you-know-and-I-know<br>
basis, we&#39;all could ignore the &quot;theoretical&quot; issues of more t=
han one<br>
collection in a message.<br>
<br>
But, this does not matter if the requirements allow lost NLRI to<br>
simply be ignored. =A0If the sender sends any MP_XXX first, then there<br>
is little chance that NLRI will be lost, and things work better. =A0If<br>
the sender follows the old rules, and particularly if it sends more<br>
than one collection at a time, then there is more chance that NLRI<br>
will be lost.<br>
<div class=3D"im"><br>
&gt; &gt; Later:<br>
&gt; &gt;<br>
&gt; &gt; =A0(2) &quot;All errors whereby the contained NLRI can be<br>
&gt; &gt; =A0 =A0 =A0 extracted are referred to as Non-Critical&quot;.<br>
&gt; &gt;<br>
&gt; &gt; And that includes:<br>
&gt; &gt;<br>
&gt; &gt; =A0(3) &quot;where the length of all path attributes contained<br=
>
&gt; &gt; =A0 =A0 =A0 within the UPDATE does not correspond to the<br>
&gt; &gt; =A0 =A0 =A0 total path attribute length.&quot;<br>
&gt; &gt;<br>
&gt; &gt; That is, at least, more explicit than<br>
&gt; &gt; draft-ietf-idr-error-handling-03, which glosses over (3).<br>
&gt; &gt;<br>
&gt; &gt; But if (3) is non-critical then there is some chance that<br>
&gt; &gt; some NLRI will not be extracted, which appears to violate<br>
&gt; &gt; (1) and (2).<br>
<br>
&gt; Disclaimer: As I am sure that my comments previously have made<br>
&gt; clear, I do not maintain a code base for a BGP daemon/implementation<b=
r>
&gt; - so please feel free to correct my logic below.<br>
&gt;<br>
&gt; I do not believe that (3) implies that the NLRI cannot be correctly<br=
>
&gt; found.<br>
<br>
</div>What (3) implies to me is that somewhere amongst the attributes<br>
something is broken. =A0As Jakob Heitz succinctly puts it: &quot;Once you<b=
r>
have a malformed update, NOTHING is certain.&quot; =A0In particular, you<br=
>
cannot be certain that all NLRI referred to by the sender can be found<br>
by the receiver.<br>
<div class=3D"im"><br>
&gt; If the sum of total length is incorrect, then we can still<br>
&gt; extract the individual attributes - we just find that there is not<br>
&gt; enough data to fill the overall length we were told and/or we have<br>
&gt; too much attribute data compared to the total attribute length.<br>
<br>
</div>This is precisely where I think the requirements come unstuck. =A0The=
re<br>
is almost no redundancy in the attribute encoding. =A0And almost all the<br=
>
redundancy there is is discarded by the new-form-error-handling. =A0So,<br>
if the sum of the attribute lengths is incorrect, then (inter alia)<br>
the receiver CANNOT know whether the attributes it has extracted are<br>
the attributes the sender intended to send. =A0In particular, the<br>
receiver cannot know it has extracted all NLRI (except in the,<br>
probably obscure, case of having extracted both MP_REACH_NLRI and<br>
MP_UNREACH_NLRI). =A0&quot;Once you have a malformed update, NOTHING is<br>
certain.&quot;<br>
<br>
Conversely, if the sum of attribute lengths is correct, then there is<br>
a fighting chance that the attributes received are the attributes<br>
sent. =A0But, if one of those attributes is (say) nominally an<br>
ATOMIC_AGGREGATE which is apparently 200 octets long, that might be a<br>
worry.<br>
<br>
...<br>
<div class=3D"im">&gt; &gt; Then (4) &quot;In order to maximise the number =
of cases whereby the<br>
&gt; &gt; NLRI attributes [plural, now, BTW] can be reliably extracted<br>
&gt; &gt; from a received message...&quot;. =A0Ah. =A0So it is not a Critic=
al<br>
&gt; &gt; Error if &quot;the NLRI attribute cannot be correctly extracted&q=
uot;.<br>
<br>
&gt; No - it is a Critical error if we cannot extract the NLRI. This<br>
&gt; recommendation is to give an increased chance that the NLRI can be<br>
&gt; extracted as per the IDR error handling draft. This then (by virtue<br=
>
&gt; of resulting in the NLRI being extracted) minimises the number of<br>
&gt; cases that result in a Critical error.<br>
<br>
</div>By this do you mean &quot;all&quot; NLRI or only &quot;some&quot; NLR=
I ? =A0As above.<br>
<div class=3D"im"><br>
&gt; The plural here is to reflect<br>
&gt; that the existence of &gt;1 type of NLRI attribute.<br>
<br>
</div>That would imply that it should be &quot;NLRI attributes&quot; throug=
hout ?<br>
Though in most places where the draft speaks of &quot;NLRI attribute&quot; =
I<br>
think it means one, some, any or all of the &quot;collections of NLRI&quot;=
 (as<br>
defined above).<br>
<div class=3D"im"><br>
&gt; &gt; For me the requirement remains &quot;conflicted&quot;. =A0On the =
one hand it<br>
&gt; &gt; seems to say that it is a Critical Error if the NLRI cannot be<br=
>
&gt; &gt; extracted and parsed. =A0On the other it seems to say it&#39;s OK=
 if<br>
&gt; &gt; you cannot extract some NLRI.<br>
<br>
&gt; If you&#39;ll forgive me for removing a significant proportion of your=
<br>
&gt; message, I think that we need to take another step back here. It<br>
&gt; seems to me that the key question that you are highlighting is &quot;W=
hat<br>
&gt; level of confidence do we need to have before we declare that the<br>
&gt; NLRI cannot be extracted?&quot; -- do you agree?<br>
<br>
</div>Absolutely.<br>
<div class=3D"im"><br>
&gt; From an operator perspective, I would like to compromise *certainty*<b=
r>
&gt; for *robustness*. You are right, we are compromising correctness<br>
&gt; here, we might end up withdrawing an incorrect NLRI and impacting<br>
&gt; service operation for that prefix - however, it is somewhat<br>
&gt; preferable to me to withdraw a a subset of the NLRI incorrectly,<br>
&gt; rather than impact all NLRI in one single action. We clearly need to<b=
r>
&gt; provide some bounds on how much we compromise the certainty (and<br>
&gt; live within the realms of possibility, such that we are not just<br>
&gt; taking a shot in the dark). This is what the definitions of Critical<b=
r>
&gt; and Non-Critical within the document are intended to provide. Once<br>
&gt; again I will refer to the requirement that there is a balance<br>
&gt; between correctness and robustness - rather than a locally risk<br>
&gt; averse approach that results in harmful wider behaviour.<br>
<br>
</div>Where one can extract the NLRI from a broken message, then<br>
treat-as-withdraw is AFAICS no worse for the NLRI in question than<br>
session-reset, and hugely better for all the other NLRI received from<br>
the peer.<br>
<br>
Where not *all* the NLRI in a message have been extracted, then we<br>
take a step beyond withdrawing something which should not have been<br>
withdrawn. =A0For each of those NLRI the receiver will (unknowingly) do<br>
one of:<br>
<br>
=A0 (a) continue using a route which the sender has<br>
=A0 =A0 =A0 withdrawn;<br>
<br>
=A0 (b) continue to use an out of date version of a route<br>
=A0 =A0 =A0 which the sender has changed in some way, possibly<br>
=A0 =A0 =A0 materially;<br>
<br>
=A0 (c) to fail to use a new route the sender has now made<br>
=A0 =A0 =A0 available.<br>
<br>
Of these (a) looks serious and (b) could be, but certainly the effect<br>
is different from session-reset; while (c) is, essentially,<br>
treat-as-withdraw. =A0If some risk &quot;lost routes&quot; is taken, then t=
he<br>
impact of that clearly must be weighed against the impact of<br>
session-reset. =A0I have not found a discussion of the impact of &quot;lost=
<br>
routes&quot; in the draft, so I have no idea what the trade-off is, here.<b=
r>
<br>
For completeness, if all forms and degrees of attribute broken-ness<br>
are required to be acceptable, then there are cases where the receiver<br>
cannot know whether *all* NLRI have been extracted, or not.<br>
<div class=3D"im"><br>
&gt; Is it acceptable that we leave this as guidance within the<br>
&gt; requirements? If not, please could you suggest how the definitions<br>
&gt; of Critical/Non-Critical could be altered to address your concerns?<br=
>
<br>
</div>Without a change at the sender end, the problem is that: &quot;Once y=
ou<br>
have a malformed update, NOTHING is certain.&quot;<br>
<br>
So, as I said earlier, the fundamental question is:<br>
<br>
=A0 (F) is it OK to continue with a session after processing<br>
=A0 =A0 =A0 a message which may have contained some NLRI which<br>
=A0 =A0 =A0 could not be extracted ?<br>
<br>
As you say, the question hinges on the &quot;which may have contained&quot;=
. =A0As<br>
above, it seems to me that the draft skates over the issue of &quot;lost<br=
>
routes&quot;.<br>
<br>
It also hinges on the operational impact of such &quot;lost routes&quot;, o=
n<br>
which I am unqualified to pronounce.<br>
<br>
It seems to me there is a range of possibilities:<br>
<br>
=A0 1) at one end of the spectrum, if &quot;lost routes&quot; are to<br>
=A0 =A0 =A0be avoided at all costs, then the rules for parsing<br>
=A0 =A0 =A0attributes need to be tight and without a change at<br>
=A0 =A0 =A0the sender end, what is achievable is constrained.<br>
<br>
=A0 2) if &quot;lost routes&quot; are acceptable in the &quot;theoretical&q=
uot;<br>
=A0 =A0 =A0cases, but you-know-and-I-know that most of the<br>
=A0 =A0 =A0time the case will not arise...<br>
<br>
=A0 =A0 =A0...or &quot;lost routes&quot; are acceptable provided some<br>
=A0 =A0 =A0defined steps are taken to identify as much NLRI<br>
=A0 =A0 =A0as is reasonably possible...<br>
<br>
=A0 =A0 =A0...then the requirements need to (a) make the case<br>
=A0 =A0 =A0and (b) describe the trade-off (so that designs<br>
=A0 =A0 =A0made to follow the requirements can be judged in<br>
=A0 =A0 =A0those terms).<br>
<br>
=A0 3) at the other end of the spectrum, if &quot;lost routes&quot;<br>
=A0 =A0 =A0are always preferable to session-reset, then this<br>
=A0 =A0 =A0opens up an even looser approach to UPDATE message<br>
=A0 =A0 =A0handling, in which practically nothing need cause<br>
=A0 =A0 =A0a session-reset -- ie, practically nothing is a<br>
=A0 =A0 =A0Critical Error.<br>
<br>
And, as I have suggested in previous discussions, it is possible to<br>
consider case (2) in two parts:<br>
<br>
=A0 a) where the sender is unchanged.<br>
<br>
=A0 =A0 =A0In which case the issue is the degree of attribute<br>
=A0 =A0 =A0broken-ness vs the likelihood of &quot;lost routes&quot;.<br>
<br>
=A0 =A0 =A0If this is viewed as an intermediate step, then<br>
=A0 =A0 =A0perhaps the requirements could err on the side of<br>
=A0 =A0 =A0safety ?<br>
<br>
=A0 =A0 =A0Perhaps this can be addressed by a &quot;knob&quot; allowing<br>
=A0 =A0 =A0more or less strictness in the parsing of<br>
=A0 =A0 =A0attributes ?<br>
<br>
=A0 b) where the sender is changed.<br>
<br>
=A0 =A0 =A0In which case the issue ought to be moot, because<br>
=A0 =A0 =A0the receiver will then be able to know that there<br>
=A0 =A0 =A0are no &quot;lost routes&quot;.<br>
<div class=3D"im"><br>
&gt; I would also appreciate further input from IDR as to whether this is<b=
r>
&gt; sufficient requirement from GROW to allow a solution document to be<br=
>
&gt; written?<br>
<br>
</div>The new requirements specify only Critical (session-reset) and<br>
Non-Critical (treat-as-withdraw) Errors. =A0This appears to rule out the<br=
>
&quot;attribute discard&quot; option in draft-ietf-idr-error-handling-03. =
=A0Is<br>
that intended ?<br>
<span class=3D"HOEnZb"><font color=3D"#888888"><br>
Chris<br>
</font></span><div class=3D"HOEnZb"><div class=3D"h5"><br>
_______________________________________________<br>
Idr mailing list<br>
<a href=3D"mailto:Idr@ietf.org">Idr@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/idr" target=3D"_blank">htt=
ps://www.ietf.org/mailman/listinfo/idr</a><br>
</div></div></blockquote></div><br></div>

--047d7b66f24bbc3cc104d1ee4d5f--

From jakob.heitz@ericsson.com  Fri Dec 28 13:35:13 2012
Return-Path: <jakob.heitz@ericsson.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 79CB721F8E00; Fri, 28 Dec 2012 13:35:13 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.475
X-Spam-Level: 
X-Spam-Status: No, score=-3.475 tagged_above=-999 required=5 tests=[AWL=-3.013, BAYES_00=-2.599, BODY_ENHANCEMENT=0.309, GB_SUMOF=5, HTML_MESSAGE=0.001, J_CHICKENPOX_23=0.6, RCVD_IN_DNSWL_MED=-4, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EhDZ8q4Pf65N; Fri, 28 Dec 2012 13:35:11 -0800 (PST)
Received: from imr3.ericy.com (imr3.ericy.com [198.24.6.13]) by ietfa.amsl.com (Postfix) with ESMTP id 7080B21F8D67; Fri, 28 Dec 2012 13:35:11 -0800 (PST)
Received: from EUSAAHC006.ericsson.se ([147.117.188.90]) by imr3.ericy.com (8.13.8/8.13.8) with ESMTP id qBSLZ4QQ031744 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 28 Dec 2012 15:35:05 -0600
Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC006.ericsson.se ([147.117.188.90]) with mapi id 14.02.0318.004; Fri, 28 Dec 2012 16:35:04 -0500
From: Jakob Heitz <jakob.heitz@ericsson.com>
To: Brian Dickson <brian.peter.dickson@gmail.com>, Chris Hall <chris.hall@highwayman.com>
Thread-Topic: [GROW] [Idr] Fwd: I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
Thread-Index: AQHN5S3xsGJpUADsWEuoY0HJqT8HEpguuk7g
Date: Fri, 28 Dec 2012 21:35:03 +0000
Message-ID: <2F3EBB88EC3A454AAB08915FBF0B8C7E14368F@eusaamb109.ericsson.se>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com>
In-Reply-To: <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [147.117.188.135]
Content-Type: multipart/alternative; boundary="_000_2F3EBB88EC3A454AAB08915FBF0B8C7E14368Feusaamb109ericsso_"
MIME-Version: 1.0
Cc: "idr@ietf.org" <idr@ietf.org>, "grow@ietf.org" <grow@ietf.org>
Subject: Re: [GROW] [Idr] Fwd: I-D Action:	draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 28 Dec 2012 21:35:13 -0000

--_000_2F3EBB88EC3A454AAB08915FBF0B8C7E14368Feusaamb109ericsso_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Treat-as-withdraw is not a cure.
Tear down the session is not a cure.

All we can hope for after a malformed update is a
temporary mitigation until human intervention can
fix the problem.

IMO, the goal of error handling is to limit the
damage, not to cure the problem.

--
Jakob Heitz.



________________________________
From: grow-bounces@ietf.org [mailto:grow-bounces@ietf.org] On Behalf Of Bri=
an Dickson
Sent: Friday, December 28, 2012 11:02 AM
To: Chris Hall
Cc: idr@ietf.org; grow@ietf.org
Subject: Re: [GROW] [Idr] Fwd: I-D Action: draft-ietf-grow-ops-reqs-for-bgp=
-error-handling-06.txt

Here's a quick question, about possible mechanisms for determining whether =
or not we need to tear down a session.

If there is a chance that some NLRI weren't properly decoded:

- What about requesting (presuming the option was negotiated) a route-refre=
sh, or a "confirm per-AFI-SAFI prefix list"?

It's an expensive "parity check" but is one way of ensuring that we haven't=
 missed a withdrawn prefix.
If it isn't in the subsequently received list of prefixes, we missed it and=
 should withdraw it.
Everything else is a no-op - maybe we missed a new prefix (with new attribu=
te?), but that is what "treat as withdraw" gets you.

Just trying to make sure the cure isn't worse than the disease, in all situ=
ations.
(Where "cure" is fail to withdraw because we "lost" the withdrawal because =
of malformed update, or "cure" is tear down the session.)

Brian

On Fri, Dec 28, 2012 at 1:04 PM, Chris Hall <chris.hall@highwayman.com<mail=
to:chris.hall@highwayman.com>> wrote:
Rob Shakir wrote (on Fri 28-Dec-2012 at 13:10 +0000):
>
> (re: CCing IDR & GROW)
>
> On 28 Dec 2012, at 12:28, Chris Hall wrote:
>
> > Rob Shakir wrote (on Thu 27-Dec-2012 at 18:44):
> >>
> >> Any comments very welcome (to me or grow@).

> > I'm afraid I still don't get it :-(  What am I missing ?
> >
> > UPDATE Message Length errors are Critical because they:
> >
> > (1) "result in cases whereby the NLRI attribute cannot
> >      be correctly extracted".
> >
> > The implication is that a failure to extract all NLRI is Critical.
> > Is that a requirement ?

> If the NLRI cannot be determined, then this is a Critical error,
> yes. I left the wording relatively open on whether this is *all*
> NLRI, ...

OK.  So, if the message is broken (in any way):

  * if no NLRI can be found, that's a Critical Error.

    That would include the case where some broken
    attribute upstream of any MP_XXX obscures that
    MP_XXX.

  * if some NLRI can be found, that's a Non-Critical error.

    Such NLRI as can be found are treated-as-withdraw.

    AND any NLRI that may or may not have been lost, are
    ignored.

    AND whatever dangers lost NLRI may pose are covered
    by the caveats in 4.1.

Yes ?

> ... as I am not sure that in the requirements draft we should
> specify direct solutions to specific issues, to e.g., say how to
> handle cases where MP_REACH_NLRI and MP_UNREACH_NLRI are in the same
> message [this is a case that I do not believe is forbidden by
> rfc2858 - if the working group could clarify whether this is
> something that we feel the draft needs to handle or can explicitly
> be omitted, then that would be appreciated].

The RFCs also allow any mix of IPv4 Unicast Reachable/Unreachable in
the body of the message, with or without MP_REACH_NLRI and/or
MP_UNREACH_NLRI.  If we call each of these a "collection of NLRI",
then there may be between 1 and 4 collections of NLRI in a message
(ignoring the End-of-RIB pseudo-UPDATE).  It is an error to have more
than one MP_REACH_NLRI attribute or more than one MP_UNREACH_NLRI
attribute.  It is not clear to me whether those should be Critical
Errors -- but while we are relaxing all error checking, I don't see
why they should be.

It is possible that most implementations actually only send one
collection per message... so, as you suggest, on a you-know-and-I-know
basis, we'all could ignore the "theoretical" issues of more than one
collection in a message.

But, this does not matter if the requirements allow lost NLRI to
simply be ignored.  If the sender sends any MP_XXX first, then there
is little chance that NLRI will be lost, and things work better.  If
the sender follows the old rules, and particularly if it sends more
than one collection at a time, then there is more chance that NLRI
will be lost.

> > Later:
> >
> >  (2) "All errors whereby the contained NLRI can be
> >       extracted are referred to as Non-Critical".
> >
> > And that includes:
> >
> >  (3) "where the length of all path attributes contained
> >       within the UPDATE does not correspond to the
> >       total path attribute length."
> >
> > That is, at least, more explicit than
> > draft-ietf-idr-error-handling-03, which glosses over (3).
> >
> > But if (3) is non-critical then there is some chance that
> > some NLRI will not be extracted, which appears to violate
> > (1) and (2).

> Disclaimer: As I am sure that my comments previously have made
> clear, I do not maintain a code base for a BGP daemon/implementation
> - so please feel free to correct my logic below.
>
> I do not believe that (3) implies that the NLRI cannot be correctly
> found.

What (3) implies to me is that somewhere amongst the attributes
something is broken.  As Jakob Heitz succinctly puts it: "Once you
have a malformed update, NOTHING is certain."  In particular, you
cannot be certain that all NLRI referred to by the sender can be found
by the receiver.

> If the sum of total length is incorrect, then we can still
> extract the individual attributes - we just find that there is not
> enough data to fill the overall length we were told and/or we have
> too much attribute data compared to the total attribute length.

This is precisely where I think the requirements come unstuck.  There
is almost no redundancy in the attribute encoding.  And almost all the
redundancy there is is discarded by the new-form-error-handling.  So,
if the sum of the attribute lengths is incorrect, then (inter alia)
the receiver CANNOT know whether the attributes it has extracted are
the attributes the sender intended to send.  In particular, the
receiver cannot know it has extracted all NLRI (except in the,
probably obscure, case of having extracted both MP_REACH_NLRI and
MP_UNREACH_NLRI).  "Once you have a malformed update, NOTHING is
certain."

Conversely, if the sum of attribute lengths is correct, then there is
a fighting chance that the attributes received are the attributes
sent.  But, if one of those attributes is (say) nominally an
ATOMIC_AGGREGATE which is apparently 200 octets long, that might be a
worry.

...
> > Then (4) "In order to maximise the number of cases whereby the
> > NLRI attributes [plural, now, BTW] can be reliably extracted
> > from a received message...".  Ah.  So it is not a Critical
> > Error if "the NLRI attribute cannot be correctly extracted".

> No - it is a Critical error if we cannot extract the NLRI. This
> recommendation is to give an increased chance that the NLRI can be
> extracted as per the IDR error handling draft. This then (by virtue
> of resulting in the NLRI being extracted) minimises the number of
> cases that result in a Critical error.

By this do you mean "all" NLRI or only "some" NLRI ?  As above.

> The plural here is to reflect
> that the existence of >1 type of NLRI attribute.

That would imply that it should be "NLRI attributes" throughout ?
Though in most places where the draft speaks of "NLRI attribute" I
think it means one, some, any or all of the "collections of NLRI" (as
defined above).

> > For me the requirement remains "conflicted".  On the one hand it
> > seems to say that it is a Critical Error if the NLRI cannot be
> > extracted and parsed.  On the other it seems to say it's OK if
> > you cannot extract some NLRI.

> If you'll forgive me for removing a significant proportion of your
> message, I think that we need to take another step back here. It
> seems to me that the key question that you are highlighting is "What
> level of confidence do we need to have before we declare that the
> NLRI cannot be extracted?" -- do you agree?

Absolutely.

> From an operator perspective, I would like to compromise *certainty*
> for *robustness*. You are right, we are compromising correctness
> here, we might end up withdrawing an incorrect NLRI and impacting
> service operation for that prefix - however, it is somewhat
> preferable to me to withdraw a a subset of the NLRI incorrectly,
> rather than impact all NLRI in one single action. We clearly need to
> provide some bounds on how much we compromise the certainty (and
> live within the realms of possibility, such that we are not just
> taking a shot in the dark). This is what the definitions of Critical
> and Non-Critical within the document are intended to provide. Once
> again I will refer to the requirement that there is a balance
> between correctness and robustness - rather than a locally risk
> averse approach that results in harmful wider behaviour.

Where one can extract the NLRI from a broken message, then
treat-as-withdraw is AFAICS no worse for the NLRI in question than
session-reset, and hugely better for all the other NLRI received from
the peer.

Where not *all* the NLRI in a message have been extracted, then we
take a step beyond withdrawing something which should not have been
withdrawn.  For each of those NLRI the receiver will (unknowingly) do
one of:

  (a) continue using a route which the sender has
      withdrawn;

  (b) continue to use an out of date version of a route
      which the sender has changed in some way, possibly
      materially;

  (c) to fail to use a new route the sender has now made
      available.

Of these (a) looks serious and (b) could be, but certainly the effect
is different from session-reset; while (c) is, essentially,
treat-as-withdraw.  If some risk "lost routes" is taken, then the
impact of that clearly must be weighed against the impact of
session-reset.  I have not found a discussion of the impact of "lost
routes" in the draft, so I have no idea what the trade-off is, here.

For completeness, if all forms and degrees of attribute broken-ness
are required to be acceptable, then there are cases where the receiver
cannot know whether *all* NLRI have been extracted, or not.

