
From nobody Tue Feb  1 08:45:10 2022
Return-Path: <touch@strayalpha.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C92943A15BD; Tue,  1 Feb 2022 08:45:08 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.523
X-Spam-Level: 
X-Spam-Status: No, score=-0.523 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HAS_X_OUTGOING_SPAM_STAT=0.795, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_NEUTRAL=0.779] autolearn=no autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=strayalpha.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id woECd-ym4JSL; Tue,  1 Feb 2022 08:45:04 -0800 (PST)
Received: from server217-1.web-hosting.com (server217-1.web-hosting.com [198.54.114.226]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3353F3A15B1; Tue,  1 Feb 2022 08:45:04 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=strayalpha.com; s=default; h=Message-Id:To:References:Subject:Date: Mime-Version:Content-Type:From:Sender:Reply-To:Cc:Content-Transfer-Encoding: Content-ID:Content-Description:Resent-Date:Resent-From:Resent-Sender: Resent-To:Resent-Cc:Resent-Message-ID:In-Reply-To:List-Id:List-Help: List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=ySUtlGARcRUoGCweYgNwTebWkHDDP08/6rHSHFdd79k=; b=CL9/1/+vQlWEkFHVYrGTw3kiUe /qZKIi4Q8eXjkeU/H804tG9RYI0CtrGpzAPQc/duIIZKgxd1iX5pYNq6NNWhDGqFKrDOQqQi9j5l5 lAZTLUFpDg8a/Qa/aWR6jRC7UGw8dpYELxLD9tpfHykIuq/N7l8RC4r/2zTKyOi72KASE3oifDbSV vXn0ZGj6YGL1YzWTsxAYhghjYBfhWsROZ6o5dVcUeY+rJUFUeHEztzpnC51uphoxmOmmdKLUTdFFA v3f0de970MtjGo2w/w1T5Q3BftcdcJJBrQ21vWfUI8f6uGhZlRAn0tQ21NXx/hfxU+2CmrSmswsux C7NKCcWQ==;
Received: from cpe-172-114-237-88.socal.res.rr.com ([172.114.237.88]:52043 helo=smtpclient.apple) by server217.web-hosting.com with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from <touch@strayalpha.com>) id 1nEwH5-000OJk-AY; Tue, 01 Feb 2022 11:45:03 -0500
From: "touch@strayalpha.com" <touch@strayalpha.com>
Content-Type: multipart/alternative; boundary="Apple-Mail=_AB10C52F-6964-4CDD-A555-E575BA521B6A"
Mime-Version: 1.0 (Mac OS X Mail 15.0 \(3693.40.0.1.81\))
Date: Tue, 1 Feb 2022 08:44:58 -0800
References: <20220201071749.E12CBECB8D@rfc-editor.org>
To: tcpm IETF list <tcpm@ietf.org>, TSVWG <tsvwg@ietf.org>
Message-Id: <13D17C3E-85B7-4832-87B7-7244529795E6@strayalpha.com>
X-Mailer: Apple Mail (2.3693.40.0.1.81)
X-OutGoing-Spam-Status: No, score=-1.0
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - server217.web-hosting.com
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - strayalpha.com
X-Get-Message-Sender-Via: server217.web-hosting.com: authenticated_id: touch@strayalpha.com
X-Authenticated-Sender: server217.web-hosting.com: touch@strayalpha.com
X-Source: 
X-Source-Args: 
X-Source-Dir: 
X-From-Rewrite: unmodified, already matched
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/NwAhRid_SuewgrfHkBJmz0WDCpk>
Subject: [tcpm] Fwd: RFC 9187 on Sequence Number Extension for Windowed Protocols
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 01 Feb 2022 16:45:09 -0000

--Apple-Mail=_AB10C52F-6964-4CDD-A555-E575BA521B6A
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

Hi, all,

The following may be of interest.