> Is it acceptable that we leave this as guidance within the
> requirements? If not, please could you suggest how the definitions
> of Critical/Non-Critical could be altered to address your concerns?

Without a change at the sender end, the problem is that: "Once you
have a malformed update, NOTHING is certain."

So, as I said earlier, the fundamental question is:

  (F) is it OK to continue with a session after processing
      a message which may have contained some NLRI which
      could not be extracted ?

As you say, the question hinges on the "which may have contained".  As
above, it seems to me that the draft skates over the issue of "lost
routes".

It also hinges on the operational impact of such "lost routes", on
which I am unqualified to pronounce.

It seems to me there is a range of possibilities:

  1) at one end of the spectrum, if "lost routes" are to
     be avoided at all costs, then the rules for parsing
     attributes need to be tight and without a change at
     the sender end, what is achievable is constrained.

  2) if "lost routes" are acceptable in the "theoretical"
     cases, but you-know-and-I-know that most of the
     time the case will not arise...

     ...or "lost routes" are acceptable provided some
     defined steps are taken to identify as much NLRI
     as is reasonably possible...

     ...then the requirements need to (a) make the case
     and (b) describe the trade-off (so that designs
     made to follow the requirements can be judged in
     those terms).

  3) at the other end of the spectrum, if "lost routes"
     are always preferable to session-reset, then this
     opens up an even looser approach to UPDATE message
     handling, in which practically nothing need cause
     a session-reset -- ie, practically nothing is a
     Critical Error.

And, as I have suggested in previous discussions, it is possible to
consider case (2) in two parts:

  a) where the sender is unchanged.

     In which case the issue is the degree of attribute
     broken-ness vs the likelihood of "lost routes".

     If this is viewed as an intermediate step, then
     perhaps the requirements could err on the side of
     safety ?

     Perhaps this can be addressed by a "knob" allowing
     more or less strictness in the parsing of
     attributes ?

  b) where the sender is changed.

     In which case the issue ought to be moot, because
     the receiver will then be able to know that there
     are no "lost routes".

> I would also appreciate further input from IDR as to whether this is
> sufficient requirement from GROW to allow a solution document to be
> written?

The new requirements specify only Critical (session-reset) and
Non-Critical (treat-as-withdraw) Errors.  This appears to rule out the
"attribute discard" option in draft-ietf-idr-error-handling-03.  Is
that intended ?

Chris

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


--_000_2F3EBB88EC3A454AAB08915FBF0B8C7E14368Feusaamb109ericsso_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"=
>
<meta content=3D"MSHTML 6.00.6002.18727" name=3D"GENERATOR">
</head>
<body>
<div dir=3D"ltr" align=3D"left"><font face=3D"Lucida Console" color=3D"#800=
080" size=3D"2"><span class=3D"105522821-28122012">Treat-as-withdraw is not=
 a cure.</span></font></div>
<div dir=3D"ltr" align=3D"left"><font face=3D"Lucida Console" color=3D"#800=
080" size=3D"2"><span class=3D"105522821-28122012">Tear down the session is=
 not a cure.</span></font></div>
<div dir=3D"ltr" align=3D"left"><font face=3D"Lucida Console" color=3D"#800=
080" size=3D"2"><span class=3D"105522821-28122012"></span></font>&nbsp;</di=
v>
<div dir=3D"ltr" align=3D"left"><font face=3D"Lucida Console" color=3D"#800=
080" size=3D"2"><span class=3D"105522821-28122012">All we can hope for afte=
r a malformed update is a</span></font></div>
<div dir=3D"ltr" align=3D"left"><font face=3D"Lucida Console" color=3D"#800=
080" size=3D"2"><span class=3D"105522821-28122012">temporary mitigation unt=
il human intervention can</span></font></div>
<div dir=3D"ltr" align=3D"left"><font face=3D"Lucida Console" color=3D"#800=
080" size=3D"2"><span class=3D"105522821-28122012">fix the problem.</span><=
/font></div>
<div dir=3D"ltr" align=3D"left"><font face=3D"Lucida Console" color=3D"#800=
080" size=3D"2"><span class=3D"105522821-28122012"></span></font>&nbsp;</di=
v>
<div dir=3D"ltr" align=3D"left"><font face=3D"Lucida Console" color=3D"#800=
080" size=3D"2"><span class=3D"105522821-28122012">IMO, the goal of error h=
andling is to limit the</span></font></div>
<div dir=3D"ltr" align=3D"left"><font face=3D"Lucida Console" color=3D"#800=
080" size=3D"2"><span class=3D"105522821-28122012">damage, not to cure the =
problem.</span></font></div>
<!-- Converted from text/rtf format -->
<p><span lang=3D"en-us"><font face=3D"Arial" size=3D"2">--</font></span> <b=
r>
<span lang=3D"en-us"><font face=3D"Arial" size=3D"2">Jakob Heitz.</font></s=
pan></p>
<div>&nbsp;</div>
<br>
<blockquote dir=3D"ltr" style=3D"PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDE=
R-LEFT: #800080 2px solid; MARGIN-RIGHT: 0px">
<div class=3D"OutlookMessageHeader" lang=3D"en-us" dir=3D"ltr" align=3D"lef=
t">
<hr tabindex=3D"-1">
<font face=3D"Tahoma" size=3D"2"><b>From:</b> grow-bounces@ietf.org [mailto=
:grow-bounces@ietf.org]
<b>On Behalf Of </b>Brian Dickson<br>
<b>Sent:</b> Friday, December 28, 2012 11:02 AM<br>
<b>To:</b> Chris Hall<br>
<b>Cc:</b> idr@ietf.org; grow@ietf.org<br>
<b>Subject:</b> Re: [GROW] [Idr] Fwd: I-D Action: draft-ietf-grow-ops-reqs-=
for-bgp-error-handling-06.txt<br>
</font><br>
</div>
<div></div>
Here's a quick question, about possible mechanisms for determining whether =
or not we need to tear down a session.
<div><br>
</div>
<div>If there is a chance that some NLRI weren't properly decoded:</div>
<div><br>
</div>
<div>- What about requesting (presuming the option was negotiated) a route-=
refresh, or a &quot;confirm per-AFI-SAFI prefix list&quot;?</div>
<div><br>
</div>
<div>It's an expensive &quot;parity check&quot; but is one way of ensuring =
that we haven't missed a withdrawn prefix.</div>
<div>If it isn't in the subsequently received list of prefixes, we missed i=
t and should withdraw it.</div>
<div>Everything else is a no-op - maybe we missed a new prefix (with new at=
tribute?), but that is what &quot;treat as withdraw&quot; gets you.</div>
<div><br>
</div>
<div>Just trying to make sure the cure isn't worse than the disease, in all=
 situations.</div>
<div>(Where &quot;cure&quot; is fail to withdraw because we &quot;lost&quot=
; the withdrawal because of malformed update, or &quot;cure&quot; is tear d=
own the session.)</div>
<div><br>
</div>
<div>Brian<br>
<br>
<div class=3D"gmail_quote">On Fri, Dec 28, 2012 at 1:04 PM, Chris Hall <spa=
n dir=3D"ltr">
&lt;<a href=3D"mailto:chris.hall@highwayman.com" target=3D"_blank">chris.ha=
ll@highwayman.com</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"PADDING-LEFT: 1ex; MARGIN: 0px 0=
px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
Rob Shakir wrote (on Fri 28-Dec-2012 at 13:10 &#43;0000):<br>
<div class=3D"im">&gt;<br>
&gt; (re: CCing IDR &amp; GROW)<br>
&gt;<br>
&gt; On 28 Dec 2012, at 12:28, Chris Hall wrote:<br>
&gt;<br>
&gt; &gt; Rob Shakir wrote (on Thu 27-Dec-2012 at 18:44):<br>
&gt; &gt;&gt;<br>
&gt; &gt;&gt; Any comments very welcome (to me or grow@).<br>
<br>
&gt; &gt; I'm afraid I still don't get it :-( &nbsp;What am I missing ?<br>
&gt; &gt;<br>
&gt; &gt; UPDATE Message Length errors are Critical because they:<br>
&gt; &gt;<br>
&gt; &gt; (1) &quot;result in cases whereby the NLRI attribute cannot<br>
&gt; &gt; &nbsp; &nbsp; &nbsp;be correctly extracted&quot;.<br>
&gt; &gt;<br>
&gt; &gt; The implication is that a failure to extract all NLRI is Critical=
.<br>
&gt; &gt; Is that a requirement ?<br>
<br>
&gt; If the NLRI cannot be determined, then this is a Critical error,<br>
&gt; yes. I left the wording relatively open on whether this is *all*<br>
</div>
&gt; NLRI, ...<br>
<br>
OK. &nbsp;So, if the message is broken (in any way):<br>
<br>
&nbsp; * if no NLRI can be found, that's a Critical Error.<br>
<br>
&nbsp; &nbsp; That would include the case where some broken<br>
&nbsp; &nbsp; attribute upstream of any MP_XXX obscures that<br>
&nbsp; &nbsp; MP_XXX.<br>
<br>
&nbsp; * if some NLRI can be found, that's a Non-Critical error.<br>
<br>
&nbsp; &nbsp; Such NLRI as can be found are treated-as-withdraw.<br>
<br>
&nbsp; &nbsp; AND any NLRI that may or may not have been lost, are<br>
&nbsp; &nbsp; ignored.<br>
<br>
&nbsp; &nbsp; AND whatever dangers lost NLRI may pose are covered<br>
&nbsp; &nbsp; by the caveats in 4.1.<br>
<br>
Yes ?<br>
<br>
&gt; ... as I am not sure that in the requirements draft we should<br>
<div class=3D"im">&gt; specify direct solutions to specific issues, to e.g.=
, say how to<br>
&gt; handle cases where MP_REACH_NLRI and MP_UNREACH_NLRI are in the same<b=
r>
&gt; message [this is a case that I do not believe is forbidden by<br>
&gt; rfc2858 - if the working group could clarify whether this is<br>
&gt; something that we feel the draft needs to handle or can explicitly<br>
&gt; be omitted, then that would be appreciated].<br>
<br>
</div>
The RFCs also allow any mix of IPv4 Unicast Reachable/Unreachable in<br>
the body of the message, with or without MP_REACH_NLRI and/or<br>
MP_UNREACH_NLRI. &nbsp;If we call each of these a &quot;collection of NLRI&=
quot;,<br>
then there may be between 1 and 4 collections of NLRI in a message<br>
(ignoring the End-of-RIB pseudo-UPDATE). &nbsp;It is an error to have more<=
br>
than one MP_REACH_NLRI attribute or more than one MP_UNREACH_NLRI<br>
attribute. &nbsp;It is not clear to me whether those should be Critical<br>
Errors -- but while we are relaxing all error checking, I don't see<br>
why they should be.<br>
<br>
It is possible that most implementations actually only send one<br>
collection per message... so, as you suggest, on a you-know-and-I-know<br>
basis, we'all could ignore the &quot;theoretical&quot; issues of more than =
one<br>
collection in a message.<br>
<br>
But, this does not matter if the requirements allow lost NLRI to<br>
simply be ignored. &nbsp;If the sender sends any MP_XXX first, then there<b=
r>
is little chance that NLRI will be lost, and things work better. &nbsp;If<b=
r>
the sender follows the old rules, and particularly if it sends more<br>
than one collection at a time, then there is more chance that NLRI<br>
will be lost.<br>
<div class=3D"im"><br>
&gt; &gt; Later:<br>
&gt; &gt;<br>
&gt; &gt; &nbsp;(2) &quot;All errors whereby the contained NLRI can be<br>
&gt; &gt; &nbsp; &nbsp; &nbsp; extracted are referred to as Non-Critical&qu=
ot;.<br>
&gt; &gt;<br>
&gt; &gt; And that includes:<br>
&gt; &gt;<br>
&gt; &gt; &nbsp;(3) &quot;where the length of all path attributes contained=
<br>
&gt; &gt; &nbsp; &nbsp; &nbsp; within the UPDATE does not correspond to the=
<br>
&gt; &gt; &nbsp; &nbsp; &nbsp; total path attribute length.&quot;<br>
&gt; &gt;<br>
&gt; &gt; That is, at least, more explicit than<br>
&gt; &gt; draft-ietf-idr-error-handling-03, which glosses over (3).<br>
&gt; &gt;<br>
&gt; &gt; But if (3) is non-critical then there is some chance that<br>
&gt; &gt; some NLRI will not be extracted, which appears to violate<br>
&gt; &gt; (1) and (2).<br>
<br>
&gt; Disclaimer: As I am sure that my comments previously have made<br>
&gt; clear, I do not maintain a code base for a BGP daemon/implementation<b=
r>
&gt; - so please feel free to correct my logic below.<br>
&gt;<br>
&gt; I do not believe that (3) implies that the NLRI cannot be correctly<br=
>
&gt; found.<br>
<br>
</div>
What (3) implies to me is that somewhere amongst the attributes<br>
something is broken. &nbsp;As Jakob Heitz succinctly puts it: &quot;Once yo=
u<br>
have a malformed update, NOTHING is certain.&quot; &nbsp;In particular, you=
<br>
cannot be certain that all NLRI referred to by the sender can be found<br>
by the receiver.<br>
<div class=3D"im"><br>
&gt; If the sum of total length is incorrect, then we can still<br>
&gt; extract the individual attributes - we just find that there is not<br>
&gt; enough data to fill the overall length we were told and/or we have<br>
&gt; too much attribute data compared to the total attribute length.<br>
<br>
</div>
This is precisely where I think the requirements come unstuck. &nbsp;There<=
br>
is almost no redundancy in the attribute encoding. &nbsp;And almost all the=
<br>
redundancy there is is discarded by the new-form-error-handling. &nbsp;So,<=
br>
if the sum of the attribute lengths is incorrect, then (inter alia)<br>
the receiver CANNOT know whether the attributes it has extracted are<br>
the attributes the sender intended to send. &nbsp;In particular, the<br>
receiver cannot know it has extracted all NLRI (except in the,<br>
probably obscure, case of having extracted both MP_REACH_NLRI and<br>
MP_UNREACH_NLRI). &nbsp;&quot;Once you have a malformed update, NOTHING is<=
br>
certain.&quot;<br>
<br>
Conversely, if the sum of attribute lengths is correct, then there is<br>
a fighting chance that the attributes received are the attributes<br>
sent. &nbsp;But, if one of those attributes is (say) nominally an<br>
ATOMIC_AGGREGATE which is apparently 200 octets long, that might be a<br>
worry.<br>
<br>
...<br>
<div class=3D"im">&gt; &gt; Then (4) &quot;In order to maximise the number =
of cases whereby the<br>
&gt; &gt; NLRI attributes [plural, now, BTW] can be reliably extracted<br>
&gt; &gt; from a received message...&quot;. &nbsp;Ah. &nbsp;So it is not a =
Critical<br>
&gt; &gt; Error if &quot;the NLRI attribute cannot be correctly extracted&q=
uot;.<br>
<br>
&gt; No - it is a Critical error if we cannot extract the NLRI. This<br>
&gt; recommendation is to give an increased chance that the NLRI can be<br>
&gt; extracted as per the IDR error handling draft. This then (by virtue<br=
>
&gt; of resulting in the NLRI being extracted) minimises the number of<br>
&gt; cases that result in a Critical error.<br>
<br>
</div>
By this do you mean &quot;all&quot; NLRI or only &quot;some&quot; NLRI ? &n=
bsp;As above.<br>
<div class=3D"im"><br>
&gt; The plural here is to reflect<br>
&gt; that the existence of &gt;1 type of NLRI attribute.<br>
<br>
</div>
That would imply that it should be &quot;NLRI attributes&quot; throughout ?=
<br>
Though in most places where the draft speaks of &quot;NLRI attribute&quot; =
I<br>
think it means one, some, any or all of the &quot;collections of NLRI&quot;=
 (as<br>
defined above).<br>
<div class=3D"im"><br>
&gt; &gt; For me the requirement remains &quot;conflicted&quot;. &nbsp;On t=
he one hand it<br>
&gt; &gt; seems to say that it is a Critical Error if the NLRI cannot be<br=
>
&gt; &gt; extracted and parsed. &nbsp;On the other it seems to say it's OK =
if<br>
&gt; &gt; you cannot extract some NLRI.<br>
<br>
&gt; If you'll forgive me for removing a significant proportion of your<br>
&gt; message, I think that we need to take another step back here. It<br>
&gt; seems to me that the key question that you are highlighting is &quot;W=
hat<br>
&gt; level of confidence do we need to have before we declare that the<br>
&gt; NLRI cannot be extracted?&quot; -- do you agree?<br>
<br>
</div>
Absolutely.<br>
<div class=3D"im"><br>
&gt; From an operator perspective, I would like to compromise *certainty*<b=
r>
&gt; for *robustness*. You are right, we are compromising correctness<br>
&gt; here, we might end up withdrawing an incorrect NLRI and impacting<br>
&gt; service operation for that prefix - however, it is somewhat<br>
&gt; preferable to me to withdraw a a subset of the NLRI incorrectly,<br>
&gt; rather than impact all NLRI in one single action. We clearly need to<b=
r>
&gt; provide some bounds on how much we compromise the certainty (and<br>
&gt; live within the realms of possibility, such that we are not just<br>
&gt; taking a shot in the dark). This is what the definitions of Critical<b=
r>
&gt; and Non-Critical within the document are intended to provide. Once<br>
&gt; again I will refer to the requirement that there is a balance<br>
&gt; between correctness and robustness - rather than a locally risk<br>
&gt; averse approach that results in harmful wider behaviour.<br>
<br>
</div>
Where one can extract the NLRI from a broken message, then<br>
treat-as-withdraw is AFAICS no worse for the NLRI in question than<br>
session-reset, and hugely better for all the other NLRI received from<br>
the peer.<br>
<br>
Where not *all* the NLRI in a message have been extracted, then we<br>
take a step beyond withdrawing something which should not have been<br>
withdrawn. &nbsp;For each of those NLRI the receiver will (unknowingly) do<=
br>
one of:<br>
<br>
&nbsp; (a) continue using a route which the sender has<br>
&nbsp; &nbsp; &nbsp; withdrawn;<br>
<br>
&nbsp; (b) continue to use an out of date version of a route<br>
&nbsp; &nbsp; &nbsp; which the sender has changed in some way, possibly<br>
&nbsp; &nbsp; &nbsp; materially;<br>
<br>
&nbsp; (c) to fail to use a new route the sender has now made<br>
&nbsp; &nbsp; &nbsp; available.<br>
<br>
Of these (a) looks serious and (b) could be, but certainly the effect<br>
is different from session-reset; while (c) is, essentially,<br>
treat-as-withdraw. &nbsp;If some risk &quot;lost routes&quot; is taken, the=
n the<br>
impact of that clearly must be weighed against the impact of<br>
session-reset. &nbsp;I have not found a discussion of the impact of &quot;l=
ost<br>
routes&quot; in the draft, so I have no idea what the trade-off is, here.<b=
r>
<br>
For completeness, if all forms and degrees of attribute broken-ness<br>
are required to be acceptable, then there are cases where the receiver<br>
cannot know whether *all* NLRI have been extracted, or not.<br>
<div class=3D"im"><br>
&gt; Is it acceptable that we leave this as guidance within the<br>
&gt; requirements? If not, please could you suggest how the definitions<br>
&gt; of Critical/Non-Critical could be altered to address your concerns?<br=
>
<br>
</div>
Without a change at the sender end, the problem is that: &quot;Once you<br>
have a malformed update, NOTHING is certain.&quot;<br>
<br>
So, as I said earlier, the fundamental question is:<br>
<br>
&nbsp; (F) is it OK to continue with a session after processing<br>
&nbsp; &nbsp; &nbsp; a message which may have contained some NLRI which<br>
&nbsp; &nbsp; &nbsp; could not be extracted ?<br>
<br>
As you say, the question hinges on the &quot;which may have contained&quot;=
. &nbsp;As<br>
above, it seems to me that the draft skates over the issue of &quot;lost<br=
>
routes&quot;.<br>
<br>
It also hinges on the operational impact of such &quot;lost routes&quot;, o=
n<br>
which I am unqualified to pronounce.<br>
<br>
It seems to me there is a range of possibilities:<br>
<br>
&nbsp; 1) at one end of the spectrum, if &quot;lost routes&quot; are to<br>
&nbsp; &nbsp; &nbsp;be avoided at all costs, then the rules for parsing<br>
&nbsp; &nbsp; &nbsp;attributes need to be tight and without a change at<br>
&nbsp; &nbsp; &nbsp;the sender end, what is achievable is constrained.<br>
<br>
&nbsp; 2) if &quot;lost routes&quot; are acceptable in the &quot;theoretica=
l&quot;<br>
&nbsp; &nbsp; &nbsp;cases, but you-know-and-I-know that most of the<br>
&nbsp; &nbsp; &nbsp;time the case will not arise...<br>
<br>
&nbsp; &nbsp; &nbsp;...or &quot;lost routes&quot; are acceptable provided s=
ome<br>
&nbsp; &nbsp; &nbsp;defined steps are taken to identify as much NLRI<br>
&nbsp; &nbsp; &nbsp;as is reasonably possible...<br>
<br>
&nbsp; &nbsp; &nbsp;...then the requirements need to (a) make the case<br>
&nbsp; &nbsp; &nbsp;and (b) describe the trade-off (so that designs<br>
&nbsp; &nbsp; &nbsp;made to follow the requirements can be judged in<br>
&nbsp; &nbsp; &nbsp;those terms).<br>
<br>
&nbsp; 3) at the other end of the spectrum, if &quot;lost routes&quot;<br>
&nbsp; &nbsp; &nbsp;are always preferable to session-reset, then this<br>
&nbsp; &nbsp; &nbsp;opens up an even looser approach to UPDATE message<br>
&nbsp; &nbsp; &nbsp;handling, in which practically nothing need cause<br>
&nbsp; &nbsp; &nbsp;a session-reset -- ie, practically nothing is a<br>
&nbsp; &nbsp; &nbsp;Critical Error.<br>
<br>
And, as I have suggested in previous discussions, it is possible to<br>
consider case (2) in two parts:<br>
<br>
&nbsp; a) where the sender is unchanged.<br>
<br>
&nbsp; &nbsp; &nbsp;In which case the issue is the degree of attribute<br>
&nbsp; &nbsp; &nbsp;broken-ness vs the likelihood of &quot;lost routes&quot=
;.<br>
<br>
&nbsp; &nbsp; &nbsp;If this is viewed as an intermediate step, then<br>
&nbsp; &nbsp; &nbsp;perhaps the requirements could err on the side of<br>
&nbsp; &nbsp; &nbsp;safety ?<br>
<br>
&nbsp; &nbsp; &nbsp;Perhaps this can be addressed by a &quot;knob&quot; all=
owing<br>
&nbsp; &nbsp; &nbsp;more or less strictness in the parsing of<br>
&nbsp; &nbsp; &nbsp;attributes ?<br>
<br>
&nbsp; b) where the sender is changed.<br>
<br>
&nbsp; &nbsp; &nbsp;In which case the issue ought to be moot, because<br>
&nbsp; &nbsp; &nbsp;the receiver will then be able to know that there<br>
&nbsp; &nbsp; &nbsp;are no &quot;lost routes&quot;.<br>
<div class=3D"im"><br>
&gt; I would also appreciate further input from IDR as to whether this is<b=
r>
&gt; sufficient requirement from GROW to allow a solution document to be<br=
>
&gt; written?<br>
<br>
</div>
The new requirements specify only Critical (session-reset) and<br>
Non-Critical (treat-as-withdraw) Errors. &nbsp;This appears to rule out the=
<br>
&quot;attribute discard&quot; option in draft-ietf-idr-error-handling-03. &=
nbsp;Is<br>
that intended ?<br>
<span class=3D"HOEnZb"><font color=3D"#888888"><br>
Chris<br>
</font></span>
<div class=3D"HOEnZb">
<div class=3D"h5"><br>
_______________________________________________<br>
Idr mailing list<br>
<a href=3D"mailto:Idr@ietf.org">Idr@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/idr" target=3D"_blank">htt=
ps://www.ietf.org/mailman/listinfo/idr</a><br>
</div>
</div>
</blockquote>
</div>
<br>
</div>
</blockquote>
</body>
</html>

--_000_2F3EBB88EC3A454AAB08915FBF0B8C7E14368Feusaamb109ericsso_--

From rjs@rob.sh  Sat Dec 29 16:05:17 2012
Return-Path: <rjs@rob.sh>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A32E121F8496; Sat, 29 Dec 2012 16:05:16 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 3.229
X-Spam-Level: ***
X-Spam-Status: No, score=3.229 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, GB_SUMOF=5, HTML_MESSAGE=0.001, J_CHICKENPOX_23=0.6, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wXtgLkHQ2ltr; Sat, 29 Dec 2012 16:05:14 -0800 (PST)
Received: from cappuccino.rob.sh (cappuccino.rob.sh [IPv6:2001:b98:201:101::10:cafe]) by ietfa.amsl.com (Postfix) with ESMTP id 0280221F845F; Sat, 29 Dec 2012 16:05:14 -0800 (PST)
Received: from [86.0.24.33] (helo=[192.168.0.19]) by cappuccino.rob.sh with esmtpa (Exim 4.72) (envelope-from <rjs@rob.sh>) id 1Tp6Lo-0007uX-CB; Sun, 30 Dec 2012 00:02:00 +0000
Content-Type: multipart/alternative; boundary="Apple-Mail=_81DB352D-E036-47D8-9A50-6E15BB53EFB2"
Mime-Version: 1.0 (Mac OS X Mail 6.2 \(1499\))
From: Rob Shakir <rjs@rob.sh>
In-Reply-To: <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com>
Date: Sun, 30 Dec 2012 00:05:08 +0000
Message-Id: <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com>
To: Brian Dickson <brian.peter.dickson@gmail.com>
X-Mailer: Apple Mail (2.1499)
Cc: idr@ietf.org, grow@ietf.org, Chris Hall <chris.hall@highwayman.com>
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 30 Dec 2012 00:05:17 -0000

--Apple-Mail=_81DB352D-E036-47D8-9A50-6E15BB53EFB2
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=iso-8859-1

Hi Brian,

We discussed this at some length previously (the -00 of the requirements =
draft discussed automatically triggering some form of means to recover =
from the error). I think that there is some merit in doing so, but there =
are clearly also scaling considerations around this (in terms of =
requiring both BGP speakers to re-generate/re-process UPDATEs).=20

The consensus that was reached in the previous discussion was that this =
is not necessarily going to be of advantage in terms of causing a =
different UPDATE message to be generated. Particularly, since a ROUTE =
REFRESH would result in the same UPDATE packing (generally). To that =
end, what the requirements draft currently says is that:

- ROUTE-REFRESH is a reasonable way to re-run a consistency check =
between two peers (especially with start-of-refresh and end-of-refresh =
markers, since this means that we can purge any NLRI that we missed =
being withdrawn).
- If there is anything automatically triggered, it would be prudent to =
try and ensure that we can make as specific a request to the other =
speaker as possible - this is where mechanisms such as One-Time ORF were =
suggested, and is a similar space to that which would be solved through =
having the UPDATE-VERSION message described in =
draft-ietf-idr-enhanced-gr if one were to be able to have a REFRESH =
following a "last known good" position. Making a more specific request =
may result in a different UPDATE being generated, which may mean the =
error is fixed, but this is not guaranteed.
- If the selected best path changes on the remote speaker, then it will =
be re-advertised anyway - resulting in another chance to re-parse these =
NLRI.

So, certainly, if we relax the "correctness" requirement, then it would =
seem a good idea to have some means to be able to consistency check -- =
it's just a matter of ensuring that this is done in an effective and =
scalable manner, that does not affect normal BGP operation.

Cheers,
r.


On 28 Dec 2012, at 19:02, Brian Dickson <brian.peter.dickson@gmail.com> =
wrote:

> Here's a quick question, about possible mechanisms for determining =
whether or not we need to tear down a session.
>=20
> If there is a chance that some NLRI weren't properly decoded:
>=20
> - What about requesting (presuming the option was negotiated) a =
route-refresh, or a "confirm per-AFI-SAFI prefix list"?
>=20
> It's an expensive "parity check" but is one way of ensuring that we =
haven't missed a withdrawn prefix.
> If it isn't in the subsequently received list of prefixes, we missed =
it and should withdraw it.
> Everything else is a no-op - maybe we missed a new prefix (with new =
attribute?), but that is what "treat as withdraw" gets you.
>=20
> Just trying to make sure the cure isn't worse than the disease, in all =
situations.
> (Where "cure" is fail to withdraw because we "lost" the withdrawal =
because of malformed update, or "cure" is tear down the session.)
>=20
> Brian
>=20
> On Fri, Dec 28, 2012 at 1:04 PM, Chris Hall =
<chris.hall@highwayman.com> wrote:
> Rob Shakir wrote (on Fri 28-Dec-2012 at 13:10 +0000):
> >
> > (re: CCing IDR & GROW)
> >
> > On 28 Dec 2012, at 12:28, Chris Hall wrote:
> >
> > > Rob Shakir wrote (on Thu 27-Dec-2012 at 18:44):
> > >>
> > >> Any comments very welcome (to me or grow@).
>=20
> > > I'm afraid I still don't get it :-(  What am I missing ?
> > >
> > > UPDATE Message Length errors are Critical because they:
> > >
> > > (1) "result in cases whereby the NLRI attribute cannot
> > >      be correctly extracted".
> > >
> > > The implication is that a failure to extract all NLRI is Critical.
> > > Is that a requirement ?
>=20
> > If the NLRI cannot be determined, then this is a Critical error,
> > yes. I left the wording relatively open on whether this is *all*
> > NLRI, ...
>=20
> OK.  So, if the message is broken (in any way):
>=20
>   * if no NLRI can be found, that's a Critical Error.
>=20
>     That would include the case where some broken
>     attribute upstream of any MP_XXX obscures that
>     MP_XXX.
>=20
>   * if some NLRI can be found, that's a Non-Critical error.
>=20
>     Such NLRI as can be found are treated-as-withdraw.
>=20
>     AND any NLRI that may or may not have been lost, are
>     ignored.
>=20
>     AND whatever dangers lost NLRI may pose are covered
>     by the caveats in 4.1.
>=20
> Yes ?
>=20
> > ... as I am not sure that in the requirements draft we should
> > specify direct solutions to specific issues, to e.g., say how to
> > handle cases where MP_REACH_NLRI and MP_UNREACH_NLRI are in the same
> > message [this is a case that I do not believe is forbidden by
> > rfc2858 - if the working group could clarify whether this is
> > something that we feel the draft needs to handle or can explicitly
> > be omitted, then that would be appreciated].
>=20
> The RFCs also allow any mix of IPv4 Unicast Reachable/Unreachable in
> the body of the message, with or without MP_REACH_NLRI and/or
> MP_UNREACH_NLRI.  If we call each of these a "collection of NLRI",
> then there may be between 1 and 4 collections of NLRI in a message
> (ignoring the End-of-RIB pseudo-UPDATE).  It is an error to have more
> than one MP_REACH_NLRI attribute or more than one MP_UNREACH_NLRI
> attribute.  It is not clear to me whether those should be Critical
> Errors -- but while we are relaxing all error checking, I don't see
> why they should be.
>=20
> It is possible that most implementations actually only send one
> collection per message... so, as you suggest, on a you-know-and-I-know
> basis, we'all could ignore the "theoretical" issues of more than one
> collection in a message.
>=20
> But, this does not matter if the requirements allow lost NLRI to
> simply be ignored.  If the sender sends any MP_XXX first, then there
> is little chance that NLRI will be lost, and things work better.  If
> the sender follows the old rules, and particularly if it sends more
> than one collection at a time, then there is more chance that NLRI
> will be lost.
>=20
> > > Later:
> > >
> > >  (2) "All errors whereby the contained NLRI can be
> > >       extracted are referred to as Non-Critical".
> > >
> > > And that includes:
> > >
> > >  (3) "where the length of all path attributes contained
> > >       within the UPDATE does not correspond to the
> > >       total path attribute length."
> > >
> > > That is, at least, more explicit than
> > > draft-ietf-idr-error-handling-03, which glosses over (3).
> > >
> > > But if (3) is non-critical then there is some chance that
> > > some NLRI will not be extracted, which appears to violate
> > > (1) and (2).
>=20
> > Disclaimer: As I am sure that my comments previously have made
> > clear, I do not maintain a code base for a BGP daemon/implementation
> > - so please feel free to correct my logic below.
> >
> > I do not believe that (3) implies that the NLRI cannot be correctly
> > found.
>=20
> What (3) implies to me is that somewhere amongst the attributes
> something is broken.  As Jakob Heitz succinctly puts it: "Once you
> have a malformed update, NOTHING is certain."  In particular, you
> cannot be certain that all NLRI referred to by the sender can be found
> by the receiver.
>=20
> > If the sum of total length is incorrect, then we can still
> > extract the individual attributes - we just find that there is not
> > enough data to fill the overall length we were told and/or we have
> > too much attribute data compared to the total attribute length.
>=20
> This is precisely where I think the requirements come unstuck.  There
> is almost no redundancy in the attribute encoding.  And almost all the
> redundancy there is is discarded by the new-form-error-handling.  So,
> if the sum of the attribute lengths is incorrect, then (inter alia)
> the receiver CANNOT know whether the attributes it has extracted are
> the attributes the sender intended to send.  In particular, the
> receiver cannot know it has extracted all NLRI (except in the,
> probably obscure, case of having extracted both MP_REACH_NLRI and
> MP_UNREACH_NLRI).  "Once you have a malformed update, NOTHING is
> certain."
>=20
> Conversely, if the sum of attribute lengths is correct, then there is
> a fighting chance that the attributes received are the attributes
> sent.  But, if one of those attributes is (say) nominally an
> ATOMIC_AGGREGATE which is apparently 200 octets long, that might be a
> worry.
>=20
> ...
> > > Then (4) "In order to maximise the number of cases whereby the
> > > NLRI attributes [plural, now, BTW] can be reliably extracted
> > > from a received message...".  Ah.  So it is not a Critical
> > > Error if "the NLRI attribute cannot be correctly extracted".
>=20
> > No - it is a Critical error if we cannot extract the NLRI. This
> > recommendation is to give an increased chance that the NLRI can be
> > extracted as per the IDR error handling draft. This then (by virtue
> > of resulting in the NLRI being extracted) minimises the number of
> > cases that result in a Critical error.
>=20
> By this do you mean "all" NLRI or only "some" NLRI ?  As above.
>=20
> > The plural here is to reflect
> > that the existence of >1 type of NLRI attribute.
>=20
> That would imply that it should be "NLRI attributes" throughout ?
> Though in most places where the draft speaks of "NLRI attribute" I
> think it means one, some, any or all of the "collections of NLRI" (as
> defined above).
>=20
> > > For me the requirement remains "conflicted".  On the one hand it
> > > seems to say that it is a Critical Error if the NLRI cannot be
> > > extracted and parsed.  On the other it seems to say it's OK if
> > > you cannot extract some NLRI.
>=20
> > If you'll forgive me for removing a significant proportion of your
> > message, I think that we need to take another step back here. It
> > seems to me that the key question that you are highlighting is "What
> > level of confidence do we need to have before we declare that the
> > NLRI cannot be extracted?" -- do you agree?
>=20
> Absolutely.
>=20
> > =46rom an operator perspective, I would like to compromise =
*certainty*
> > for *robustness*. You are right, we are compromising correctness
> > here, we might end up withdrawing an incorrect NLRI and impacting
> > service operation for that prefix - however, it is somewhat
> > preferable to me to withdraw a a subset of the NLRI incorrectly,
> > rather than impact all NLRI in one single action. We clearly need to
> > provide some bounds on how much we compromise the certainty (and
> > live within the realms of possibility, such that we are not just
> > taking a shot in the dark). This is what the definitions of Critical
> > and Non-Critical within the document are intended to provide. Once
> > again I will refer to the requirement that there is a balance
> > between correctness and robustness - rather than a locally risk
> > averse approach that results in harmful wider behaviour.
>=20
> Where one can extract the NLRI from a broken message, then
> treat-as-withdraw is AFAICS no worse for the NLRI in question than
> session-reset, and hugely better for all the other NLRI received from
> the peer.
>=20
> Where not *all* the NLRI in a message have been extracted, then we
> take a step beyond withdrawing something which should not have been
> withdrawn.  For each of those NLRI the receiver will (unknowingly) do
> one of:
>=20
>   (a) continue using a route which the sender has
>       withdrawn;
>=20
>   (b) continue to use an out of date version of a route
>       which the sender has changed in some way, possibly
>       materially;
>=20
>   (c) to fail to use a new route the sender has now made
>       available.
>=20
> Of these (a) looks serious and (b) could be, but certainly the effect
> is different from session-reset; while (c) is, essentially,
> treat-as-withdraw.  If some risk "lost routes" is taken, then the
> impact of that clearly must be weighed against the impact of
> session-reset.  I have not found a discussion of the impact of "lost
> routes" in the draft, so I have no idea what the trade-off is, here.
>=20
> For completeness, if all forms and degrees of attribute broken-ness
> are required to be acceptable, then there are cases where the receiver
> cannot know whether *all* NLRI have been extracted, or not.
>=20
> > Is it acceptable that we leave this as guidance within the
> > requirements? If not, please could you suggest how the definitions
> > of Critical/Non-Critical could be altered to address your concerns?
>=20
> Without a change at the sender end, the problem is that: "Once you
> have a malformed update, NOTHING is certain."
>=20
> So, as I said earlier, the fundamental question is:
>=20
>   (F) is it OK to continue with a session after processing
>       a message which may have contained some NLRI which
>       could not be extracted ?
>=20
> As you say, the question hinges on the "which may have contained".  As
> above, it seems to me that the draft skates over the issue of "lost
> routes".
>=20
> It also hinges on the operational impact of such "lost routes", on
> which I am unqualified to pronounce.
>=20
> It seems to me there is a range of possibilities:
>=20
>   1) at one end of the spectrum, if "lost routes" are to
>      be avoided at all costs, then the rules for parsing
>      attributes need to be tight and without a change at
>      the sender end, what is achievable is constrained.
>=20
>   2) if "lost routes" are acceptable in the "theoretical"
>      cases, but you-know-and-I-know that most of the
>      time the case will not arise...
>=20
>      ...or "lost routes" are acceptable provided some
>      defined steps are taken to identify as much NLRI
>      as is reasonably possible...
>=20
>      ...then the requirements need to (a) make the case
>      and (b) describe the trade-off (so that designs
>      made to follow the requirements can be judged in
>      those terms).
>=20
>   3) at the other end of the spectrum, if "lost routes"
>      are always preferable to session-reset, then this
>      opens up an even looser approach to UPDATE message
>      handling, in which practically nothing need cause
>      a session-reset -- ie, practically nothing is a
>      Critical Error.
>=20
> And, as I have suggested in previous discussions, it is possible to
> consider case (2) in two parts:
>=20
>   a) where the sender is unchanged.
>=20
>      In which case the issue is the degree of attribute
>      broken-ness vs the likelihood of "lost routes".
>=20
>      If this is viewed as an intermediate step, then
>      perhaps the requirements could err on the side of
>      safety ?
>=20
>      Perhaps this can be addressed by a "knob" allowing
>      more or less strictness in the parsing of
>      attributes ?
>=20
>   b) where the sender is changed.
>=20
>      In which case the issue ought to be moot, because
>      the receiver will then be able to know that there
>      are no "lost routes".
>=20
> > I would also appreciate further input from IDR as to whether this is
> > sufficient requirement from GROW to allow a solution document to be
> > written?
>=20
> The new requirements specify only Critical (session-reset) and
> Non-Critical (treat-as-withdraw) Errors.  This appears to rule out the
> "attribute discard" option in draft-ietf-idr-error-handling-03.  Is
> that intended ?
>=20
> Chris
>=20
> _______________________________________________
> Idr mailing list
> Idr@ietf.org
> https://www.ietf.org/mailman/listinfo/idr
>=20


--Apple-Mail=_81DB352D-E036-47D8-9A50-6E15BB53EFB2
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=iso-8859-1