Joe

=E2=80=94
Joe Touch, temporal epistemologist
www.strayalpha.com

> Begin forwarded message:
>=20
> From: rfc-editor@rfc-editor.org
> Subject: RFC 9187 on Sequence Number Extension for Windowed Protocols
> Date: January 31, 2022 at 11:17:49 PM PST
> To: ietf-announce@ietf.org, rfc-dist@rfc-editor.org
> Cc: drafts-update-ref@iana.org, rfc-editor@rfc-editor.org
>=20
> A new Request for Comments is now available in online RFC libraries.
>=20
>=20
>        RFC 9187
>=20
>        Title:      Sequence Number Extension for Windowed Protocols=20
>        Author:     J. Touch
>        Status:     Informational
>        Stream:     Independent
>        Date:       January 2022
>        Mailbox:    touch@strayalpha.com
>        Pages:      10
>        Updates/Obsoletes/SeeAlso:   None
>=20
>        I-D Tag:    draft-touch-sne-02.txt
>=20
>        URL:        https://www.rfc-editor.org/info/rfc9187
>=20
>        DOI:        10.17487/RFC9187
>=20
> Sliding window protocols use finite sequence numbers to determine
> segment placement and order. These sequence number spaces wrap around
> and are reused during the operation of such protocols. This document
> describes a way to extend the size of these sequence numbers at the
> endpoints to avoid the impact of that wrap and reuse without
> transmitting additional information in the packet header. The
> resulting extended sequence numbers can be used at the endpoints in
> encryption and authentication algorithms to ensure input bit patterns
> do not repeat over the lifetime of a connection.
>=20
>=20
> INFORMATIONAL: This memo provides information for the Internet =
community.
> It does not specify an Internet standard of any kind. Distribution of
> this memo is unlimited.
>=20
> This announcement is sent to the IETF-Announce and rfc-dist lists.
> To subscribe or unsubscribe, see
>  https://www.ietf.org/mailman/listinfo/ietf-announce
>  https://mailman.rfc-editor.org/mailman/listinfo/rfc-dist
>=20
> For searching the RFC series, see https://www.rfc-editor.org/search
> For downloading RFCs, see https://www.rfc-editor.org/retrieve/bulk
>=20
> Requests for special distribution should be addressed to either the
> author of the RFC in question, or to rfc-editor@rfc-editor.org.  =
Unless
> specifically noted otherwise on the RFC itself, all RFCs are for
> unlimited distribution.
>=20
>=20
> The RFC Editor Team
> Association Management Solutions, LLC
>=20
>=20
> _______________________________________________
> IETF-Announce mailing list
> IETF-Announce@ietf.org
> https://www.ietf.org/mailman/listinfo/ietf-announce


--Apple-Mail=_AB10C52F-6964-4CDD-A555-E575BA521B6A
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=utf-8

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3Dutf-8"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: after-white-space;" class=3D"">Hi, =
all,<div class=3D""><br class=3D""></div><div class=3D"">The following =
may be of interest.</div><div class=3D""><br class=3D""></div><div =
class=3D"">Joe</div><div class=3D""><br class=3D""><div class=3D"">
<meta charset=3D"UTF-8" class=3D""><div dir=3D"auto" style=3D"caret-color:=
 rgb(0, 0, 0); color: rgb(0, 0, 0); letter-spacing: normal; text-align: =
start; text-indent: 0px; text-transform: none; white-space: normal; =
word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: =
none; word-wrap: break-word; -webkit-nbsp-mode: space; line-break: =
after-white-space;" class=3D"">=E2=80=94<div class=3D"">Joe Touch, =
temporal epistemologist<div class=3D""><a =
href=3D"http://www.strayalpha.com" =
class=3D"">www.strayalpha.com</a></div></div></div>
</div>