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html =
charset=3Diso-8859-1"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Hi =
Brian,<div><br></div><div>We discussed this at some length previously =
(the -00 of the requirements draft discussed automatically triggering =
some form of means to recover from the error). I think that there is =
some merit in doing so, but there are clearly also scaling =
considerations around this (in terms of requiring both BGP speakers to =
re-generate/re-process UPDATEs).&nbsp;</div><div><br></div><div>The =
consensus that was reached in the previous discussion was that this is =
not necessarily going to be of advantage in terms of causing a different =
UPDATE message to be generated. Particularly, since a ROUTE REFRESH =
would result in the same UPDATE packing (generally). To that end, what =
the requirements draft currently says is =
that:</div><div><br></div><div>- ROUTE-REFRESH is a reasonable way to =
re-run a consistency check between two peers (especially with =
start-of-refresh and end-of-refresh markers, since this means that we =
can purge any NLRI that we missed being withdrawn).</div><div>- If there =
is anything automatically triggered, it would be prudent to try and =
ensure that we can make as specific a request to the other speaker as =
possible - this is where mechanisms such as One-Time ORF were suggested, =
and is a similar space to that which would be solved through having the =
UPDATE-VERSION message described in draft-ietf-idr-enhanced-gr if one =
were to be able to have a REFRESH following a "last known good" =
position. Making a more specific request may result in a different =
UPDATE being generated, which may mean the error is fixed, but this is =
not guaranteed.</div><div>- If the selected best path changes on the =
remote speaker, then it will be re-advertised anyway - resulting in =
another chance to re-parse these NLRI.</div><div><br></div><div>So, =
certainly, if we relax the "correctness" requirement, then it would seem =
a good idea to have some means to be able to consistency check -- it's =
just a matter of ensuring that this is done in an effective and scalable =
manner, that does not affect normal BGP =
operation.</div><div><br></div><div>Cheers,</div><div>r.</div><div><br></d=
iv><div><br><div><div>On 28 Dec 2012, at 19:02, Brian Dickson &lt;<a =
href=3D"mailto:brian.peter.dickson@gmail.com">brian.peter.dickson@gmail.co=
m</a>&gt; wrote:</div><br class=3D"Apple-interchange-newline"><blockquote =
type=3D"cite">Here's a quick question, about possible mechanisms for =
determining whether or not we need to tear down a =
session.<div><br></div><div>If there is a chance that some NLRI weren't =
properly decoded:</div><div><br></div>
<div>- What about requesting (presuming the option was negotiated) a =
route-refresh, or a "confirm per-AFI-SAFI prefix =
list"?</div><div><br></div><div>It's an expensive "parity check" but is =
one way of ensuring that we haven't missed a withdrawn prefix.</div>
<div>If it isn't in the subsequently received list of prefixes, we =
missed it and should withdraw it.</div><div>Everything else is a no-op - =
maybe we missed a new prefix (with new attribute?), but that is what =
"treat as withdraw" gets you.</div>
<div><br></div><div>Just trying to make sure the cure isn't worse than =
the disease, in all situations.</div><div>(Where "cure" is fail to =
withdraw because we "lost" the withdrawal because of malformed update, =
or "cure" is tear down the session.)</div>
<div><br></div><div>Brian<br><br><div class=3D"gmail_quote">On Fri, Dec =
28, 2012 at 1:04 PM, Chris Hall <span dir=3D"ltr">&lt;<a =
href=3D"mailto:chris.hall@highwayman.com" =
target=3D"_blank">chris.hall@highwayman.com</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 =
.8ex;border-left:1px #ccc solid;padding-left:1ex">Rob Shakir wrote (on =
Fri 28-Dec-2012 at 13:10 +0000):<br>
<div class=3D"im">&gt;<br>
&gt; (re: CCing IDR &amp; GROW)<br>
&gt;<br>
&gt; On 28 Dec 2012, at 12:28, Chris Hall wrote:<br>
&gt;<br>
&gt; &gt; Rob Shakir wrote (on Thu 27-Dec-2012 at 18:44):<br>
&gt; &gt;&gt;<br>
&gt; &gt;&gt; Any comments very welcome (to me or grow@).<br>
<br>
&gt; &gt; I'm afraid I still don't get it :-( &nbsp;What am I missing =
?<br>
&gt; &gt;<br>
&gt; &gt; UPDATE Message Length errors are Critical because they:<br>
&gt; &gt;<br>
&gt; &gt; (1) "result in cases whereby the NLRI attribute cannot<br>
&gt; &gt; &nbsp; &nbsp; &nbsp;be correctly extracted".<br>
&gt; &gt;<br>
&gt; &gt; The implication is that a failure to extract all NLRI is =
Critical.<br>
&gt; &gt; Is that a requirement ?<br>
<br>
&gt; If the NLRI cannot be determined, then this is a Critical =
error,<br>
&gt; yes. I left the wording relatively open on whether this is =
*all*<br>
</div>&gt; NLRI, ...<br>
<br>
OK. &nbsp;So, if the message is broken (in any way):<br>
<br>
&nbsp; * if no NLRI can be found, that's a Critical Error.<br>
<br>
&nbsp; &nbsp; That would include the case where some broken<br>
&nbsp; &nbsp; attribute upstream of any MP_XXX obscures that<br>
&nbsp; &nbsp; MP_XXX.<br>
<br>
&nbsp; * if some NLRI can be found, that's a Non-Critical error.<br>
<br>
&nbsp; &nbsp; Such NLRI as can be found are treated-as-withdraw.<br>
<br>
&nbsp; &nbsp; AND any NLRI that may or may not have been lost, are<br>
&nbsp; &nbsp; ignored.<br>
<br>
&nbsp; &nbsp; AND whatever dangers lost NLRI may pose are covered<br>
&nbsp; &nbsp; by the caveats in 4.1.<br>
<br>
Yes ?<br>
<br>
&gt; ... as I am not sure that in the requirements draft we should<br>
<div class=3D"im">&gt; specify direct solutions to specific issues, to =
e.g., say how to<br>
&gt; handle cases where MP_REACH_NLRI and MP_UNREACH_NLRI are in the =
same<br>
&gt; message [this is a case that I do not believe is forbidden by<br>
&gt; rfc2858 - if the working group could clarify whether this is<br>
&gt; something that we feel the draft needs to handle or can =
explicitly<br>
&gt; be omitted, then that would be appreciated].<br>
<br>
</div>The RFCs also allow any mix of IPv4 Unicast Reachable/Unreachable =
in<br>
the body of the message, with or without MP_REACH_NLRI and/or<br>
MP_UNREACH_NLRI. &nbsp;If we call each of these a "collection of =
NLRI",<br>
then there may be between 1 and 4 collections of NLRI in a message<br>
(ignoring the End-of-RIB pseudo-UPDATE). &nbsp;It is an error to have =
more<br>
than one MP_REACH_NLRI attribute or more than one MP_UNREACH_NLRI<br>
attribute. &nbsp;It is not clear to me whether those should be =
Critical<br>
Errors -- but while we are relaxing all error checking, I don't see<br>
why they should be.<br>
<br>
It is possible that most implementations actually only send one<br>
collection per message... so, as you suggest, on a =
you-know-and-I-know<br>
basis, we'all could ignore the "theoretical" issues of more than one<br>
collection in a message.<br>
<br>
But, this does not matter if the requirements allow lost NLRI to<br>
simply be ignored. &nbsp;If the sender sends any MP_XXX first, then =
there<br>
is little chance that NLRI will be lost, and things work better. =
&nbsp;If<br>
the sender follows the old rules, and particularly if it sends more<br>
than one collection at a time, then there is more chance that NLRI<br>
will be lost.<br>
<div class=3D"im"><br>
&gt; &gt; Later:<br>
&gt; &gt;<br>
&gt; &gt; &nbsp;(2) "All errors whereby the contained NLRI can be<br>
&gt; &gt; &nbsp; &nbsp; &nbsp; extracted are referred to as =
Non-Critical".<br>
&gt; &gt;<br>
&gt; &gt; And that includes:<br>
&gt; &gt;<br>
&gt; &gt; &nbsp;(3) "where the length of all path attributes =
contained<br>
&gt; &gt; &nbsp; &nbsp; &nbsp; within the UPDATE does not correspond to =
the<br>
&gt; &gt; &nbsp; &nbsp; &nbsp; total path attribute length."<br>
&gt; &gt;<br>
&gt; &gt; That is, at least, more explicit than<br>
&gt; &gt; draft-ietf-idr-error-handling-03, which glosses over (3).<br>
&gt; &gt;<br>
&gt; &gt; But if (3) is non-critical then there is some chance that<br>
&gt; &gt; some NLRI will not be extracted, which appears to violate<br>
&gt; &gt; (1) and (2).<br>
<br>
&gt; Disclaimer: As I am sure that my comments previously have made<br>
&gt; clear, I do not maintain a code base for a BGP =
daemon/implementation<br>
&gt; - so please feel free to correct my logic below.<br>
&gt;<br>
&gt; I do not believe that (3) implies that the NLRI cannot be =
correctly<br>
&gt; found.<br>
<br>
</div>What (3) implies to me is that somewhere amongst the =
attributes<br>
something is broken. &nbsp;As Jakob Heitz succinctly puts it: "Once =
you<br>
have a malformed update, NOTHING is certain." &nbsp;In particular, =
you<br>
cannot be certain that all NLRI referred to by the sender can be =
found<br>
by the receiver.<br>
<div class=3D"im"><br>
&gt; If the sum of total length is incorrect, then we can still<br>
&gt; extract the individual attributes - we just find that there is =
not<br>
&gt; enough data to fill the overall length we were told and/or we =
have<br>
&gt; too much attribute data compared to the total attribute length.<br>
<br>
</div>This is precisely where I think the requirements come unstuck. =
&nbsp;There<br>
is almost no redundancy in the attribute encoding. &nbsp;And almost all =
the<br>
redundancy there is is discarded by the new-form-error-handling. =
&nbsp;So,<br>
if the sum of the attribute lengths is incorrect, then (inter alia)<br>
the receiver CANNOT know whether the attributes it has extracted are<br>
the attributes the sender intended to send. &nbsp;In particular, the<br>
receiver cannot know it has extracted all NLRI (except in the,<br>
probably obscure, case of having extracted both MP_REACH_NLRI and<br>
MP_UNREACH_NLRI). &nbsp;"Once you have a malformed update, NOTHING =
is<br>
certain."<br>
<br>
Conversely, if the sum of attribute lengths is correct, then there =
is<br>
a fighting chance that the attributes received are the attributes<br>
sent. &nbsp;But, if one of those attributes is (say) nominally an<br>
ATOMIC_AGGREGATE which is apparently 200 octets long, that might be =
a<br>
worry.<br>
<br>
...<br>
<div class=3D"im">&gt; &gt; Then (4) "In order to maximise the number of =
cases whereby the<br>
&gt; &gt; NLRI attributes [plural, now, BTW] can be reliably =
extracted<br>
&gt; &gt; from a received message...". &nbsp;Ah. &nbsp;So it is not a =
Critical<br>
&gt; &gt; Error if "the NLRI attribute cannot be correctly =
extracted".<br>
<br>
&gt; No - it is a Critical error if we cannot extract the NLRI. This<br>
&gt; recommendation is to give an increased chance that the NLRI can =
be<br>
&gt; extracted as per the IDR error handling draft. This then (by =
virtue<br>
&gt; of resulting in the NLRI being extracted) minimises the number =
of<br>
&gt; cases that result in a Critical error.<br>
<br>
</div>By this do you mean "all" NLRI or only "some" NLRI ? &nbsp;As =
above.<br>
<div class=3D"im"><br>
&gt; The plural here is to reflect<br>
&gt; that the existence of &gt;1 type of NLRI attribute.<br>
<br>
</div>That would imply that it should be "NLRI attributes" throughout =
?<br>
Though in most places where the draft speaks of "NLRI attribute" I<br>
think it means one, some, any or all of the "collections of NLRI" =
(as<br>
defined above).<br>
<div class=3D"im"><br>
&gt; &gt; For me the requirement remains "conflicted". &nbsp;On the one =
hand it<br>
&gt; &gt; seems to say that it is a Critical Error if the NLRI cannot =
be<br>
&gt; &gt; extracted and parsed. &nbsp;On the other it seems to say it's =
OK if<br>
&gt; &gt; you cannot extract some NLRI.<br>
<br>
&gt; If you'll forgive me for removing a significant proportion of =
your<br>
&gt; message, I think that we need to take another step back here. =
It<br>
&gt; seems to me that the key question that you are highlighting is =
"What<br>
&gt; level of confidence do we need to have before we declare that =
the<br>
&gt; NLRI cannot be extracted?" -- do you agree?<br>
<br>
</div>Absolutely.<br>
<div class=3D"im"><br>
&gt; =46rom an operator perspective, I would like to compromise =
*certainty*<br>
&gt; for *robustness*. You are right, we are compromising =
correctness<br>
&gt; here, we might end up withdrawing an incorrect NLRI and =
impacting<br>
&gt; service operation for that prefix - however, it is somewhat<br>
&gt; preferable to me to withdraw a a subset of the NLRI =
incorrectly,<br>
&gt; rather than impact all NLRI in one single action. We clearly need =
to<br>
&gt; provide some bounds on how much we compromise the certainty =
(and<br>
&gt; live within the realms of possibility, such that we are not =
just<br>
&gt; taking a shot in the dark). This is what the definitions of =
Critical<br>
&gt; and Non-Critical within the document are intended to provide. =
Once<br>
&gt; again I will refer to the requirement that there is a balance<br>
&gt; between correctness and robustness - rather than a locally risk<br>
&gt; averse approach that results in harmful wider behaviour.<br>
<br>
</div>Where one can extract the NLRI from a broken message, then<br>
treat-as-withdraw is AFAICS no worse for the NLRI in question than<br>
session-reset, and hugely better for all the other NLRI received =
from<br>
the peer.<br>
<br>
Where not *all* the NLRI in a message have been extracted, then we<br>
take a step beyond withdrawing something which should not have been<br>
withdrawn. &nbsp;For each of those NLRI the receiver will (unknowingly) =
do<br>
one of:<br>
<br>
&nbsp; (a) continue using a route which the sender has<br>
&nbsp; &nbsp; &nbsp; withdrawn;<br>
<br>
&nbsp; (b) continue to use an out of date version of a route<br>
&nbsp; &nbsp; &nbsp; which the sender has changed in some way, =
possibly<br>
&nbsp; &nbsp; &nbsp; materially;<br>
<br>
&nbsp; (c) to fail to use a new route the sender has now made<br>
&nbsp; &nbsp; &nbsp; available.<br>
<br>
Of these (a) looks serious and (b) could be, but certainly the =
effect<br>
is different from session-reset; while (c) is, essentially,<br>
treat-as-withdraw. &nbsp;If some risk "lost routes" is taken, then =
the<br>
impact of that clearly must be weighed against the impact of<br>
session-reset. &nbsp;I have not found a discussion of the impact of =
"lost<br>
routes" in the draft, so I have no idea what the trade-off is, here.<br>
<br>
For completeness, if all forms and degrees of attribute broken-ness<br>
are required to be acceptable, then there are cases where the =
receiver<br>
cannot know whether *all* NLRI have been extracted, or not.<br>
<div class=3D"im"><br>
&gt; Is it acceptable that we leave this as guidance within the<br>
&gt; requirements? If not, please could you suggest how the =
definitions<br>
&gt; of Critical/Non-Critical could be altered to address your =
concerns?<br>
<br>
</div>Without a change at the sender end, the problem is that: "Once =
you<br>
have a malformed update, NOTHING is certain."<br>
<br>
So, as I said earlier, the fundamental question is:<br>
<br>
&nbsp; (F) is it OK to continue with a session after processing<br>
&nbsp; &nbsp; &nbsp; a message which may have contained some NLRI =
which<br>
&nbsp; &nbsp; &nbsp; could not be extracted ?<br>
<br>
As you say, the question hinges on the "which may have contained". =
&nbsp;As<br>
above, it seems to me that the draft skates over the issue of "lost<br>
routes".<br>
<br>
It also hinges on the operational impact of such "lost routes", on<br>
which I am unqualified to pronounce.<br>
<br>
It seems to me there is a range of possibilities:<br>
<br>
&nbsp; 1) at one end of the spectrum, if "lost routes" are to<br>
&nbsp; &nbsp; &nbsp;be avoided at all costs, then the rules for =
parsing<br>
&nbsp; &nbsp; &nbsp;attributes need to be tight and without a change =
at<br>
&nbsp; &nbsp; &nbsp;the sender end, what is achievable is =
constrained.<br>
<br>
&nbsp; 2) if "lost routes" are acceptable in the "theoretical"<br>
&nbsp; &nbsp; &nbsp;cases, but you-know-and-I-know that most of the<br>
&nbsp; &nbsp; &nbsp;time the case will not arise...<br>
<br>
&nbsp; &nbsp; &nbsp;...or "lost routes" are acceptable provided some<br>
&nbsp; &nbsp; &nbsp;defined steps are taken to identify as much NLRI<br>
&nbsp; &nbsp; &nbsp;as is reasonably possible...<br>
<br>
&nbsp; &nbsp; &nbsp;...then the requirements need to (a) make the =
case<br>
&nbsp; &nbsp; &nbsp;and (b) describe the trade-off (so that designs<br>
&nbsp; &nbsp; &nbsp;made to follow the requirements can be judged in<br>
&nbsp; &nbsp; &nbsp;those terms).<br>
<br>
&nbsp; 3) at the other end of the spectrum, if "lost routes"<br>
&nbsp; &nbsp; &nbsp;are always preferable to session-reset, then =
this<br>
&nbsp; &nbsp; &nbsp;opens up an even looser approach to UPDATE =
message<br>
&nbsp; &nbsp; &nbsp;handling, in which practically nothing need =
cause<br>
&nbsp; &nbsp; &nbsp;a session-reset -- ie, practically nothing is a<br>
&nbsp; &nbsp; &nbsp;Critical Error.<br>
<br>
And, as I have suggested in previous discussions, it is possible to<br>
consider case (2) in two parts:<br>
<br>
&nbsp; a) where the sender is unchanged.<br>
<br>
&nbsp; &nbsp; &nbsp;In which case the issue is the degree of =
attribute<br>
&nbsp; &nbsp; &nbsp;broken-ness vs the likelihood of "lost routes".<br>
<br>
&nbsp; &nbsp; &nbsp;If this is viewed as an intermediate step, then<br>
&nbsp; &nbsp; &nbsp;perhaps the requirements could err on the side =
of<br>
&nbsp; &nbsp; &nbsp;safety ?<br>
<br>
&nbsp; &nbsp; &nbsp;Perhaps this can be addressed by a "knob" =
allowing<br>
&nbsp; &nbsp; &nbsp;more or less strictness in the parsing of<br>
&nbsp; &nbsp; &nbsp;attributes ?<br>
<br>
&nbsp; b) where the sender is changed.<br>
<br>
&nbsp; &nbsp; &nbsp;In which case the issue ought to be moot, =
because<br>
&nbsp; &nbsp; &nbsp;the receiver will then be able to know that =
there<br>
&nbsp; &nbsp; &nbsp;are no "lost routes".<br>
<div class=3D"im"><br>
&gt; I would also appreciate further input from IDR as to whether this =
is<br>
&gt; sufficient requirement from GROW to allow a solution document to =
be<br>
&gt; written?<br>
<br>
</div>The new requirements specify only Critical (session-reset) and<br>
Non-Critical (treat-as-withdraw) Errors. &nbsp;This appears to rule out =
the<br>
"attribute discard" option in draft-ietf-idr-error-handling-03. =
&nbsp;Is<br>
that intended ?<br>
<span class=3D"HOEnZb"><font color=3D"#888888"><br>
Chris<br>
</font></span><div class=3D"HOEnZb"><div class=3D"h5"><br>
_______________________________________________<br>
Idr mailing list<br>
<a href=3D"mailto:Idr@ietf.org">Idr@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/idr" =
target=3D"_blank">https://www.ietf.org/mailman/listinfo/idr</a><br>
</div></div></blockquote></div><br></div>
</blockquote></div><br></div></body></html>=

--Apple-Mail=_81DB352D-E036-47D8-9A50-6E15BB53EFB2--

From chris.hall@highwayman.com  Fri Dec 28 10:04:23 2012
Return-Path: <chris.hall@highwayman.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 96CD221F8BC9; Fri, 28 Dec 2012 10:04:23 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 2.874
X-Spam-Level: **
X-Spam-Status: No, score=2.874 tagged_above=-999 required=5 tests=[AWL=-2.414,  BAYES_00=-2.599, GB_SUMOF=5, HELO_MISMATCH_UK=1.749, HOST_MISMATCH_NET=0.311, J_CHICKENPOX_23=0.6, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HsCChMzzEC3Q; Fri, 28 Dec 2012 10:04:22 -0800 (PST)
Received: from smtp.demon.co.uk (mdfmta010.mxout.tch.inty.net [91.221.169.51]) by ietfa.amsl.com (Postfix) with ESMTP id F27F121F8923; Fri, 28 Dec 2012 10:04:17 -0800 (PST)
Received: from mdfmta010.tch.inty.net (unknown [127.0.0.1]) by mdfmta010.tch.inty.net (Postfix) with ESMTP id 5EFB9400594; Fri, 28 Dec 2012 18:04:16 +0000 (GMT)
Received: from mdfmta010.tch.inty.net (unknown [127.0.0.1])	by mdfmta010.tch.inty.net (Postfix) with ESMTP id 29CE8400590; Fri, 28 Dec 2012 18:04:16 +0000 (GMT)
Received: from hestia.halldom.com (unknown [80.177.246.130])	(using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits))	(No client certificate requested)	by mdfmta010.tch.inty.net (Postfix) with ESMTP; Fri, 28 Dec 2012 18:04:15 +0000 (GMT)
Received: from hyperion.halldom.com ([80.177.246.170] helo=HYPERION)	by hestia.halldom.com with esmtpsa (TLSv1:AES128-SHA:128)	(Exim 4.76)	(envelope-from <chris.hall@highwayman.com>)	id 1ToeI2-0000vV-S7; Fri, 28 Dec 2012 18:04:14 +0000
From: "Chris Hall" <chris.hall@highwayman.com>
To: "'Rob Shakir'" <rjs@rob.sh>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh>
In-Reply-To: <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh>
Date: Fri, 28 Dec 2012 18:04:09 -0000
Organization: Highwayman
Message-ID: <028701cde525$bf262070$3d726150$@highwayman.com>
MIME-Version: 1.0
Content-Type: text/plain;	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQFuOjybZ81Jyr3vkBQVN4QFzf3TcAHhILBDmN6LQYA=
Content-Language: en-gb
X-MDF-HostID: 19
X-Mailman-Approved-At: Sat, 29 Dec 2012 19:26:25 -0800
Cc: idr@ietf.org, grow@ietf.org
Subject: Re: [GROW] [Idr] Fwd: I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 28 Dec 2012 18:04:23 -0000

Rob Shakir wrote (on Fri 28-Dec-2012 at 13:10 +0000):
> 
> (re: CCing IDR & GROW)
> 
> On 28 Dec 2012, at 12:28, Chris Hall wrote:
> 
> > Rob Shakir wrote (on Thu 27-Dec-2012 at 18:44):
> >>
> >> Any comments very welcome (to me or grow@).

> > I'm afraid I still don't get it :-(  What am I missing ?
> >
> > UPDATE Message Length errors are Critical because they:
> > 
> > (1) "result in cases whereby the NLRI attribute cannot
> >      be correctly extracted".
> >
> > The implication is that a failure to extract all NLRI is Critical.
> > Is that a requirement ?
 
> If the NLRI cannot be determined, then this is a Critical error,
> yes. I left the wording relatively open on whether this is *all*
> NLRI, ...

OK.  So, if the message is broken (in any way):

  * if no NLRI can be found, that's a Critical Error.

    That would include the case where some broken
    attribute upstream of any MP_XXX obscures that
    MP_XXX. 

  * if some NLRI can be found, that's a Non-Critical error.

    Such NLRI as can be found are treated-as-withdraw.

    AND any NLRI that may or may not have been lost, are
    ignored.

    AND whatever dangers lost NLRI may pose are covered
    by the caveats in 4.1.

Yes ?

> ... as I am not sure that in the requirements draft we should
> specify direct solutions to specific issues, to e.g., say how to
> handle cases where MP_REACH_NLRI and MP_UNREACH_NLRI are in the same
> message [this is a case that I do not believe is forbidden by
> rfc2858 - if the working group could clarify whether this is
> something that we feel the draft needs to handle or can explicitly
> be omitted, then that would be appreciated].

The RFCs also allow any mix of IPv4 Unicast Reachable/Unreachable in
the body of the message, with or without MP_REACH_NLRI and/or
MP_UNREACH_NLRI.  If we call each of these a "collection of NLRI",
then there may be between 1 and 4 collections of NLRI in a message
(ignoring the End-of-RIB pseudo-UPDATE).  It is an error to have more
than one MP_REACH_NLRI attribute or more than one MP_UNREACH_NLRI
attribute.  It is not clear to me whether those should be Critical
Errors -- but while we are relaxing all error checking, I don't see
why they should be.

It is possible that most implementations actually only send one
collection per message... so, as you suggest, on a you-know-and-I-know
basis, we'all could ignore the "theoretical" issues of more than one
collection in a message.

But, this does not matter if the requirements allow lost NLRI to
simply be ignored.  If the sender sends any MP_XXX first, then there
is little chance that NLRI will be lost, and things work better.  If
the sender follows the old rules, and particularly if it sends more
than one collection at a time, then there is more chance that NLRI
will be lost.

> > Later:
> >
> >  (2) "All errors whereby the contained NLRI can be
> >       extracted are referred to as Non-Critical".
> >
> > And that includes:
> >
> >  (3) "where the length of all path attributes contained
> >       within the UPDATE does not correspond to the
> >       total path attribute length."
> >
> > That is, at least, more explicit than
> > draft-ietf-idr-error-handling-03, which glosses over (3).
> >
> > But if (3) is non-critical then there is some chance that
> > some NLRI will not be extracted, which appears to violate
> > (1) and (2).

> Disclaimer: As I am sure that my comments previously have made
> clear, I do not maintain a code base for a BGP daemon/implementation
> - so please feel free to correct my logic below.
> 
> I do not believe that (3) implies that the NLRI cannot be correctly
> found.

What (3) implies to me is that somewhere amongst the attributes
something is broken.  As Jakob Heitz succinctly puts it: "Once you
have a malformed update, NOTHING is certain."  In particular, you
cannot be certain that all NLRI referred to by the sender can be found
by the receiver.

> If the sum of total length is incorrect, then we can still
> extract the individual attributes - we just find that there is not
> enough data to fill the overall length we were told and/or we have
> too much attribute data compared to the total attribute length.

This is precisely where I think the requirements come unstuck.  There
is almost no redundancy in the attribute encoding.  And almost all the
redundancy there is is discarded by the new-form-error-handling.  So,
if the sum of the attribute lengths is incorrect, then (inter alia)
the receiver CANNOT know whether the attributes it has extracted are
the attributes the sender intended to send.  In particular, the
receiver cannot know it has extracted all NLRI (except in the,
probably obscure, case of having extracted both MP_REACH_NLRI and
MP_UNREACH_NLRI).  "Once you have a malformed update, NOTHING is
certain."

Conversely, if the sum of attribute lengths is correct, then there is
a fighting chance that the attributes received are the attributes
sent.  But, if one of those attributes is (say) nominally an
ATOMIC_AGGREGATE which is apparently 200 octets long, that might be a
worry.

...
> > Then (4) "In order to maximise the number of cases whereby the
> > NLRI attributes [plural, now, BTW] can be reliably extracted
> > from a received message...".  Ah.  So it is not a Critical
> > Error if "the NLRI attribute cannot be correctly extracted".

> No - it is a Critical error if we cannot extract the NLRI. This
> recommendation is to give an increased chance that the NLRI can be
> extracted as per the IDR error handling draft. This then (by virtue
> of resulting in the NLRI being extracted) minimises the number of
> cases that result in a Critical error.

By this do you mean "all" NLRI or only "some" NLRI ?  As above.

> The plural here is to reflect
> that the existence of >1 type of NLRI attribute.

That would imply that it should be "NLRI attributes" throughout ?
Though in most places where the draft speaks of "NLRI attribute" I
think it means one, some, any or all of the "collections of NLRI" (as
defined above).
 
> > For me the requirement remains "conflicted".  On the one hand it
> > seems to say that it is a Critical Error if the NLRI cannot be
> > extracted and parsed.  On the other it seems to say it's OK if
> > you cannot extract some NLRI.

> If you'll forgive me for removing a significant proportion of your
> message, I think that we need to take another step back here. It
> seems to me that the key question that you are highlighting is "What
> level of confidence do we need to have before we declare that the
> NLRI cannot be extracted?" -- do you agree?

Absolutely.

> From an operator perspective, I would like to compromise *certainty*
> for *robustness*. You are right, we are compromising correctness
> here, we might end up withdrawing an incorrect NLRI and impacting
> service operation for that prefix - however, it is somewhat
> preferable to me to withdraw a a subset of the NLRI incorrectly,
> rather than impact all NLRI in one single action. We clearly need to
> provide some bounds on how much we compromise the certainty (and
> live within the realms of possibility, such that we are not just
> taking a shot in the dark). This is what the definitions of Critical
> and Non-Critical within the document are intended to provide. Once
> again I will refer to the requirement that there is a balance
> between correctness and robustness - rather than a locally risk
> averse approach that results in harmful wider behaviour.

Where one can extract the NLRI from a broken message, then
treat-as-withdraw is AFAICS no worse for the NLRI in question than
session-reset, and hugely better for all the other NLRI received from
the peer.

Where not *all* the NLRI in a message have been extracted, then we
take a step beyond withdrawing something which should not have been
withdrawn.  For each of those NLRI the receiver will (unknowingly) do
one of:

  (a) continue using a route which the sender has
      withdrawn;

  (b) continue to use an out of date version of a route
      which the sender has changed in some way, possibly
      materially;

  (c) to fail to use a new route the sender has now made
      available.

Of these (a) looks serious and (b) could be, but certainly the effect
is different from session-reset; while (c) is, essentially,
treat-as-withdraw.  If some risk "lost routes" is taken, then the
impact of that clearly must be weighed against the impact of
session-reset.  I have not found a discussion of the impact of "lost
routes" in the draft, so I have no idea what the trade-off is, here.

For completeness, if all forms and degrees of attribute broken-ness
are required to be acceptable, then there are cases where the receiver
cannot know whether *all* NLRI have been extracted, or not.

> Is it acceptable that we leave this as guidance within the
> requirements? If not, please could you suggest how the definitions
> of Critical/Non-Critical could be altered to address your concerns?

Without a change at the sender end, the problem is that: "Once you
have a malformed update, NOTHING is certain."

So, as I said earlier, the fundamental question is:

  (F) is it OK to continue with a session after processing
      a message which may have contained some NLRI which
      could not be extracted ?

As you say, the question hinges on the "which may have contained".  As
above, it seems to me that the draft skates over the issue of "lost
routes".

It also hinges on the operational impact of such "lost routes", on
which I am unqualified to pronounce.

It seems to me there is a range of possibilities:

  1) at one end of the spectrum, if "lost routes" are to
     be avoided at all costs, then the rules for parsing
     attributes need to be tight and without a change at
     the sender end, what is achievable is constrained.

  2) if "lost routes" are acceptable in the "theoretical"
     cases, but you-know-and-I-know that most of the
     time the case will not arise...

     ...or "lost routes" are acceptable provided some
     defined steps are taken to identify as much NLRI
     as is reasonably possible...

     ...then the requirements need to (a) make the case
     and (b) describe the trade-off (so that designs
     made to follow the requirements can be judged in
     those terms).

  3) at the other end of the spectrum, if "lost routes"
     are always preferable to session-reset, then this
     opens up an even looser approach to UPDATE message
     handling, in which practically nothing need cause
     a session-reset -- ie, practically nothing is a
     Critical Error.

And, as I have suggested in previous discussions, it is possible to
consider case (2) in two parts:

  a) where the sender is unchanged.

     In which case the issue is the degree of attribute
     broken-ness vs the likelihood of "lost routes".

     If this is viewed as an intermediate step, then
     perhaps the requirements could err on the side of
     safety ?

     Perhaps this can be addressed by a "knob" allowing
     more or less strictness in the parsing of
     attributes ?

  b) where the sender is changed.

     In which case the issue ought to be moot, because
     the receiver will then be able to know that there
     are no "lost routes".

> I would also appreciate further input from IDR as to whether this is
> sufficient requirement from GROW to allow a solution document to be
> written?

The new requirements specify only Critical (session-reset) and
Non-Critical (treat-as-withdraw) Errors.  This appears to rule out the
"attribute discard" option in draft-ietf-idr-error-handling-03.  Is
that intended ?

Chris


From chris.hall@highwayman.com  Sat Dec 29 03:09:12 2012
Return-Path: <chris.hall@highwayman.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3067021F85AB; Sat, 29 Dec 2012 03:09:12 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.208
X-Spam-Level: 
X-Spam-Status: No, score=0.208 tagged_above=-999 required=5 tests=[AWL=0.520,  BAYES_00=-2.599, HELO_MISMATCH_UK=1.749, HOST_MISMATCH_NET=0.311, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id VfMF5psNPb4n; Sat, 29 Dec 2012 03:09:11 -0800 (PST)
Received: from smtp.demon.co.uk (mdfmta005.mxout.tbr.inty.net [91.221.168.46]) by ietfa.amsl.com (Postfix) with ESMTP id 67FED21F85A7; Sat, 29 Dec 2012 03:09:11 -0800 (PST)
Received: from mdfmta005.tbr.inty.net (unknown [127.0.0.1]) by mdfmta005.tbr.inty.net (Postfix) with ESMTP id B087BA64315; Sat, 29 Dec 2012 11:09:09 +0000 (GMT)
Received: from mdfmta005.tbr.inty.net (unknown [127.0.0.1])	by mdfmta005.tbr.inty.net (Postfix) with ESMTP id 8B655A642CC; Sat, 29 Dec 2012 11:09:09 +0000 (GMT)
Received: from hestia.halldom.com (unknown [80.177.246.130])	(using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits))	(No client certificate requested)	by mdfmta005.tbr.inty.net (Postfix) with ESMTP; Sat, 29 Dec 2012 11:09:09 +0000 (GMT)
Received: from hyperion.halldom.com ([80.177.246.170] helo=HYPERION)	by hestia.halldom.com with esmtpsa (TLSv1:AES128-SHA:128)	(Exim 4.76)	(envelope-from <chris.hall@highwayman.com>)	id 1TouHs-0000wE-NR; Sat, 29 Dec 2012 11:09:08 +0000
From: "Chris Hall" <chris.hall@highwayman.com>
To: <idr@ietf.org>,	<grow@ietf.org>
Date: Sat, 29 Dec 2012 11:09:03 -0000
Organization: Highwayman
Message-ID: <02bd01cde5b4$ec6c58d0$c5450a70$@highwayman.com>
MIME-Version: 1.0
Content-Type: text/plain;	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: Ac3ltOeobOE5p/5+RsmrJ6jcHRxD7Q==
Content-Language: en-gb
X-MDF-HostID: 8
X-Mailman-Approved-At: Sat, 29 Dec 2012 19:26:25 -0800
Subject: Re: [GROW] [Idr] Fwd: I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 29 Dec 2012 11:09:12 -0000

Jakob Heitz wrote (on Fri 28-Dec-2012 at 21:35 +0000):
....
> All we can hope for after a malformed update is a
> temporary mitigation until human intervention can
> fix the problem.
>
> IMO, the goal of error handling is to limit the
> damage, not to cure the problem.

>From previous discussions it appears that to "limit the damage" you
would avoid session-reset at all costs.  Yes ?

Seems to me the goal of "Enhanced Error Handling" is, simply, to be
"not as bad as session-reset".

Where all NLRI in a message can be identified, treat-as-withdraw seems
to me to be not as bad as session-reset.

But: "Once you have a malformed update, NOTHING is certain."  So,
there is some risk of "lost NLRI" when tolerating (not resetting the
session) malformed updates.

I don't know if "lost NLRI" can cause problems which are worse than
session-reset.  I cannot find anything in
draft-ietf-grow-ops-reqs-for-bgp-error-handling-06 or
draft-ietf-idr-error-handling-03 which addresses the issue.  (Though
draft-ietf-idr-error-handling-03 waxes lyrical on "Why not discard
UPDATE messages ?")  

If the strategy is to treat-as-withdraw where you can, and continue
with whatever "lost NLRI" there may be, then that's simple enough.
But if that is the strategy, why do the drafts not say so ?  The
"ops-reqs" draft sets out to define Critical and Non-Critical errors
in terms of the inability/ability to extract NLRI, which is not
obviously consistent with this strategy.

It may be that "lost NLRI" are, indeed, potentially (in theory) worse
than session-reset, but that the risk is (in practice) reduced to a
tolerable level if a certain amount of care is taken to extract NLRI
-- falling back to session-reset if that process fails.  This seems to
me to be quite a sophisticated position -- too sophisticated to be
simply implied by what the drafts do not specify.

On the other hand, if "lost NLRI" are not as bad as session-reset,
then:

  a) when parsing the NLRI themselves, why not treat-as-
     withdraw everything up to the point where a prefix
     length overruns the NLRI collection ?

  b) when parsing MP_XXX attributes, why worry about
     duplicates ?  Why not treat-as-withdraw everything
     in sight ?

  c) when parsing the message layer, why worry about
     consistency of message length and withdrawn
     routes and attributes length ?

     Since "lost NLRI" are tolerable, why worry about
     losing some IPv4 Unicast NLRI ?

  d) the presence of 16 x 0xFF is surely sufficient
     to signal the start of a BGP Message.  So, when
     starting to read a message (after having read
     message length octets for the previous one),
     why not scan forwards looking for 16 x 0xFF ?

I suspect the drafts do not go this far because there are unspoken
assumptions about what is worse than session-reset; either that, or
because the issue of "lost NLRI" has been glossed over.  Neither of
which, I suggest, are very satisfactory.

Chris



From chris.hall@highwayman.com  Sat Dec 29 06:21:10 2012
Return-Path: <chris.hall@highwayman.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F2BF721F85DF; Sat, 29 Dec 2012 06:21:09 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.181
X-Spam-Level: 
X-Spam-Status: No, score=0.181 tagged_above=-999 required=5 tests=[AWL=0.493,  BAYES_00=-2.599, HELO_MISMATCH_UK=1.749, HOST_MISMATCH_NET=0.311, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id B6pLlAJG8tRX; Sat, 29 Dec 2012 06:21:09 -0800 (PST)
Received: from smtp.demon.co.uk (mdfmta010.mxout.tch.inty.net [91.221.169.51]) by ietfa.amsl.com (Postfix) with ESMTP id 52F2621F8505; Sat, 29 Dec 2012 06:21:09 -0800 (PST)
Received: from mdfmta010.tch.inty.net (unknown [127.0.0.1]) by mdfmta010.tch.inty.net (Postfix) with ESMTP id C382B400516; Sat, 29 Dec 2012 14:21:07 +0000 (GMT)
Received: from mdfmta010.tch.inty.net (unknown [127.0.0.1])	by mdfmta010.tch.inty.net (Postfix) with ESMTP id A2E16400514; Sat, 29 Dec 2012 14:21:07 +0000 (GMT)
Received: from hestia.halldom.com (unknown [80.177.246.130])	(using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits))	(No client certificate requested)	by mdfmta010.tch.inty.net (Postfix) with ESMTP; Sat, 29 Dec 2012 14:21:07 +0000 (GMT)
Received: from hyperion.halldom.com ([80.177.246.170] helo=HYPERION)	by hestia.halldom.com with esmtpsa (TLSv1:AES128-SHA:128)	(Exim 4.76)	(envelope-from <chris.hall@highwayman.com>)	id 1ToxHe-0000wS-Nh; Sat, 29 Dec 2012 14:21:06 +0000
From: "Chris Hall" <chris.hall@highwayman.com>
To: <idr@ietf.org>,	<grow@ietf.org>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com>	<3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh>	<028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com>
In-Reply-To: <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com>
Date: Sat, 29 Dec 2012 14:21:01 -0000
Organization: Highwayman
Message-ID: <02c101cde5cf$bd94d0d0$38be7270$@highwayman.com>
MIME-Version: 1.0
Content-Type: text/plain;	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQFuOjybZ81Jyr3vkBQVN4QFzf3TcAHhILBDAeLn4UsCePM9v5i9DclQ
Content-Language: en-gb
X-MDF-HostID: 19
X-Mailman-Approved-At: Sat, 29 Dec 2012 19:26:25 -0800
Subject: Re: [GROW] [Idr] Fwd: I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 29 Dec 2012 14:21:10 -0000

Brian Dickson wrote (on Fri 28-Dec-2012 at 19:02 +0000):
>
> If there is a chance that some NLRI weren't properly decoded:
>
>  - What about requesting (presuming the option was negotiated)
>    a route-refresh, or a "confirm per-AFI-SAFI prefix list"?

If the sender is sending broken updates, then a route-refresh might
repeat the error.  But that doesn't make things any worse.  In fact,
since one would only keep those routes which the route-refresh sent
correctly, that would improve things :-)  The draft (section 5) talks
about Graceful Restart and extensions thereof, but if those are not
available, Route-Refresh might be.

Is Route-Refresh a sufficiently common option that one can stop trying
to square the "Once you have a malformed update, NOTHING is certain"
circle ?

It might be nice to only refresh/confirm the affected AFI/SAFI... but
if MP_XXX attributes have gone walkabout, it may not be clear which
AFI/SAFI those are :-(

Chris


From brian.peter.dickson@gmail.com  Sun Dec 30 13:34:41 2012
Return-Path: <brian.peter.dickson@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9E04221F8A90; Sun, 30 Dec 2012 13:34:41 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.684
X-Spam-Level: 
X-Spam-Status: No, score=-0.684 tagged_above=-999 required=5 tests=[AWL=-2.913, BAYES_00=-2.599, GB_SUMOF=5, HTML_MESSAGE=0.001, J_CHICKENPOX_23=0.6, RCVD_IN_DNSWL_LOW=-1, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 28XwVK2c+11e; Sun, 30 Dec 2012 13:34:39 -0800 (PST)
Received: from mail-ea0-f173.google.com (mail-ea0-f173.google.com [209.85.215.173]) by ietfa.amsl.com (Postfix) with ESMTP id B14AE21F8A8F; Sun, 30 Dec 2012 13:34:38 -0800 (PST)
Received: by mail-ea0-f173.google.com with SMTP id i13so4898737eaa.18 for <multiple recipients>; Sun, 30 Dec 2012 13:34:37 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=aZNEA/DXolsdywbVDMA7T/CsQDdpVPBkZaej74CivQA=; b=gUEPjeQF730gH6BBtZmV4+J/1/1YE5tOH0VidE/ggNc5ogp0so+kkSfwXTc5FwB4qo CsH6a7zqHBkHtUeRO8m0xr2lYdeIrVLmFz6vYroDuv0WMJ4r5OakBv0dORBdMPNi2cgY 6m2S8NS9gqhxO7bKCskFD0HD+SHgGovAsliSjvwWK42rtGcXo9llAx3RXBkA9ShvrSan ZxcDTrpH39i2ihNQ2Vt3C8decUQOw/NTToRkSZ0HY38SRNsRU/kOZd82gwd5CIr9AkX4 elDNakWL8ogkFkJD6FZmcFAf4ap9+NTePwFNVqWew1sZB0m2R5yYZrHQZ9p9unld0etc 1m8Q==
MIME-Version: 1.0
Received: by 10.14.223.200 with SMTP id v48mr105497810eep.24.1356903277757; Sun, 30 Dec 2012 13:34:37 -0800 (PST)
Received: by 10.223.173.199 with HTTP; Sun, 30 Dec 2012 13:34:37 -0800 (PST)
In-Reply-To: <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh>
Date: Sun, 30 Dec 2012 16:34:37 -0500
Message-ID: <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com>
From: Brian Dickson <brian.peter.dickson@gmail.com>
To: Rob Shakir <rjs@rob.sh>
Content-Type: multipart/alternative; boundary=047d7b622820dd839904d218a9ee
Cc: idr@ietf.org, grow@ietf.org, Chris Hall <chris.hall@highwayman.com>
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 30 Dec 2012 21:34:41 -0000

--047d7b622820dd839904d218a9ee
Content-Type: text/plain; charset=ISO-8859-1

Here's the basic problem in a nutshell:

If a WITHDRAW is missed (either standard or MP), there are two
possibilities concerning withdrawn NLRIs:
- the WITHDRAW was sent because the sender no longer HAS a route to the
destination, or
- the WITHDRAW was sent because the sender now prefers the route via the
recipient

In the former case, losing the WITHDRAW means that the recipient, if it
selects that route as BEST, will now BLACKHOLE the traffic.
In the latter case, it might be possible for a routing loop to be caused
(although, I think most of the time this won't be possible unless there was
already a missed WITHDRAW in the other direction).

But, the basic problem is this: missing an UPDATE won't trigger either
condition, it will at worst cause sub-optimal routing. Missing a WITHDRAW
_CAN_ cause Bad Things (TM) to happen.

So, it is important to accommodate this in the requirements, possibly
requiring a very slight change in required message packing, in order to
guarantee that these Bad Things do not happen.

Today, those Bad Things can't happen when a session tear-down occurs.

There are a couple of easy ways to ensure the WITHDRAW is not missed:
- Pack all the WITHDRAW stuff at the start
- Require that no UPDATE be sent in the same MESSAGE as a WITHDRAW, and
preferably limit WITHDRAW to one AFI/SAFI per MESSAGE

It's clean, has relatively modest requirements, and ensures that a broken
WITHDRAW can trigger the correct result (tear-down), and any other broken
UPDATE can be handled with Treat-As-Withdraw.

IMNSHO, routing loops and/or black holes (for even a single prefix) are not
an acceptable situation.

There are other approaches that can be take to avoid this as well, but I
think KISS is the best approach.

Brian

On Sat, Dec 29, 2012 at 7:05 PM, Rob Shakir <rjs@rob.sh> wrote:

> Hi Brian,
>
> We discussed this at some length previously (the -00 of the requirements
> draft discussed automatically triggering some form of means to recover from
> the error). I think that there is some merit in doing so, but there are
> clearly also scaling considerations around this (in terms of requiring both
> BGP speakers to re-generate/re-process UPDATEs).
>
> The consensus that was reached in the previous discussion was that this is
> not necessarily going to be of advantage in terms of causing a different
> UPDATE message to be generated. Particularly, since a ROUTE REFRESH would
> result in the same UPDATE packing (generally). To that end, what the
> requirements draft currently says is that:
>
> - ROUTE-REFRESH is a reasonable way to re-run a consistency check between
> two peers (especially with start-of-refresh and end-of-refresh markers,
> since this means that we can purge any NLRI that we missed being withdrawn).
> - If there is anything automatically triggered, it would be prudent to try
> and ensure that we can make as specific a request to the other speaker as
> possible - this is where mechanisms such as One-Time ORF were suggested,
> and is a similar space to that which would be solved through having the
> UPDATE-VERSION message described in draft-ietf-idr-enhanced-gr if one were
> to be able to have a REFRESH following a "last known good" position. Making
> a more specific request may result in a different UPDATE being generated,
> which may mean the error is fixed, but this is not guaranteed.
> - If the selected best path changes on the remote speaker, then it will be
> re-advertised anyway - resulting in another chance to re-parse these NLRI.
>
> So, certainly, if we relax the "correctness" requirement, then it would
> seem a good idea to have some means to be able to consistency check -- it's
> just a matter of ensuring that this is done in an effective and scalable
> manner, that does not affect normal BGP operation.
>
> Cheers,
> r.
>
>
> On 28 Dec 2012, at 19:02, Brian Dickson <brian.peter.dickson@gmail.com>
> wrote:
>
> Here's a quick question, about possible mechanisms for determining whether
> or not we need to tear down a session.
>
> If there is a chance that some NLRI weren't properly decoded:
>
> - What about requesting (presuming the option was negotiated) a
> route-refresh, or a "confirm per-AFI-SAFI prefix list"?
>
> It's an expensive "parity check" but is one way of ensuring that we
> haven't missed a withdrawn prefix.
> If it isn't in the subsequently received list of prefixes, we missed it
> and should withdraw it.
> Everything else is a no-op - maybe we missed a new prefix (with new
> attribute?), but that is what "treat as withdraw" gets you.
>
> Just trying to make sure the cure isn't worse than the disease, in all
> situations.
> (Where "cure" is fail to withdraw because we "lost" the withdrawal because
> of malformed update, or "cure" is tear down the session.)
>
> Brian
>
> On Fri, Dec 28, 2012 at 1:04 PM, Chris Hall <chris.hall@highwayman.com>wrote:
>
>> Rob Shakir wrote (on Fri 28-Dec-2012 at 13:10 +0000):
>> >
>> > (re: CCing IDR & GROW)
>> >
>> > On 28 Dec 2012, at 12:28, Chris Hall wrote:
>> >
>> > > Rob Shakir wrote (on Thu 27-Dec-2012 at 18:44):
>> > >>
>> > >> Any comments very welcome (to me or grow@).
>>
>> > > I'm afraid I still don't get it :-(  What am I missing ?
>> > >
>> > > UPDATE Message Length errors are Critical because they:
>> > >
>> > > (1) "result in cases whereby the NLRI attribute cannot
>> > >      be correctly extracted".
>> > >
>> > > The implication is that a failure to extract all NLRI is Critical.
>> > > Is that a requirement ?
>>
>> > If the NLRI cannot be determined, then this is a Critical error,
>> > yes. I left the wording relatively open on whether this is *all*
>> > NLRI, ...
>>
>> OK.  So, if the message is broken (in any way):
>>
>>   * if no NLRI can be found, that's a Critical Error.
>>
>>     That would include the case where some broken
>>     attribute upstream of any MP_XXX obscures that
>>     MP_XXX.
>>
>>   * if some NLRI can be found, that's a Non-Critical error.
>>
>>     Such NLRI as can be found are treated-as-withdraw.
>>
>>     AND any NLRI that may or may not have been lost, are
>>     ignored.
>>
>>     AND whatever dangers lost NLRI may pose are covered
>>     by the caveats in 4.1.
>>
>> Yes ?
>>
>> > ... as I am not sure that in the requirements draft we should
>> > specify direct solutions to specific issues, to e.g., say how to
>> > handle cases where MP_REACH_NLRI and MP_UNREACH_NLRI are in the same
>> > message [this is a case that I do not believe is forbidden by
>> > rfc2858 - if the working group could clarify whether this is
>> > something that we feel the draft needs to handle or can explicitly
>> > be omitted, then that would be appreciated].
>>
>> The RFCs also allow any mix of IPv4 Unicast Reachable/Unreachable in
>> the body of the message, with or without MP_REACH_NLRI and/or
>> MP_UNREACH_NLRI.  If we call each of these a "collection of NLRI",
>> then there may be between 1 and 4 collections of NLRI in a message
>> (ignoring the End-of-RIB pseudo-UPDATE).  It is an error to have more
>> than one MP_REACH_NLRI attribute or more than one MP_UNREACH_NLRI
>> attribute.  It is not clear to me whether those should be Critical
>> Errors -- but while we are relaxing all error checking, I don't see
>> why they should be.
>>
>> It is possible that most implementations actually only send one
>> collection per message... so, as you suggest, on a you-know-and-I-know
>> basis, we'all could ignore the "theoretical" issues of more than one
>> collection in a message.
>>
>> But, this does not matter if the requirements allow lost NLRI to
>> simply be ignored.  If the sender sends any MP_XXX first, then there
>> is little chance that NLRI will be lost, and things work better.  If
>> the sender follows the old rules, and particularly if it sends more
>> than one collection at a time, then there is more chance that NLRI
>> will be lost.
>>
>> > > Later:
>> > >
>> > >  (2) "All errors whereby the contained NLRI can be
>> > >       extracted are referred to as Non-Critical".
>> > >
>> > > And that includes:
>> > >
>> > >  (3) "where the length of all path attributes contained
>> > >       within the UPDATE does not correspond to the
>> > >       total path attribute length."
>> > >
>> > > That is, at least, more explicit than
>> > > draft-ietf-idr-error-handling-03, which glosses over (3).
>> > >
>> > > But if (3) is non-critical then there is some chance that
>> > > some NLRI will not be extracted, which appears to violate
>> > > (1) and (2).
>>
>> > Disclaimer: As I am sure that my comments previously have made
>> > clear, I do not maintain a code base for a BGP daemon/implementation
>> > - so please feel free to correct my logic below.
>> >
>> > I do not believe that (3) implies that the NLRI cannot be correctly
>> > found.
>>
>> What (3) implies to me is that somewhere amongst the attributes
>> something is broken.  As Jakob Heitz succinctly puts it: "Once you
>> have a malformed update, NOTHING is certain."  In particular, you
>> cannot be certain that all NLRI referred to by the sender can be found
>> by the receiver.
>>
>> > If the sum of total length is incorrect, then we can still
>> > extract the individual attributes - we just find that there is not
>> > enough data to fill the overall length we were told and/or we have
>> > too much attribute data compared to the total attribute length.
>>
>> This is precisely where I think the requirements come unstuck.  There
>> is almost no redundancy in the attribute encoding.  And almost all the
>> redundancy there is is discarded by the new-form-error-handling.  So,
>> if the sum of the attribute lengths is incorrect, then (inter alia)
>> the receiver CANNOT know whether the attributes it has extracted are
>> the attributes the sender intended to send.  In particular, the
>> receiver cannot know it has extracted all NLRI (except in the,
>> probably obscure, case of having extracted both MP_REACH_NLRI and
>> MP_UNREACH_NLRI).  "Once you have a malformed update, NOTHING is
>> certain."
>>
>> Conversely, if the sum of attribute lengths is correct, then there is
>> a fighting chance that the attributes received are the attributes
>> sent.  But, if one of those attributes is (say) nominally an
>> ATOMIC_AGGREGATE which is apparently 200 octets long, that might be a
>> worry.
>>
>> ...
>> > > Then (4) "In order to maximise the number of cases whereby the
>> > > NLRI attributes [plural, now, BTW] can be reliably extracted
>> > > from a received message...".  Ah.  So it is not a Critical
>> > > Error if "the NLRI attribute cannot be correctly extracted".
>>
>> > No - it is a Critical error if we cannot extract the NLRI. This
>> > recommendation is to give an increased chance that the NLRI can be
>> > extracted as per the IDR error handling draft. This then (by virtue
>> > of resulting in the NLRI being extracted) minimises the number of
>> > cases that result in a Critical error.
>>
>> By this do you mean "all" NLRI or only "some" NLRI ?  As above.
>>
>> > The plural here is to reflect
>> > that the existence of >1 type of NLRI attribute.
>>
>> That would imply that it should be "NLRI attributes" throughout ?
>> Though in most places where the draft speaks of "NLRI attribute" I
>> think it means one, some, any or all of the "collections of NLRI" (as
>> defined above).
>>
>> > > For me the requirement remains "conflicted".  On the one hand it
>> > > seems to say that it is a Critical Error if the NLRI cannot be
>> > > extracted and parsed.  On the other it seems to say it's OK if
>> > > you cannot extract some NLRI.
>>
>> > If you'll forgive me for removing a significant proportion of your
>> > message, I think that we need to take another step back here. It
>> > seems to me that the key question that you are highlighting is "What
>> > level of confidence do we need to have before we declare that the
>> > NLRI cannot be extracted?" -- do you agree?
>>
>> Absolutely.
>>
>> > From an operator perspective, I would like to compromise *certainty*
>> > for *robustness*. You are right, we are compromising correctness
>> > here, we might end up withdrawing an incorrect NLRI and impacting
>> > service operation for that prefix - however, it is somewhat
>> > preferable to me to withdraw a a subset of the NLRI incorrectly,
>> > rather than impact all NLRI in one single action. We clearly need to
>> > provide some bounds on how much we compromise the certainty (and
>> > live within the realms of possibility, such that we are not just
>> > taking a shot in the dark). This is what the definitions of Critical
>> > and Non-Critical within the document are intended to provide. Once
>> > again I will refer to the requirement that there is a balance
>> > between correctness and robustness - rather than a locally risk
>> > averse approach that results in harmful wider behaviour.
>>
>> Where one can extract the NLRI from a broken message, then
>> treat-as-withdraw is AFAICS no worse for the NLRI in question than
>> session-reset, and hugely better for all the other NLRI received from
>> the peer.
>>
>> Where not *all* the NLRI in a message have been extracted, then we
>> take a step beyond withdrawing something which should not have been
>> withdrawn.  For each of those NLRI the receiver will (unknowingly) do
>> one of:
>>
>>   (a) continue using a route which the sender has
>>       withdrawn;
>>
>>   (b) continue to use an out of date version of a route
>>       which the sender has changed in some way, possibly
>>       materially;
>>
>>   (c) to fail to use a new route the sender has now made
>>       available.
>>
>> Of these (a) looks serious and (b) could be, but certainly the effect
>> is different from session-reset; while (c) is, essentially,
>> treat-as-withdraw.  If some risk "lost routes" is taken, then the
>> impact of that clearly must be weighed against the impact of
>> session-reset.  I have not found a discussion of the impact of "lost
>> routes" in the draft, so I have no idea what the trade-off is, here.
>>
>> For completeness, if all forms and degrees of attribute broken-ness
>> are required to be acceptable, then there are cases where the receiver
>> cannot know whether *all* NLRI have been extracted, or not.
>>
>> > Is it acceptable that we leave this as guidance within the
>> > requirements? If not, please could you suggest how the definitions
>> > of Critical/Non-Critical could be altered to address your concerns?
>>
>> Without a change at the sender end, the problem is that: "Once you
>> have a malformed update, NOTHING is certain."
>>
>> So, as I said earlier, the fundamental question is:
>>
>>   (F) is it OK to continue with a session after processing
>>       a message which may have contained some NLRI which
>>       could not be extracted ?
>>
>> As you say, the question hinges on the "which may have contained".  As
>> above, it seems to me that the draft skates over the issue of "lost
>> routes".
>>
>> It also hinges on the operational impact of such "lost routes", on
>> which I am unqualified to pronounce.
>>
>> It seems to me there is a range of possibilities:
>>
>>   1) at one end of the spectrum, if "lost routes" are to
>>      be avoided at all costs, then the rules for parsing
>>      attributes need to be tight and without a change at
>>      the sender end, what is achievable is constrained.
>>
>>   2) if "lost routes" are acceptable in the "theoretical"
>>      cases, but you-know-and-I-know that most of the
>>      time the case will not arise...
>>
>>      ...or "lost routes" are acceptable provided some
>>      defined steps are taken to identify as much NLRI
>>      as is reasonably possible...
>>
>>      ...then the requirements need to (a) make the case
>>      and (b) describe the trade-off (so that designs
>>      made to follow the requirements can be judged in
>>      those terms).
>>
>>   3) at the other end of the spectrum, if "lost routes"
>>      are always preferable to session-reset, then this
>>      opens up an even looser approach to UPDATE message
>>      handling, in which practically nothing need cause
>>      a session-reset -- ie, practically nothing is a
>>      Critical Error.
>>
>> And, as I have suggested in previous discussions, it is possible to
>> consider case (2) in two parts:
>>
>>   a) where the sender is unchanged.
>>
>>      In which case the issue is the degree of attribute
>>      broken-ness vs the likelihood of "lost routes".
>>
>>      If this is viewed as an intermediate step, then
>>      perhaps the requirements could err on the side of
>>      safety ?
>>
>>      Perhaps this can be addressed by a "knob" allowing
>>      more or less strictness in the parsing of
>>      attributes ?
>>
>>   b) where the sender is changed.
>>
>>      In which case the issue ought to be moot, because
>>      the receiver will then be able to know that there
>>      are no "lost routes".
>>
>> > I would also appreciate further input from IDR as to whether this is
>> > sufficient requirement from GROW to allow a solution document to be
>> > written?
>>
>> The new requirements specify only Critical (session-reset) and
>> Non-Critical (treat-as-withdraw) Errors.  This appears to rule out the
>> "attribute discard" option in draft-ietf-idr-error-handling-03.  Is
>> that intended ?
>>
>> Chris
>>
>> _______________________________________________
>> Idr mailing list
>> Idr@ietf.org
>> https://www.ietf.org/mailman/listinfo/idr
>>
>
>
>

--047d7b622820dd839904d218a9ee
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Here&#39;s the basic problem in a nutshell:<div><br></div><div>If a WITHDRA=
W is missed (either standard or MP), there are two possibilities concerning=
 withdrawn NLRIs:</div><div>- the WITHDRAW was sent because the sender no l=
onger HAS a route to the destination, or</div>
<div>- the WITHDRAW was sent because the sender now prefers the route via t=
he recipient</div><div><br></div><div>In the former case, losing the WITHDR=
AW means that the recipient, if it selects that route as BEST, will now BLA=
CKHOLE the traffic.</div>
<div>In the latter case, it might be possible for a routing loop to be caus=
ed (although, I think most of the time this won&#39;t be possible unless th=
ere was already a missed WITHDRAW in the other direction).</div><div><br>
</div><div>But, the basic problem is this: missing an UPDATE won&#39;t trig=
ger either condition, it will at worst cause sub-optimal routing. Missing a=
 WITHDRAW _CAN_ cause Bad Things (TM) to happen.</div><div><br></div><div>
So, it is important to accommodate this in the requirements, possibly requi=
ring a very slight change in required message packing, in order to guarante=
e that these Bad Things do not happen.</div><div><br></div><div>Today, thos=
e Bad Things can&#39;t happen when a session tear-down occurs.</div>
<div><br></div><div>There are a couple of easy ways to ensure the WITHDRAW =
is not missed:</div><div>- Pack all the WITHDRAW stuff at the start</div><d=
iv>- Require that no UPDATE be sent in the same MESSAGE as a WITHDRAW, and =
preferably limit WITHDRAW to one AFI/SAFI per MESSAGE</div>
<div><br></div><div>It&#39;s clean, has relatively modest requirements, and=
 ensures that a broken WITHDRAW can trigger the correct result (tear-down),=
 and any other broken UPDATE can be handled with Treat-As-Withdraw.</div>
<div><br></div><div>IMNSHO, routing loops and/or black holes (for even a si=
ngle prefix) are not an acceptable situation.</div><div><br></div><div>Ther=
e are other approaches that can be take to avoid this as well, but I think =
KISS is the best approach.</div>
<div><br></div><div>Brian</div><div><br><div class=3D"gmail_quote">On Sat, =
Dec 29, 2012 at 7:05 PM, Rob Shakir <span dir=3D"ltr">&lt;<a href=3D"mailto=
:rjs@rob.sh" target=3D"_blank">rjs@rob.sh</a>&gt;</span> wrote:<br><blockqu=
ote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc s=
olid;padding-left:1ex">
<div style=3D"word-wrap:break-word">Hi Brian,<div><br></div><div>We discuss=
ed this at some length previously (the -00 of the requirements draft discus=
sed automatically triggering some form of means to recover from the error).=
 I think that there is some merit in doing so, but there are clearly also s=
caling considerations around this (in terms of requiring both BGP speakers =
to re-generate/re-process UPDATEs).=A0</div>
<div><br></div><div>The consensus that was reached in the previous discussi=
on was that this is not necessarily going to be of advantage in terms of ca=
using a different UPDATE message to be generated. Particularly, since a ROU=
TE REFRESH would result in the same UPDATE packing (generally). To that end=
, what the requirements draft currently says is that:</div>
<div><br></div><div>- ROUTE-REFRESH is a reasonable way to re-run a consist=
ency check between two peers (especially with start-of-refresh and end-of-r=
efresh markers, since this means that we can purge any NLRI that we missed =
being withdrawn).</div>
<div>- If there is anything automatically triggered, it would be prudent to=
 try and ensure that we can make as specific a request to the other speaker=
 as possible - this is where mechanisms such as One-Time ORF were suggested=
, and is a similar space to that which would be solved through having the U=
PDATE-VERSION message described in draft-ietf-idr-enhanced-gr if one were t=
o be able to have a REFRESH following a &quot;last known good&quot; positio=
n. Making a more specific request may result in a different UPDATE being ge=
nerated, which may mean the error is fixed, but this is not guaranteed.</di=
v>
<div>- If the selected best path changes on the remote speaker, then it wil=
l be re-advertised anyway - resulting in another chance to re-parse these N=
LRI.</div><div><br></div><div>So, certainly, if we relax the &quot;correctn=
ess&quot; requirement, then it would seem a good idea to have some means to=
 be able to consistency check -- it&#39;s just a matter of ensuring that th=
is is done in an effective and scalable manner, that does not affect normal=
 BGP operation.</div>
<div><br></div><div>Cheers,</div><div>r.</div><div><div class=3D"h5"><div><=
br></div><div><br><div><div>On 28 Dec 2012, at 19:02, Brian Dickson &lt;<a =
href=3D"mailto:brian.peter.dickson@gmail.com" target=3D"_blank">brian.peter=
.dickson@gmail.com</a>&gt; wrote:</div>
<br><blockquote type=3D"cite">Here&#39;s a quick question, about possible m=
echanisms for determining whether or not we need to tear down a session.<di=
v><br></div><div>If there is a chance that some NLRI weren&#39;t properly d=
ecoded:</div>
<div><br></div>
<div>- What about requesting (presuming the option was negotiated) a route-=
refresh, or a &quot;confirm per-AFI-SAFI prefix list&quot;?</div><div><br><=
/div><div>It&#39;s an expensive &quot;parity check&quot; but is one way of =
ensuring that we haven&#39;t missed a withdrawn prefix.</div>

<div>If it isn&#39;t in the subsequently received list of prefixes, we miss=
ed it and should withdraw it.</div><div>Everything else is a no-op - maybe =
we missed a new prefix (with new attribute?), but that is what &quot;treat =
as withdraw&quot; gets you.</div>

<div><br></div><div>Just trying to make sure the cure isn&#39;t worse than =
the disease, in all situations.</div><div>(Where &quot;cure&quot; is fail t=
o withdraw because we &quot;lost&quot; the withdrawal because of malformed =
update, or &quot;cure&quot; is tear down the session.)</div>

<div><br></div><div>Brian<br><br><div class=3D"gmail_quote">On Fri, Dec 28,=
 2012 at 1:04 PM, Chris Hall <span dir=3D"ltr">&lt;<a href=3D"mailto:chris.=
hall@highwayman.com" target=3D"_blank">chris.hall@highwayman.com</a>&gt;</s=
pan> wrote:<br>

<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">Rob Shakir wrote (on Fri 28-Dec-2012 at 13:1=
0 +0000):<br>
<div>&gt;<br>
&gt; (re: CCing IDR &amp; GROW)<br>
&gt;<br>
&gt; On 28 Dec 2012, at 12:28, Chris Hall wrote:<br>
&gt;<br>
&gt; &gt; Rob Shakir wrote (on Thu 27-Dec-2012 at 18:44):<br>
&gt; &gt;&gt;<br>
&gt; &gt;&gt; Any comments very welcome (to me or grow@).<br>
<br>
&gt; &gt; I&#39;m afraid I still don&#39;t get it :-( =A0What am I missing =
?<br>
&gt; &gt;<br>
&gt; &gt; UPDATE Message Length errors are Critical because they:<br>
&gt; &gt;<br>
&gt; &gt; (1) &quot;result in cases whereby the NLRI attribute cannot<br>
&gt; &gt; =A0 =A0 =A0be correctly extracted&quot;.<br>
&gt; &gt;<br>
&gt; &gt; The implication is that a failure to extract all NLRI is Critical=
.<br>
&gt; &gt; Is that a requirement ?<br>
<br>
&gt; If the NLRI cannot be determined, then this is a Critical error,<br>
&gt; yes. I left the wording relatively open on whether this is *all*<br>
</div>&gt; NLRI, ...<br>
<br>
OK. =A0So, if the message is broken (in any way):<br>
<br>
=A0 * if no NLRI can be found, that&#39;s a Critical Error.<br>
<br>
=A0 =A0 That would include the case where some broken<br>
=A0 =A0 attribute upstream of any MP_XXX obscures that<br>
=A0 =A0 MP_XXX.<br>
<br>
=A0 * if some NLRI can be found, that&#39;s a Non-Critical error.<br>
<br>
=A0 =A0 Such NLRI as can be found are treated-as-withdraw.<br>
<br>
=A0 =A0 AND any NLRI that may or may not have been lost, are<br>
=A0 =A0 ignored.<br>
<br>
=A0 =A0 AND whatever dangers lost NLRI may pose are covered<br>
=A0 =A0 by the caveats in 4.1.<br>
<br>
Yes ?<br>
<br>
&gt; ... as I am not sure that in the requirements draft we should<br>
<div>&gt; specify direct solutions to specific issues, to e.g., say how to<=
br>
&gt; handle cases where MP_REACH_NLRI and MP_UNREACH_NLRI are in the same<b=
r>
&gt; message [this is a case that I do not believe is forbidden by<br>
&gt; rfc2858 - if the working group could clarify whether this is<br>
&gt; something that we feel the draft needs to handle or can explicitly<br>
&gt; be omitted, then that would be appreciated].<br>
<br>
</div>The RFCs also allow any mix of IPv4 Unicast Reachable/Unreachable in<=
br>
the body of the message, with or without MP_REACH_NLRI and/or<br>
MP_UNREACH_NLRI. =A0If we call each of these a &quot;collection of NLRI&quo=
t;,<br>
then there may be between 1 and 4 collections of NLRI in a message<br>
(ignoring the End-of-RIB pseudo-UPDATE). =A0It is an error to have more<br>
than one MP_REACH_NLRI attribute or more than one MP_UNREACH_NLRI<br>
attribute. =A0It is not clear to me whether those should be Critical<br>
Errors -- but while we are relaxing all error checking, I don&#39;t see<br>
why they should be.<br>
<br>
It is possible that most implementations actually only send one<br>
collection per message... so, as you suggest, on a you-know-and-I-know<br>
basis, we&#39;all could ignore the &quot;theoretical&quot; issues of more t=
han one<br>
collection in a message.<br>
<br>
But, this does not matter if the requirements allow lost NLRI to<br>
simply be ignored. =A0If the sender sends any MP_XXX first, then there<br>
is little chance that NLRI will be lost, and things work better. =A0If<br>
the sender follows the old rules, and particularly if it sends more<br>
than one collection at a time, then there is more chance that NLRI<br>
will be lost.<br>
<div><br>
&gt; &gt; Later:<br>
&gt; &gt;<br>
&gt; &gt; =A0(2) &quot;All errors whereby the contained NLRI can be<br>
&gt; &gt; =A0 =A0 =A0 extracted are referred to as Non-Critical&quot;.<br>
&gt; &gt;<br>
&gt; &gt; And that includes:<br>
&gt; &gt;<br>
&gt; &gt; =A0(3) &quot;where the length of all path attributes contained<br=
>
&gt; &gt; =A0 =A0 =A0 within the UPDATE does not correspond to the<br>
&gt; &gt; =A0 =A0 =A0 total path attribute length.&quot;<br>
&gt; &gt;<br>
&gt; &gt; That is, at least, more explicit than<br>
&gt; &gt; draft-ietf-idr-error-handling-03, which glosses over (3).<br>
&gt; &gt;<br>
&gt; &gt; But if (3) is non-critical then there is some chance that<br>
&gt; &gt; some NLRI will not be extracted, which appears to violate<br>
&gt; &gt; (1) and (2).<br>
<br>
&gt; Disclaimer: As I am sure that my comments previously have made<br>
&gt; clear, I do not maintain a code base for a BGP daemon/implementation<b=
r>
&gt; - so please feel free to correct my logic below.<br>
&gt;<br>
&gt; I do not believe that (3) implies that the NLRI cannot be correctly<br=
>
&gt; found.<br>
<br>
</div>What (3) implies to me is that somewhere amongst the attributes<br>
something is broken. =A0As Jakob Heitz succinctly puts it: &quot;Once you<b=
r>
have a malformed update, NOTHING is certain.&quot; =A0In particular, you<br=
>
cannot be certain that all NLRI referred to by the sender can be found<br>
by the receiver.<br>
<div><br>
&gt; If the sum of total length is incorrect, then we can still<br>
&gt; extract the individual attributes - we just find that there is not<br>
&gt; enough data to fill the overall length we were told and/or we have<br>
&gt; too much attribute data compared to the total attribute length.<br>
<br>
</div>This is precisely where I think the requirements come unstuck. =A0The=
re<br>
is almost no redundancy in the attribute encoding. =A0And almost all the<br=
>
redundancy there is is discarded by the new-form-error-handling. =A0So,<br>
if the sum of the attribute lengths is incorrect, then (inter alia)<br>
the receiver CANNOT know whether the attributes it has extracted are<br>
the attributes the sender intended to send. =A0In particular, the<br>
receiver cannot know it has extracted all NLRI (except in the,<br>
probably obscure, case of having extracted both MP_REACH_NLRI and<br>
MP_UNREACH_NLRI). =A0&quot;Once you have a malformed update, NOTHING is<br>
certain.&quot;<br>
<br>
Conversely, if the sum of attribute lengths is correct, then there is<br>
a fighting chance that the attributes received are the attributes<br>
sent. =A0But, if one of those attributes is (say) nominally an<br>
ATOMIC_AGGREGATE which is apparently 200 octets long, that might be a<br>
worry.<br>
<br>
...<br>
<div>&gt; &gt; Then (4) &quot;In order to maximise the number of cases wher=
eby the<br>
&gt; &gt; NLRI attributes [plural, now, BTW] can be reliably extracted<br>
&gt; &gt; from a received message...&quot;. =A0Ah. =A0So it is not a Critic=
al<br>
&gt; &gt; Error if &quot;the NLRI attribute cannot be correctly extracted&q=
uot;.<br>
<br>
&gt; No - it is a Critical error if we cannot extract the NLRI. This<br>
&gt; recommendation is to give an increased chance that the NLRI can be<br>
&gt; extracted as per the IDR error handling draft. This then (by virtue<br=
>
&gt; of resulting in the NLRI being extracted) minimises the number of<br>
&gt; cases that result in a Critical error.<br>
<br>
</div>By this do you mean &quot;all&quot; NLRI or only &quot;some&quot; NLR=
I ? =A0As above.<br>
<div><br>
&gt; The plural here is to reflect<br>
&gt; that the existence of &gt;1 type of NLRI attribute.<br>
<br>
</div>That would imply that it should be &quot;NLRI attributes&quot; throug=
hout ?<br>
Though in most places where the draft speaks of &quot;NLRI attribute&quot; =
I<br>
think it means one, some, any or all of the &quot;collections of NLRI&quot;=
 (as<br>
defined above).<br>
<div><br>
&gt; &gt; For me the requirement remains &quot;conflicted&quot;. =A0On the =
one hand it<br>
&gt; &gt; seems to say that it is a Critical Error if the NLRI cannot be<br=
>
&gt; &gt; extracted and parsed. =A0On the other it seems to say it&#39;s OK=
 if<br>
&gt; &gt; you cannot extract some NLRI.<br>
<br>
&gt; If you&#39;ll forgive me for removing a significant proportion of your=
<br>
&gt; message, I think that we need to take another step back here. It<br>
&gt; seems to me that the key question that you are highlighting is &quot;W=
hat<br>
&gt; level of confidence do we need to have before we declare that the<br>
&gt; NLRI cannot be extracted?&quot; -- do you agree?<br>
<br>
</div>Absolutely.<br>
<div><br>
&gt; From an operator perspective, I would like to compromise *certainty*<b=
r>
&gt; for *robustness*. You are right, we are compromising correctness<br>
&gt; here, we might end up withdrawing an incorrect NLRI and impacting<br>
&gt; service operation for that prefix - however, it is somewhat<br>
&gt; preferable to me to withdraw a a subset of the NLRI incorrectly,<br>
&gt; rather than impact all NLRI in one single action. We clearly need to<b=
r>
&gt; provide some bounds on how much we compromise the certainty (and<br>
&gt; live within the realms of possibility, such that we are not just<br>
&gt; taking a shot in the dark). This is what the definitions of Critical<b=
r>
&gt; and Non-Critical within the document are intended to provide. Once<br>
&gt; again I will refer to the requirement that there is a balance<br>
&gt; between correctness and robustness - rather than a locally risk<br>
&gt; averse approach that results in harmful wider behaviour.<br>
<br>
</div>Where one can extract the NLRI from a broken message, then<br>
treat-as-withdraw is AFAICS no worse for the NLRI in question than<br>
session-reset, and hugely better for all the other NLRI received from<br>
the peer.<br>
<br>
Where not *all* the NLRI in a message have been extracted, then we<br>
take a step beyond withdrawing something which should not have been<br>
withdrawn. =A0For each of those NLRI the receiver will (unknowingly) do<br>
one of:<br>
<br>
=A0 (a) continue using a route which the sender has<br>
=A0 =A0 =A0 withdrawn;<br>
<br>
=A0 (b) continue to use an out of date version of a route<br>
=A0 =A0 =A0 which the sender has changed in some way, possibly<br>
=A0 =A0 =A0 materially;<br>
<br>
=A0 (c) to fail to use a new route the sender has now made<br>
=A0 =A0 =A0 available.<br>
<br>
Of these (a) looks serious and (b) could be, but certainly the effect<br>
is different from session-reset; while (c) is, essentially,<br>
treat-as-withdraw. =A0If some risk &quot;lost routes&quot; is taken, then t=
he<br>
impact of that clearly must be weighed against the impact of<br>
session-reset. =A0I have not found a discussion of the impact of &quot;lost=
<br>
routes&quot; in the draft, so I have no idea what the trade-off is, here.<b=
r>
<br>
For completeness, if all forms and degrees of attribute broken-ness<br>
are required to be acceptable, then there are cases where the receiver<br>
cannot know whether *all* NLRI have been extracted, or not.<br>
<div><br>
&gt; Is it acceptable that we leave this as guidance within the<br>
&gt; requirements? If not, please could you suggest how the definitions<br>
&gt; of Critical/Non-Critical could be altered to address your concerns?<br=
>
<br>
</div>Without a change at the sender end, the problem is that: &quot;Once y=
ou<br>
have a malformed update, NOTHING is certain.&quot;<br>
<br>
So, as I said earlier, the fundamental question is:<br>
<br>
=A0 (F) is it OK to continue with a session after processing<br>
=A0 =A0 =A0 a message which may have contained some NLRI which<br>
=A0 =A0 =A0 could not be extracted ?<br>
<br>
As you say, the question hinges on the &quot;which may have contained&quot;=
. =A0As<br>
above, it seems to me that the draft skates over the issue of &quot;lost<br=
>
routes&quot;.<br>
<br>
It also hinges on the operational impact of such &quot;lost routes&quot;, o=
n<br>
which I am unqualified to pronounce.<br>
<br>
It seems to me there is a range of possibilities:<br>
<br>
=A0 1) at one end of the spectrum, if &quot;lost routes&quot; are to<br>
=A0 =A0 =A0be avoided at all costs, then the rules for parsing<br>
=A0 =A0 =A0attributes need to be tight and without a change at<br>
=A0 =A0 =A0the sender end, what is achievable is constrained.<br>
<br>
=A0 2) if &quot;lost routes&quot; are acceptable in the &quot;theoretical&q=
uot;<br>
=A0 =A0 =A0cases, but you-know-and-I-know that most of the<br>
=A0 =A0 =A0time the case will not arise...<br>
<br>
=A0 =A0 =A0...or &quot;lost routes&quot; are acceptable provided some<br>
=A0 =A0 =A0defined steps are taken to identify as much NLRI<br>
=A0 =A0 =A0as is reasonably possible...<br>
<br>
=A0 =A0 =A0...then the requirements need to (a) make the case<br>
=A0 =A0 =A0and (b) describe the trade-off (so that designs<br>
=A0 =A0 =A0made to follow the requirements can be judged in<br>
=A0 =A0 =A0those terms).<br>
<br>
=A0 3) at the other end of the spectrum, if &quot;lost routes&quot;<br>
=A0 =A0 =A0are always preferable to session-reset, then this<br>
=A0 =A0 =A0opens up an even looser approach to UPDATE message<br>
=A0 =A0 =A0handling, in which practically nothing need cause<br>
=A0 =A0 =A0a session-reset -- ie, practically nothing is a<br>
=A0 =A0 =A0Critical Error.<br>
<br>
And, as I have suggested in previous discussions, it is possible to<br>
consider case (2) in two parts:<br>
<br>
=A0 a) where the sender is unchanged.<br>
<br>
=A0 =A0 =A0In which case the issue is the degree of attribute<br>
=A0 =A0 =A0broken-ness vs the likelihood of &quot;lost routes&quot;.<br>
<br>
=A0 =A0 =A0If this is viewed as an intermediate step, then<br>
=A0 =A0 =A0perhaps the requirements could err on the side of<br>
=A0 =A0 =A0safety ?<br>
<br>
=A0 =A0 =A0Perhaps this can be addressed by a &quot;knob&quot; allowing<br>
=A0 =A0 =A0more or less strictness in the parsing of<br>
=A0 =A0 =A0attributes ?<br>
<br>
=A0 b) where the sender is changed.<br>
<br>
=A0 =A0 =A0In which case the issue ought to be moot, because<br>
=A0 =A0 =A0the receiver will then be able to know that there<br>
=A0 =A0 =A0are no &quot;lost routes&quot;.<br>
<div><br>
&gt; I would also appreciate further input from IDR as to whether this is<b=
r>
&gt; sufficient requirement from GROW to allow a solution document to be<br=
>
&gt; written?<br>
<br>
</div>The new requirements specify only Critical (session-reset) and<br>
Non-Critical (treat-as-withdraw) Errors. =A0This appears to rule out the<br=
>
&quot;attribute discard&quot; option in draft-ietf-idr-error-handling-03. =
=A0Is<br>
that intended ?<br>
<span><font color=3D"#888888"><br>
Chris<br>
</font></span><div><div><br>
_______________________________________________<br>
Idr mailing list<br>
<a href=3D"mailto:Idr@ietf.org" target=3D"_blank">Idr@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/idr" target=3D"_blank">htt=
ps://www.ietf.org/mailman/listinfo/idr</a><br>
</div></div></blockquote></div><br></div>
</blockquote></div><br></div></div></div></div></blockquote></div><br></div=
>

--047d7b622820dd839904d218a9ee--

From john@jlc.net  Sun Dec 30 19:09:25 2012
Return-Path: <john@jlc.net>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B018421F8BD5; Sun, 30 Dec 2012 19:09:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -106.149
X-Spam-Level: 
X-Spam-Status: No, score=-106.149 tagged_above=-999 required=5 tests=[AWL=0.223, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, SARE_SUB_OBFU_Q1=0.227, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uCdnkv5E6csm; Sun, 30 Dec 2012 19:09:25 -0800 (PST)
Received: from mailhost.jlc.net (mailhost.jlc.net [199.201.159.4]) by ietfa.amsl.com (Postfix) with ESMTP id 1592821F8BD1; Sun, 30 Dec 2012 19:09:25 -0800 (PST)
Received: by mailhost.jlc.net (Postfix, from userid 104) id 6CDFB33C26; Sun, 30 Dec 2012 22:09:25 -0500 (EST)
Date: Sun, 30 Dec 2012 22:09:25 -0500
From: John Leslie <john@jlc.net>
To: Brian Dickson <brian.peter.dickson@gmail.com>
Message-ID: <20121231030925.GB62942@verdi>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com>
User-Agent: Mutt/1.4.1i
X-Mailman-Approved-At: Sun, 30 Dec 2012 19:23:33 -0800
Cc: idr@ietf.org, grow@ietf.org
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 03:09:25 -0000

Brian Dickson <brian.peter.dickson@gmail.com> wrote:
> 
> But, the basic problem is this: missing an UPDATE won't trigger either
> condition, it will at worst cause sub-optimal routing. Missing a WITHDRAW
> _CAN_ cause Bad Things (TM) to happen.

   +1

--
John Leslie <john@jlc.net>

From jsw@inconcepts.biz  Sun Dec 30 20:02:42 2012
Return-Path: <jsw@inconcepts.biz>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AFEE521F8B8B for <grow@ietfa.amsl.com>; Sun, 30 Dec 2012 20:02:42 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.737
X-Spam-Level: 
X-Spam-Status: No, score=-2.737 tagged_above=-999 required=5 tests=[AWL=0.013,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bywxErefhSPU for <grow@ietfa.amsl.com>; Sun, 30 Dec 2012 20:02:42 -0800 (PST)
Received: from mail-ie0-f175.google.com (mail-ie0-f175.google.com [209.85.223.175]) by ietfa.amsl.com (Postfix) with ESMTP id DC0C721F8A69 for <grow@ietf.org>; Sun, 30 Dec 2012 20:02:41 -0800 (PST)
Received: by mail-ie0-f175.google.com with SMTP id qd14so14449067ieb.6 for <grow@ietf.org>; Sun, 30 Dec 2012 20:02:41 -0800 (PST)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:x-originating-ip:in-reply-to:references:date :message-id:subject:from:to:cc:content-type:x-gm-message-state; bh=aPWyOGwBBa94G9A5Q7xPczFbDPqWmBNbXU2ctf0GXNE=; b=aN4n7pY2yo1RNp3JG3g1S1W9JSVqPCHk6jjJ9zTuSCaqIePP7yFiG9bWrJkUTAuha+ Q6FNCewqtkn9iAgUcy0RS+OUpM0CqtsbHgf6Hdmthr1jkDWIve8y0kmNdrhawrEaSey8 vam7/C/zDtGMKDW6SMzJ+G2TdckV7uwVTmMSpKN9lXpbwjzzfuGAAfGcnhKHs9Uj+gmx B9OoRL8HcGl9hvV9SrasX7NGKNg41b79vAqjx8r307irAipBA/5xZ4o62TwMI1um+nD3 VPceP40KSl7KRVuEDOieNlC3hcvm6dzmCz6rBm/pj8Q+03WQK4PPV5ztRsi6MSpb00Zt z+mg==
MIME-Version: 1.0
Received: by 10.50.104.232 with SMTP id gh8mr34420278igb.45.1356926561449; Sun, 30 Dec 2012 20:02:41 -0800 (PST)
Received: by 10.64.132.33 with HTTP; Sun, 30 Dec 2012 20:02:41 -0800 (PST)
X-Originating-IP: [74.134.22.105]
In-Reply-To: <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com>
Date: Sun, 30 Dec 2012 23:02:41 -0500
Message-ID: <CAPWAtbKj2pb4vDqadbggubnAT5k5YRh=dLuGcs33AjFd=bFw-Q@mail.gmail.com>
From: Jeff Wheeler <jsw@inconcepts.biz>
To: Brian Dickson <brian.peter.dickson@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
X-Gm-Message-State: ALoCoQmW0rzr1SZBJNGiNtjnadXk8qfDfKFG0YRF083sI919to8LiwSKR55MpyE6jmB/45XOXtGl
X-Mailman-Approved-At: Sun, 30 Dec 2012 20:11:28 -0800
Cc: idr@ietf.org, grow@ietf.org
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 04:02:42 -0000

On Sun, Dec 30, 2012 at 4:34 PM, Brian Dickson
<brian.peter.dickson@gmail.com> wrote:
> There are a couple of easy ways to ensure the WITHDRAW is not missed:
> - Pack all the WITHDRAW stuff at the start
> - Require that no UPDATE be sent in the same MESSAGE as a WITHDRAW, and
> preferably limit WITHDRAW to one AFI/SAFI per MESSAGE

The change you are proposing is sensible and has minimal cost to the
control-plane CPU.

BGP already can only have one, two, or three AFI/SAFI per Message.
The reason it's "one, two, or three" is there can be native NLRI and
MP-NLRI in one Message.

You are not allowed to send more than one Attribute of the same Type
Code in a BGP Message.  There can't be two MP_REACH_NLRI or two
MP_UNREACH_NLRI.

The rule about one attribute per type code is kind of buried in
RFC4271 section 6.3 paragraph 14.  I imagine not everyone is aware of
this because it really is not written into the spec in a way that it
jumps out at you!

Here is how this plus MP-BGP combine to allow up to three NLRI in the
same Message:
1) native NLRI which are either reachable or withdrawn
2) NLRI inside an MP_REACH_NLRI Attribute
3) NLRI inside an MP_UNREACH_NLRI Attribute

All three of those could possibly have distinct AFI/SAFI today.  But
it's not like it is possible to have NLRIs for 10 different AFIs in
one Message today.

I'm sure this discussion is related to draft-ietf-idr-error-handling.
This draft does not call for a BGP Capability Code.  It needs one if
it expects the neighbor to behave in a certain way which differs from
existing BGP requirements.  I have posted some discussion about this
on the IDR list recently but I do not read GROW.

-- 
Jeff S Wheeler <jsw@inconcepts.biz>
Sr Network Operator  /  Innovative Network Concepts

From rraszuk@gmail.com  Mon Dec 31 01:21:02 2012
Return-Path: <rraszuk@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6144021F893E; Mon, 31 Dec 2012 01:21:02 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.832
X-Spam-Level: 
X-Spam-Status: No, score=-2.832 tagged_above=-999 required=5 tests=[AWL=-0.082, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OjuKeK-92tC9; Mon, 31 Dec 2012 01:21:01 -0800 (PST)
Received: from mail-ie0-f181.google.com (mail-ie0-f181.google.com [209.85.223.181]) by ietfa.amsl.com (Postfix) with ESMTP id CE67321F8925; Mon, 31 Dec 2012 01:21:00 -0800 (PST)
Received: by mail-ie0-f181.google.com with SMTP id 16so14698976iea.26 for <multiple recipients>; Mon, 31 Dec 2012 01:21:00 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=RmD2W5z6CoMaWL+QwRXx9YKoRYRjgtls3NRES6QB1oM=; b=jZKZLwLXINKxhdaI0mFWDZF9AWS8b0+3TmCAR/4Oyc839FyusUHmzljHTxmfvt7ZkM kdDtRIVnQKL7GQiIsC761ObeX9959+o5WwgaUfrGGMS1Cth+Trh+tSA0v3v/eVB+jpvV K5zhN6CNkfICSGfvNHKTzzTzFQLMBYa3fjB1+HNFCSV2RRb2OGjEWXzkzAOlbb3zxZ09 DjQH/qxnhGCMyV5jVrGMUt+gHapu/lSzt9AyyCqIxEsUJnbWhHGHT0PRgLwGgXBcXVGc SRfcFqVUUwOW7mFoSkCpc8fjVSKj+QDL45IkAlClpwccIfaSCFwvGul98Z0XIYi+aZ7a MS0g==
MIME-Version: 1.0
Received: by 10.50.108.145 with SMTP id hk17mr35426541igb.51.1356945660446; Mon, 31 Dec 2012 01:21:00 -0800 (PST)
Sender: rraszuk@gmail.com
Received: by 10.64.167.204 with HTTP; Mon, 31 Dec 2012 01:21:00 -0800 (PST)
In-Reply-To: <20121231030925.GB62942@verdi>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com> <20121231030925.GB62942@verdi>
Date: Mon, 31 Dec 2012 10:21:00 +0100
X-Google-Sender-Auth: Zt1RVNCfd_vE5MUKIO9VyblbZOY
Message-ID: <CA+b+ERkYE61CSCxmZosxRO1LHhddRLsNmXYtB_nnBa8pVpy4Hw@mail.gmail.com>
From: Robert Raszuk <robert@raszuk.net>
To: John Leslie <john@jlc.net>, Brian Dickson <brian.peter.dickson@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Cc: idr@ietf.org, grow@ietf.org
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 09:21:02 -0000

How about if we would mandate Enhance Route Refresh request to be send
to such peer who's bad updates triggered treat-as-withdraw action ?

If we resync databases OK the likehood of "bad things" should be
minimal. If we fail to sync (get another malformed update(s)) then
IMHO resetting the session should be granted.

Yes that means that "treat-as-withdraw" should be applied only to
those peers which support Enhanced Route Refresh.

Happy New Year,
R.


> On Mon, Dec 31, 2012 at 4:09 AM, John Leslie <john@jlc.net> wrote:
>
> Brian Dickson <brian.peter.dickson@gmail.com> wrote:
>>
>> But, the basic problem is this: missing an UPDATE won't trigger either
>> condition, it will at worst cause sub-optimal routing. Missing a WITHDRAW
>> _CAN_ cause Bad Things (TM) to happen.
>
>    +1
>
> --
> John Leslie <john@jlc.net>
> _______________________________________________
> Idr mailing list
> Idr@ietf.org
> https://www.ietf.org/mailman/listinfo/idr

From rjs@rob.sh  Mon Dec 31 02:52:11 2012
Return-Path: <rjs@rob.sh>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8BBA821F85E6; Mon, 31 Dec 2012 02:52:11 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.429
X-Spam-Level: 
X-Spam-Status: No, score=-1.429 tagged_above=-999 required=5 tests=[AWL=0.343,  BAYES_00=-2.599, J_CHICKENPOX_13=0.6, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Rxbr5SBXvMYf; Mon, 31 Dec 2012 02:52:10 -0800 (PST)
Received: from cappuccino.rob.sh (cappuccino.rob.sh [IPv6:2001:b98:201:101::10:cafe]) by ietfa.amsl.com (Postfix) with ESMTP id 9752821F85DC; Mon, 31 Dec 2012 02:52:10 -0800 (PST)
Received: from [46.65.174.227] (helo=latte.munster) by cappuccino.rob.sh with esmtpa (Exim 4.72) (envelope-from <rjs@rob.sh>) id 1TpcvP-0008VP-HY; Mon, 31 Dec 2012 10:48:55 +0000
Mime-Version: 1.0 (Apple Message framework v1283)
Content-Type: text/plain; charset=windows-1252
From: Rob Shakir <rjs@rob.sh>
In-Reply-To: <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com>
Date: Mon, 31 Dec 2012 10:52:08 +0000
Content-Transfer-Encoding: quoted-printable
Message-Id: <6EC403F6-43EE-4A70-860A-3196B1CF266C@rob.sh>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com>
To: Brian Dickson <brian.peter.dickson@gmail.com>, Chris Hall <chris.hall@highwayman.com>, grow@ietf.org
X-Mailer: Apple Mail (2.1283)
Cc: idr@ietf.org
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 10:52:11 -0000

On 30 Dec 2012, at 21:34, Brian Dickson wrote:

> But, the basic problem is this: missing an UPDATE won't trigger either =
condition, it will at worst cause sub-optimal routing. Missing a =
WITHDRAW _CAN_ cause Bad Things (TM) to happen.

If we tear down the session based on a single bad UPDATE, then Bad =
Things=99 happen. Worse still, these bad things happen to  NLRI that are =
not associated with the UPDATE that was erroneous and were supporting =
live customer services. With modern deployments of BGP, the sessions =
that are torn down can be relatively fundamental to network operation, =
carry multiple (unrelated) customer topologies, and have significant =
recovery times. Based on this, the tear-down behaviour has been shown to =
have harmful effects to the operation of real networks based on real =
world errors [0], and as per the draft that this thread is discussing. =
There are a number of deployments in which this is not acceptable.

We're spinning round the same loop here that was discussed a number of =
times. Let me try and summarise this as accurately as I  can:
=20
As soon as we accept any kind of mechanism that targets error handling =
to particular NLRI, we KNOW that we will compromise correctness. (As per =
Section 4.1 of the draft) We also KNOW that this has a cost - which is =
that we cannot be sure that the routing system is entirely consistent =
any more, and we might end up with stale NLRI, routing loops, or =
blackholing. The operational requirement described in this draft =
highlight that this _IS_ the cost of robustness, and highlight the need =
for operational monitoring and recovery mechanisms to support this =
relaxation. Further to this, they highlight ways that the current =
session-reset behaviour can be improved, particularly relevant in cases =
where we CANNOT apply more targeted error handling.

We are discussing whether we can wholly address the "when there has been =
an error in an UPDATE, NOTHING is certain" - I would assert that we do =
not need to. The fundamental premise here is that we are compromising =
protocol correctness, which has a cost in terms consistency. =
Operationally, paying this cost in order to keep numerous other services =
up and running is acceptable. If this cost is not acceptable to an =
operator, then such mechanisms SHOULD NOT be deployed.

There is an argument that says that we could use a modified =
session-reset based on NOTIFICATION after erroneous UPDATE approach for =
all errors, however, since we also have a concern around longer-lived =
errors (that are sourced by a remote speaker, and hence just repeated on =
session reset), and the scalability of BGP routers, optimising such that =
where it is possible to do so, we make minimal compromises to =
correctness (i.e.., we compromise the correctness of a limited subset of =
NLRI, where we can identify them) and maintain sessions is a key =
requirement.

If there are ways that we can improve the message packing in BGP such =
that we minimise the risk of compromising correctness with NLRI-targeted =
error handling, absolutely we should try and pursue these in =
draft-ietf-idr-error-handling - however, I think that there is consensus =
that error handling as a problem space is something we need to address, =
since there are work items for this in both IDR and GROW.

Would the GROW working group be happy if we address Chris' concern =
related to "lost NLRI" (which AFAICS is really the case where we have >1 =
type of NLRI attribute within a single message) by adding a note that an =
error remains Non-Critical if _at least one_ NLRI attributes can be =
successfully parsed? I'm unclear here as to whether we're addressing a =
common situation where multiple sets of NLRI are being contained within =
a single message [1]. If so, then adding "at least one" and then a =
further point that an implementation SHOULD use a single NLRI attribute =
per UPDATE message, and put this at the start of the attributes would =
seem to be a fair way forward.

Does this sound a reasonable way forward? If so, I will update this in =
an -07, as well as addressing the editorial issues that Chris =
highlighted earlier.

Many thanks for your feedback.

HNY!
r.

[0]: I enumerated a small number of real-world incidents back in 2011 =
when I presented this work at NANOG (slides: =
http://rob.sh/files/nanog-slides.pdf). There have been numerous =
incidents since, alongside many incidents that the operators involved =
were not willing to share more public details of. It would be a massive =
shame to me that when the inevitable next incident like this turns up in =
the DFZ, or internally to a SP network, we are still questioning the =
validity of the problem space.

[1]: Trawling a number PCAPs that I have, I can't find an UPDATE that =
contained MP_REACH_NLRI and MP_UNREACH_NLRI simultaneously from a =
multi-vendor L3VPN network. Another operator informs me that they have a =
certain implementation that appears to include both in a single message.



From rraszuk@gmail.com  Mon Dec 31 04:49:44 2012
Return-Path: <rraszuk@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EAF6B21F8809; Mon, 31 Dec 2012 04:49:44 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.83
X-Spam-Level: 
X-Spam-Status: No, score=-2.83 tagged_above=-999 required=5 tests=[AWL=-0.080,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Xqa1PO5TkSlm; Mon, 31 Dec 2012 04:49:44 -0800 (PST)
Received: from mail-ia0-f174.google.com (mail-ia0-f174.google.com [209.85.210.174]) by ietfa.amsl.com (Postfix) with ESMTP id 2DE9821F87FD; Mon, 31 Dec 2012 04:49:44 -0800 (PST)
Received: by mail-ia0-f174.google.com with SMTP id y25so10539836iay.33 for <multiple recipients>; Mon, 31 Dec 2012 04:49:42 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=VR7fuH6mLNCBx17lAuSQ6nBDmEdzB8aPGXNX7F7+D0w=; b=K0/BZXw6MGF7Y2EtCccLcV+oU8oDOdiKlVbXlzFgsEZtll5/ssy5S6xKqz59cRQwF/ 747sUDd1qLTuglJuUyzLYRiq9imudmUYnu7Y5WX1HGJFwKFKWMrz9xnGbt+LfTzcqV6S NrZm38Q1s5LDhDMakh6eu774FbEAxjeNJIQ6r1AqhK4C5ba996Gp3paNzcdATaBsoVhZ TqV1oIoVXHov9TjU90I+GYLQdtrXaPtptz7kqlzMx8KIqOtJ9c5O2S/kHQCFDnTDSfpa YmMrMod0oi6jrAvuUXa2nCxR+uiNtbct9f29dakjZjamFGaTHmp/Gc56h/eQ9YM6hcnT nkOQ==
MIME-Version: 1.0
Received: by 10.42.52.204 with SMTP id k12mr31273105icg.3.1356958182107; Mon, 31 Dec 2012 04:49:42 -0800 (PST)
Sender: rraszuk@gmail.com
Received: by 10.64.167.204 with HTTP; Mon, 31 Dec 2012 04:49:41 -0800 (PST)
In-Reply-To: <6EC403F6-43EE-4A70-860A-3196B1CF266C@rob.sh>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com> <6EC403F6-43EE-4A70-860A-3196B1CF266C@rob.sh>
Date: Mon, 31 Dec 2012 13:49:41 +0100
X-Google-Sender-Auth: rkijlS_drze8eJsD437mDLMlkO0
Message-ID: <CA+b+ERmd=YiFaHX-izQEYvmeJm4sO67vudD_Aj8SzJhYhB+Ujg@mail.gmail.com>
From: Robert Raszuk <robert@raszuk.net>
To: Rob Shakir <rjs@rob.sh>
Content-Type: text/plain; charset=ISO-8859-1
Cc: idr@ietf.org, grow@ietf.org, Chris Hall <chris.hall@highwayman.com>
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 12:49:45 -0000

Hi Rob,

> we are still questioning the validity of the problem space.

I am not sure where do you draw such impression from ? I think we are
all clear that there is problem .. from time to time a bleeding wound
appears.

However it seems that there is serious concern if the cure proposed is
the right one. It appears that the treat-as-withdraw is a piece of
duck tape so the rest of the body can sort of function. One could
argue that while for some this may even work if the wound is small,
imagine such duck tape being applied on 1 meter wound ? Moreover such
cut will not recover under duck tape.

So to me it seems this is a valid debate if it is better to reset the
guy and send him to hospital for proper bandage or stitching while
counting on the co-pilot(s) (other BGP sessions) to seamlessly carry
the services for customers in the mean time.

I think your draft and even the proposed solution is the best for the
current BGP4 transport of messages for the cases where some bug is
remote and you would likely end-up resetting perhaps all of your
upstream sessions .. no doubt. But we have no way of knowing that
network wide and I think this one of the problems. Clearly in this
case it is better to put even duck tape on the pilots so they can try
to land rather then sit and watch the plane crash.

Actually real life example:

I have discovered last week in one of BGP implementation that BGP
conditional advertisement will not kick in till the trigger  (check
condition) disappears completely from BGP table even if it's next hop
is no longer present in global RIB. One could call this a bug, but it
is one example where while today I can automatically prefer other exit
when session goes down with treat-as-withdraw and keep-the-session-up
this will no longer work as trigger route will happily continue to sit
in BGP table. And with treat-as-withdraw being default in some cases
after new software upgrade your expected BGP behavior will change. I
doubt you will find this example described in the release notes too ;)

Best,
R.

From rjs@rob.sh  Mon Dec 31 07:46:49 2012
Return-Path: <rjs@rob.sh>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8A9EF21F8877; Mon, 31 Dec 2012 07:46:49 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.797
X-Spam-Level: 
X-Spam-Status: No, score=-1.797 tagged_above=-999 required=5 tests=[AWL=0.575,  BAYES_00=-2.599, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LFm-BKNU1BfC; Mon, 31 Dec 2012 07:46:49 -0800 (PST)
Received: from cappuccino.rob.sh (cappuccino.rob.sh [IPv6:2001:b98:201:101::10:cafe]) by ietfa.amsl.com (Postfix) with ESMTP id CC57221F84E9; Mon, 31 Dec 2012 07:46:43 -0800 (PST)
Received: from [46.65.174.227] (helo=latte.munster) by cappuccino.rob.sh with esmtpa (Exim 4.72) (envelope-from <rjs@rob.sh>) id 1TphWU-0001Go-4d; Mon, 31 Dec 2012 15:43:30 +0000
Mime-Version: 1.0 (Apple Message framework v1283)
Content-Type: text/plain; charset=iso-8859-1
From: Rob Shakir <rjs@rob.sh>
In-Reply-To: <CA+b+ERmd=YiFaHX-izQEYvmeJm4sO67vudD_Aj8SzJhYhB+Ujg@mail.gmail.com>
Date: Mon, 31 Dec 2012 15:46:42 +0000
Content-Transfer-Encoding: quoted-printable
Message-Id: <31A90AE9-ACBC-4C5F-9A3D-A68CDB391F2C@rob.sh>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com> <6EC403F6-43EE-4A70-860A-3196B1CF266C@rob.sh> <CA+b+ERmd=YiFaHX-izQEYvmeJm4sO67vudD_Aj8SzJhYhB+Ujg@mail.gmail.com>
To: Robert Raszuk <robert@raszuk.net>
X-Mailer: Apple Mail (2.1283)
Cc: idr@ietf.org, grow@ietf.org, Chris Hall <chris.hall@highwayman.com>
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 15:46:49 -0000

Hi Robert,

On 31 Dec 2012, at 12:49, Robert Raszuk wrote:

> I think your draft and even the proposed solution is the best for the
> current BGP4 transport of messages for the cases where some bug is
> remote and you would likely end-up resetting perhaps all of your
> upstream sessions .. no doubt. But we have no way of knowing that
> network wide and I think this one of the problems. Clearly in this
> case it is better to put even duck tape on the pilots so they can try
> to land rather then sit and watch the plane crash.

The previous discussions on this subject would seem to indicate that =
there are a significant number of cases where treat-as-withdraw is seen =
as the best approach for both non-Internet and Internet deployments. In =
my view, the presence of other paths for particular NLRI does not =
necessarily imply that they are equally viable (for both capacity and =
cost reasons) - so maintaining as much of the network in the normal =
operating state seems the most beneficial approach. To do this we need =
some form of NLRI-level error handling.

There are undoubtedly some deployment scenarios where it is not what the =
operator wants for their network - and this may be the case for your =
example(s). The requirements draft is not asserting that this is a MUST =
and should be the default -- but rather outlining the motivation for =
those operators who want error handling that applies to particular NLRI, =
as well as the tools that are needed to make such error handling =
deployable. If your preference is to err more on the side of caution (as =
an operator, or an implementor) then there is no reason that one could =
not always just have session-level error handling, with Enhanced GR =
deployed within your network.

I think we need to have both approaches - and note the caveats around =
each. At the moment, the approach of having no choice for operators is =
concerning to me.

Kind regards,
r.=

From rjs@rob.sh  Mon Dec 31 09:33:30 2012
Return-Path: <rjs@rob.sh>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0623721F88B1; Mon, 31 Dec 2012 09:33:30 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.893
X-Spam-Level: 
X-Spam-Status: No, score=-1.893 tagged_above=-999 required=5 tests=[AWL=0.479,  BAYES_00=-2.599, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9wd2IkGe2K6B; Mon, 31 Dec 2012 09:33:29 -0800 (PST)
Received: from cappuccino.rob.sh (cappuccino.rob.sh [IPv6:2001:b98:201:101::10:cafe]) by ietfa.amsl.com (Postfix) with ESMTP id 8217021F86BA; Mon, 31 Dec 2012 09:33:29 -0800 (PST)
Received: from [46.65.174.227] (helo=latte.munster) by cappuccino.rob.sh with esmtpa (Exim 4.72) (envelope-from <rjs@rob.sh>) id 1TpjBp-0001jm-EE; Mon, 31 Dec 2012 17:30:17 +0000
Mime-Version: 1.0 (Apple Message framework v1283)
Content-Type: text/plain; charset=windows-1252
From: Rob Shakir <rjs@rob.sh>
In-Reply-To: <CAPWAtbK4kOM2ruYGHSt8FuG+65GU0wpOYnNa1euodeZCp8OFdQ@mail.gmail.com>
Date: Mon, 31 Dec 2012 17:33:30 +0000
Content-Transfer-Encoding: quoted-printable
Message-Id: <A7A2EEC9-E669-4358-B6C0-981C6FA05DD2@rob.sh>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com> <6EC403F6-43EE-4A70-860A-3196B1CF266C@rob.sh> <CAPWAtbK4kOM2ruYGHSt8FuG+65GU0wpOYnNa1euodeZCp8OFdQ@mail.gmail.com>
To: Jeff Wheeler <jsw@inconcepts.biz>
X-Mailer: Apple Mail (2.1283)
Cc: idr@ietf.org, grow@ietf.org
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 17:33:30 -0000

On 31 Dec 2012, at 16:59, Jeff Wheeler wrote:

> On Mon, Dec 31, 2012 at 5:52 AM, Rob Shakir <rjs@rob.sh> wrote:
>> If we tear down the session based on a single bad UPDATE, then Bad =
Things=99 happen.
>=20
> This is why I'm confused why you aren't more interested in giving the
> vendor (ultimately, the operator) the choice to simply ignore bad
> updates.
>=20
> You know that error-handling already leaves the RIB (network) in a bad
> state.  You might as well give the option to reset the session under
> even fewer circumstances.

This is, of course, another route that we could go down, and has been =
discussed previously. If we receive an UPDATE we know that *something* =
to do with this path has changed, and if this UPDATE is erroneous know =
that *something* about the signalling path is not correct. If we do have =
other paths to that particular prefix, then treat-as-withdraw lets us =
use these. If we simply ignore the UPDATE, then we trust this path that =
has something suspect about it just as much as those that we have not =
seen any error in the messaging for.

Cheers,
r.=

From jakob.heitz@ericsson.com  Mon Dec 31 09:55:08 2012
Return-Path: <jakob.heitz@ericsson.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8C4FE21F8792; Mon, 31 Dec 2012 09:55:08 -0800 (PST)
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=[AWL=0.062,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IsRDT8uECIof; Mon, 31 Dec 2012 09:55:07 -0800 (PST)
Received: from imr4.ericy.com (imr4.ericy.com [198.24.6.9]) by ietfa.amsl.com (Postfix) with ESMTP id A5D7A21F85DC; Mon, 31 Dec 2012 09:55:07 -0800 (PST)
Received: from EUSAAHC004.ericsson.se ([147.117.188.84]) by imr4.ericy.com (8.14.3/8.14.3/Debian-9.1ubuntu1) with ESMTP id qBVI8Q1S010951; Mon, 31 Dec 2012 12:08:27 -0600
Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC004.ericsson.se ([147.117.188.84]) with mapi id 14.02.0318.004; Mon, 31 Dec 2012 12:55:00 -0500
From: Jakob Heitz <jakob.heitz@ericsson.com>
To: Robert Raszuk <robert@raszuk.net>
Thread-Topic: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
Thread-Index: AQHN5zgrlXyLlAlYv0yHiK+hIwQIfZgzMXY6
Date: Mon, 31 Dec 2012 17:54:59 +0000
Message-ID: <6D642BD9-C87F-4987-9C32-0A4463E6A542@ericsson.com>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com> <20121231030925.GB62942@verdi>, <CA+b+ERkYE61CSCxmZosxRO1LHhddRLsNmXYtB_nnBa8pVpy4Hw@mail.gmail.com>
In-Reply-To: <CA+b+ERkYE61CSCxmZosxRO1LHhddRLsNmXYtB_nnBa8pVpy4Hw@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: "idr@ietf.org" <idr@ietf.org>, "grow@ietf.org" <grow@ietf.org>, John Leslie <john@jlc.net>
Subject: Re: [GROW] [Idr] I-D Action:	draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 17:55:08 -0000

I don't think treat-as-withdraw is trying to fix a single session reset. Gr=
aceful restart can fix that. It's the rolling resets that need a human to r=
emove a buggy router or a config that triggered the bug. That takes several=
 hours. Treat-as-withdraw limits the damage during those hours.

Could we please settle on that without trying to solve the impossible?

Cheers,
Jakob

Sent from my iPoo.

On Dec 31, 2012, at 1:21 AM, "Robert Raszuk" <robert@raszuk.net> wrote:

> How about if we would mandate Enhance Route Refresh request to be send
> to such peer who's bad updates triggered treat-as-withdraw action ?
>=20
> If we resync databases OK the likehood of "bad things" should be
> minimal. If we fail to sync (get another malformed update(s)) then
> IMHO resetting the session should be granted.
>=20
> Yes that means that "treat-as-withdraw" should be applied only to
> those peers which support Enhanced Route Refresh.
>=20
> Happy New Year,
> R.
>=20
>=20
>> On Mon, Dec 31, 2012 at 4:09 AM, John Leslie <john@jlc.net> wrote:
>>=20
>> Brian Dickson <brian.peter.dickson@gmail.com> wrote:
>>>=20
>>> But, the basic problem is this: missing an UPDATE won't trigger either
>>> condition, it will at worst cause sub-optimal routing. Missing a WITHDR=
AW
>>> _CAN_ cause Bad Things (TM) to happen.
>>=20
>>   +1
>>=20
>> --
>> John Leslie <john@jlc.net>
>> _______________________________________________
>> Idr mailing list
>> Idr@ietf.org
>> https://www.ietf.org/mailman/listinfo/idr
> _______________________________________________
> GROW mailing list
> GROW@ietf.org
> https://www.ietf.org/mailman/listinfo/grow

From jakob.heitz@ericsson.com  Mon Dec 31 10:15:37 2012
Return-Path: <jakob.heitz@ericsson.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3CBB421F8830; Mon, 31 Dec 2012 10:15:37 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.312
X-Spam-Level: 
X-Spam-Status: No, score=-6.312 tagged_above=-999 required=5 tests=[AWL=0.060,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iSvUK919k67i; Mon, 31 Dec 2012 10:15:36 -0800 (PST)
Received: from imr3.ericy.com (imr3.ericy.com [198.24.6.13]) by ietfa.amsl.com (Postfix) with ESMTP id ADBCA21F8738; Mon, 31 Dec 2012 10:15:36 -0800 (PST)
Received: from EUSAAHC003.ericsson.se ([147.117.188.81]) by imr3.ericy.com (8.13.8/8.13.8) with ESMTP id qBVIFXs9020835 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 31 Dec 2012 12:15:34 -0600
Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC003.ericsson.se ([147.117.188.81]) with mapi id 14.02.0318.004; Mon, 31 Dec 2012 13:15:33 -0500
From: Jakob Heitz <jakob.heitz@ericsson.com>
To: Robert Raszuk <robert@raszuk.net>
Thread-Topic: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
Thread-Index: AQHN5tV/uxLXJIv0r0ynEwftCG2icZgzD+gAgAAg2ICAAAc6qg==
Date: Mon, 31 Dec 2012 18:15:32 +0000
Message-ID: <E3F7ABAD-E653-4E28-B2FE-498AB188FAC4@ericsson.com>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com> <6EC403F6-43EE-4A70-860A-3196B1CF266C@rob.sh>, <CA+b+ERmd=YiFaHX-izQEYvmeJm4sO67vudD_Aj8SzJhYhB+Ujg@mail.gmail.com>
In-Reply-To: <CA+b+ERmd=YiFaHX-izQEYvmeJm4sO67vudD_Aj8SzJhYhB+Ujg@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: "idr@ietf.org" <idr@ietf.org>, "grow@ietf.org" <grow@ietf.org>, Chris Hall <chris.hall@highwayman.com>
Subject: Re: [GROW] [Idr] I-D Action:	draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 18:15:37 -0000

On Dec 31, 2012, at 4:49 AM, "Robert Raszuk" <robert@raszuk.net> wrote:
>=20
> I have discovered last week in one of BGP implementation that BGP
> conditional advertisement will not kick in till the trigger  (check
> condition) disappears completely from BGP table even if it's next hop
> is no longer present in global RIB. One could call this a bug,

Before we even consider this case, perhaps someone could update (and explai=
n) rfc4271 to remove the following sentence from page 76:

The function that calculates the degree of preference for a given route SHA=
LL NOT use any of the following as its inputs: the existence of other route=
s, the non-existence of other routes, or the path attributes of other route=
s.

Cheers,
Jakob.=

From jakob.heitz@ericsson.com  Mon Dec 31 10:35:44 2012
Return-Path: <jakob.heitz@ericsson.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 687AF21F865D; Mon, 31 Dec 2012 10:35:44 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.313
X-Spam-Level: 
X-Spam-Status: No, score=-6.313 tagged_above=-999 required=5 tests=[AWL=0.059,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IkN+DAzCno+U; Mon, 31 Dec 2012 10:35:43 -0800 (PST)
Received: from imr3.ericy.com (imr3.ericy.com [198.24.6.13]) by ietfa.amsl.com (Postfix) with ESMTP id CAE4221F8651; Mon, 31 Dec 2012 10:35:43 -0800 (PST)
Received: from EUSAAHC005.ericsson.se ([147.117.188.87]) by imr3.ericy.com (8.13.8/8.13.8) with ESMTP id qBVIZgs6021549 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 31 Dec 2012 12:35:43 -0600
Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC005.ericsson.se ([147.117.188.87]) with mapi id 14.02.0318.004; Mon, 31 Dec 2012 13:35:42 -0500
From: Jakob Heitz <jakob.heitz@ericsson.com>
To: Jeff Wheeler <jsw@inconcepts.biz>
Thread-Topic: [Idr] [GROW] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
Thread-Index: AQHN54KbFisNgDpWy0mj2cTgGt+EAZgzPEEA
Date: Mon, 31 Dec 2012 18:35:42 +0000
Message-ID: <F0D96A03-6117-4D72-B23D-C1A73C3C9697@ericsson.com>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com> <20121231030925.GB62942@verdi> <CA+b+ERkYE61CSCxmZosxRO1LHhddRLsNmXYtB_nnBa8pVpy4Hw@mail.gmail.com> <6D642BD9-C87F-4987-9C32-0A4463E6A542@ericsson.com>, <CAPWAtb+v4W4qVQ6ceTdewc4u_nBOyZLztXe_m7vT_QCna2WWfQ@mail.gmail.com>
In-Reply-To: <CAPWAtb+v4W4qVQ6ceTdewc4u_nBOyZLztXe_m7vT_QCna2WWfQ@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: "idr@ietf.org" <idr@ietf.org>, "grow@ietf.org" <grow@ietf.org>
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 18:35:44 -0000

The difference between ignore and treat-as-withdraw is discussed in the dra=
ft.
Anyway, the feature will be knobbed. So far, I see 2 knob options:
1. Treat-as-withdraw / ignore
2. If stream sync is lost, search ahead for the next 16*0xFF / reset the se=
ssion.

Cheers,
Jakob

On Dec 31, 2012, at 10:14 AM, "Jeff Wheeler" <jsw@inconcepts.biz> wrote:

> On Mon, Dec 31, 2012 at 12:54 PM, Jakob Heitz <jakob.heitz@ericsson.com> =
wrote:
>> I don't think treat-as-withdraw is trying to fix a single session reset.=
 Graceful restart can fix that. It's the rolling resets that need a human t=
o remove a buggy router or a config that triggered the bug. That takes seve=
ral hours. Treat-as-withdraw limits the damage during those hours.
>>=20
>> Could we please settle on that without trying to solve the impossible?
>=20
> If you read my posts to IDR on this topic, you'll see where I explain
> how it is possible to solve the "impossible."
>=20
> Specifically, you can ignore just about any bad update, or bad message
> of any kind, as long as you can figure out where the next message
> starts.  This fixes the rolling resets.
>=20
> You may know that a lot of businesses suffered multi-hour outages in
> October simply because of 5 DFZ routes announced by LANL that had
> illegal attributes.  This is very hard for operators to troubleshoot
> on most routers.  The routers that experienced rolling resets were
> buggy but if the operators simply had a panic button, "ignore bad
> messages," their networks would have been up and they would not have
> been losing money by the minute.
>=20
> This is not the only time bad updates have propagated through the DFZ
> and caused big problems.  It has happened repeatedly.
>=20
> I believe it will begin to happen more often inside datacenter
> networks, because BGP is being used for more and more things, like
> EVPN.  Operators are going to need BGP to become more robust.
>=20
> You can make it a lot more robust just by deciding to ignore
> everything in a bad message.  This is not good, but it is a lot better
> than session-reset, in most cases.
>=20
> Please, read my posts on this topic, and do not treat this problem as
> an unsolvable one.  It can be largely solved in a way that gives a
> very useful fallback option to operators.
>=20
> This whole draft is about fallback options, and it is pretty stupid to
> have a large amount of complexity to solve a small set of potential
> bugs, when you could ALTERNATIVELY or IN ADDITION to that, have a very
> low-complexity option that solves more problems.
>=20
> --=20
> Jeff S Wheeler <jsw@inconcepts.biz>
> Sr Network Operator  /  Innovative Network Concepts

From rraszuk@gmail.com  Mon Dec 31 12:15:03 2012
Return-Path: <rraszuk@gmail.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A56E021F8596; Mon, 31 Dec 2012 12:15:03 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.828
X-Spam-Level: 
X-Spam-Status: No, score=-2.828 tagged_above=-999 required=5 tests=[AWL=-0.078, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OIJARpurcpjr; Mon, 31 Dec 2012 12:15:03 -0800 (PST)
Received: from mail-ie0-f179.google.com (mail-ie0-f179.google.com [209.85.223.179]) by ietfa.amsl.com (Postfix) with ESMTP id 1852C21F8566; Mon, 31 Dec 2012 12:15:03 -0800 (PST)
Received: by mail-ie0-f179.google.com with SMTP id k14so15542014iea.10 for <multiple recipients>; Mon, 31 Dec 2012 12:15:02 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=gKfskRheaoGxzT2b0t2PGGxiDDULkKIcRuvGYSsSaAE=; b=lNh3KdkN02SV9nxKsrI/Z9P11oh5keI7bRC+6kXcty7Y06sM0jjZbQi2Qn8ku8w6r7 /Ww7N6ci02u9RCrqlMM76876JvjadBrSO5QI77QsYMLnEFrkP9wMeTYoJ3pSCmoDy9vS 5pjG4sH1Jg77IntS/KT3ZMJj4IeWbst347RVHpc5NtrEp691MzxGqh1q8bEcGkL26bFc A+vZuTUXGf7Xla5L/9iwGzttKjjFm9kw9ty5+PQUSjGVBd04wQO1PMHyHNibPoI+edCu HeVoppy2kivuCq3dWPKYBwr77kRfjtm83PvAhdGAA8LSDmF5mtUXECjDeHp/5+xbhlcq 9b4A==
MIME-Version: 1.0
Received: by 10.42.72.132 with SMTP id o4mr21274564icj.44.1356984902692; Mon, 31 Dec 2012 12:15:02 -0800 (PST)
Sender: rraszuk@gmail.com
Received: by 10.64.167.204 with HTTP; Mon, 31 Dec 2012 12:15:02 -0800 (PST)
In-Reply-To: <CAPWAtbJTE1mnkHMHVfh0-66HmtG4ERTFbODRPYay42g2khJnnA@mail.gmail.com>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com> <6D642BD9-C87F-4987-9C32-0A4463E6A542@ericsson.com> <20121231184138.GA25817@verdi> <CAPWAtbJTE1mnkHMHVfh0-66HmtG4ERTFbODRPYay42g2khJnnA@mail.gmail.com>
Date: Mon, 31 Dec 2012 21:15:02 +0100
X-Google-Sender-Auth: dVap1hCJix_NbCKbDPyAbpFxNzs
Message-ID: <CA+b+ERmDAu3+kKcDmY1EkuniZNPXSKz0WTTLZHuJcSDTm7_=AQ@mail.gmail.com>
From: Robert Raszuk <robert@raszuk.net>
To: Jeff Wheeler <jsw@inconcepts.biz>
Content-Type: text/plain; charset=ISO-8859-1
Cc: "idr@ietf.org" <idr@ietf.org>, "grow@ietf.org" <grow@ietf.org>, John Leslie <john@jlc.net>
Subject: Re: [GROW] [Idr] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 20:15:03 -0000

Hi Jeff,

> That was a bug, but the customers had NO OPTIONS TO WORK AROUND THIS.
> Their ONLY CHOICE was to wait for a patch from their vendor (for folks
> whose vendor routers did this)

+

> the operator can buy time to diagnose it in detail, or wait for his
> vendor to provide a software update, or whatever.

I think you have just hit the nail on it's head above.

I think waiting for vendor's fix is just not acceptable for production
environments .. it just takes too long [read days if not weeks] (not
to mention that a lot of deployed hardware requires reload when you
upgrade).

When we originally discussed error handling solutions in junisco one
option was to allow operator to apply filter to BGP messages between
TCP and BGP. If the error is known it is pretty trivial to match on
arbitrary string in such BGP pre-parser and filter it out or clear or
reset the bit/bits if NOC decides it is operationally safe.

The point is that for critical protocol like BGP4 (with current choice
of transport) I am of the opinion that we should build a much more
universal mechanism rather then add fixed twicks which may or may not
always help and when operator has zero control on what such "session
saver" twick actually does.

Cheers,
R.

From jakob.heitz@ericsson.com  Mon Dec 31 12:24:40 2012
Return-Path: <jakob.heitz@ericsson.com>
X-Original-To: grow@ietfa.amsl.com
Delivered-To: grow@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3EB8021F8692; Mon, 31 Dec 2012 12:24:39 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.32
X-Spam-Level: 
X-Spam-Status: No, score=-6.32 tagged_above=-999 required=5 tests=[AWL=0.052,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, SARE_SUB_OBFU_Q1=0.227]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uiMGSAWt2Fpe; Mon, 31 Dec 2012 12:24:38 -0800 (PST)
Received: from imr4.ericy.com (imr4.ericy.com [198.24.6.9]) by ietfa.amsl.com (Postfix) with ESMTP id 89FE021F8470; Mon, 31 Dec 2012 12:24:38 -0800 (PST)
Received: from EUSAAHC007.ericsson.se ([147.117.188.93]) by imr4.ericy.com (8.14.3/8.14.3/Debian-9.1ubuntu1) with ESMTP id qBVKbwdN018281; Mon, 31 Dec 2012 14:37:59 -0600
Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC007.ericsson.se ([147.117.188.93]) with mapi id 14.02.0318.004; Mon, 31 Dec 2012 15:24:30 -0500
From: Jakob Heitz <jakob.heitz@ericsson.com>
To: Jeff Wheeler <jsw@inconcepts.biz>, John Leslie <john@jlc.net>
Thread-Topic: [Idr] [GROW] I-D Action: draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
Thread-Index: AQHN55FWi/J3CnHlp0Sb55SRXym/4ZgzWYog
Date: Mon, 31 Dec 2012 20:24:30 +0000
Message-ID: <2F3EBB88EC3A454AAB08915FBF0B8C7E147337@eusaamb109.ericsson.se>
References: <025801cde4f6$d7783700$8668a500$@highwayman.com> <3B3C1AA7-40DB-4566-9BD0-C25E5845935D@rob.sh> <028701cde525$bf262070$3d726150$@highwayman.com> <CAH1iCip=sm9rEDw8GnmejWU62sLJ32MZNOJ6EPsyF9b1MQWCZg@mail.gmail.com> <EC5968D3-A05B-4942-BE7A-B0BF1277330A@rob.sh> <CAH1iCir+wpd2LP4dXXQWtXrOt4ZfuUMN4OU4FWWuGbD7nLErQg@mail.gmail.com> <6D642BD9-C87F-4987-9C32-0A4463E6A542@ericsson.com> <20121231184138.GA25817@verdi> <CAPWAtbJTE1mnkHMHVfh0-66HmtG4ERTFbODRPYay42g2khJnnA@mail.gmail.com>
In-Reply-To: <CAPWAtbJTE1mnkHMHVfh0-66HmtG4ERTFbODRPYay42g2khJnnA@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [147.117.188.135]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: "idr@ietf.org" <idr@ietf.org>, "grow@ietf.org" <grow@ietf.org>
Subject: Re: [GROW] [Idr] I-D Action:	draft-ietf-grow-ops-reqs-for-bgp-error-handling-06.txt
X-BeenThere: grow@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Grow Working Group Mailing List <grow.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/grow>, <mailto:grow-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/grow>
List-Post: <mailto:grow@ietf.org>
List-Help: <mailto:grow-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/grow>, <mailto:grow-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 31 Dec 2012 20:24:40 -0000

On , Jeff Wheeler <> wrote:

> On Mon, Dec 31, 2012 at 1:41 PM, John Leslie <john@jlc.net> wrote:
> Malformed Attribute Flags are not
> one of the criteria it addresses.=20

http://tools.ietf.org/html/draft-ietf-idr-error-handling-03
says

   When a path attribute (other than the MP_REACH_NLRI attribute
   [RFC4760] or the MP_UNREACH_NLRI attribute [RFC4760]) in an UPDATE
   message is determined to be malformed, the UPDATE message containing
   that attribute MUST be treated as though all contained routes had
   been withdrawn...
   ...In the case of an
   attribute which has no effect on route selection or installation, the
   malformed attribute MAY instead be discarded...

That covers malformed attribute flags.


--=20
Jakob Heitz.