<div><br class=3D""><blockquote type=3D"cite" class=3D""><div =
class=3D"">Begin forwarded message:</div><br =
class=3D"Apple-interchange-newline"><div style=3D"margin-top: 0px; =
margin-right: 0px; margin-bottom: 0px; margin-left: 0px;" class=3D""><span=
 style=3D"font-family: -webkit-system-font, Helvetica Neue, Helvetica, =
sans-serif; color:rgba(0, 0, 0, 1.0);" class=3D""><b class=3D"">From: =
</b></span><span style=3D"font-family: -webkit-system-font, Helvetica =
Neue, Helvetica, sans-serif;" class=3D""><a =
href=3D"mailto:rfc-editor@rfc-editor.org" =
class=3D"">rfc-editor@rfc-editor.org</a><br class=3D""></span></div><div =
style=3D"margin-top: 0px; margin-right: 0px; margin-bottom: 0px; =
margin-left: 0px;" class=3D""><span style=3D"font-family: =
-webkit-system-font, Helvetica Neue, Helvetica, sans-serif; =
color:rgba(0, 0, 0, 1.0);" class=3D""><b class=3D"">Subject: =
</b></span><span style=3D"font-family: -webkit-system-font, Helvetica =
Neue, Helvetica, sans-serif;" class=3D""><b class=3D"">RFC 9187 on =
Sequence Number Extension for Windowed Protocols</b><br =
class=3D""></span></div><div style=3D"margin-top: 0px; margin-right: =
0px; margin-bottom: 0px; margin-left: 0px;" class=3D""><span =
style=3D"font-family: -webkit-system-font, Helvetica Neue, Helvetica, =
sans-serif; color:rgba(0, 0, 0, 1.0);" class=3D""><b class=3D"">Date: =
</b></span><span style=3D"font-family: -webkit-system-font, Helvetica =
Neue, Helvetica, sans-serif;" class=3D"">January 31, 2022 at 11:17:49 PM =
PST<br class=3D""></span></div><div style=3D"margin-top: 0px; =
margin-right: 0px; margin-bottom: 0px; margin-left: 0px;" class=3D""><span=
 style=3D"font-family: -webkit-system-font, Helvetica Neue, Helvetica, =
sans-serif; color:rgba(0, 0, 0, 1.0);" class=3D""><b class=3D"">To: =
</b></span><span style=3D"font-family: -webkit-system-font, Helvetica =
Neue, Helvetica, sans-serif;" class=3D""><a =
href=3D"mailto:ietf-announce@ietf.org" =
class=3D"">ietf-announce@ietf.org</a>, <a =
href=3D"mailto:rfc-dist@rfc-editor.org" =
class=3D"">rfc-dist@rfc-editor.org</a><br class=3D""></span></div><div =
style=3D"margin-top: 0px; margin-right: 0px; margin-bottom: 0px; =
margin-left: 0px;" class=3D""><span style=3D"font-family: =
-webkit-system-font, Helvetica Neue, Helvetica, sans-serif; =
color:rgba(0, 0, 0, 1.0);" class=3D""><b class=3D"">Cc: </b></span><span =
style=3D"font-family: -webkit-system-font, Helvetica Neue, Helvetica, =
sans-serif;" class=3D""><a href=3D"mailto:drafts-update-ref@iana.org" =
class=3D"">drafts-update-ref@iana.org</a>, <a =
href=3D"mailto:rfc-editor@rfc-editor.org" =
class=3D"">rfc-editor@rfc-editor.org</a><br class=3D""></span></div><br =
class=3D""><div class=3D""><div class=3D"">A new Request for Comments is =
now available in online RFC libraries.<br class=3D""><br class=3D""><br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;RFC 9187<br =
class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Title: =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Sequence Number Extension for Windowed =
Protocols <br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Author: =
&nbsp;&nbsp;&nbsp;&nbsp;J. Touch<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Status: =
&nbsp;&nbsp;&nbsp;&nbsp;Informational<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Stream: =
&nbsp;&nbsp;&nbsp;&nbsp;Independent<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Date: =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;January 2022<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Mailbox: &nbsp;&nbsp;&nbsp;<a =
href=3D"mailto:touch@strayalpha.com" =
class=3D"">touch@strayalpha.com</a><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Pages: =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;10<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Updates/Obsoletes/SeeAlso: =
&nbsp;&nbsp;None<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;I-D Tag: =
&nbsp;&nbsp;&nbsp;draft-touch-sne-02.txt<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;URL: =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;<a =
href=3D"https://www.rfc-editor.org/info/rfc9187" =
class=3D"">https://www.rfc-editor.org/info/rfc9187</a><br class=3D""><br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;DOI: =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;10.17487/RFC9187<br =
class=3D""><br class=3D"">Sliding window protocols use finite sequence =
numbers to determine<br class=3D"">segment placement and order. These =
sequence number spaces wrap around<br class=3D"">and are reused during =
the operation of such protocols. This document<br class=3D"">describes a =
way to extend the size of these sequence numbers at the<br =
class=3D"">endpoints to avoid the impact of that wrap and reuse =
without<br class=3D"">transmitting additional information in the packet =
header. The<br class=3D"">resulting extended sequence numbers can be =
used at the endpoints in<br class=3D"">encryption and authentication =
algorithms to ensure input bit patterns<br class=3D"">do not repeat over =
the lifetime of a connection.<br class=3D""><br class=3D""><br =
class=3D"">INFORMATIONAL: This memo provides information for the =
Internet community.<br class=3D"">It does not specify an Internet =
standard of any kind. Distribution of<br class=3D"">this memo is =
unlimited.<br class=3D""><br class=3D"">This announcement is sent to the =
IETF-Announce and rfc-dist lists.<br class=3D"">To subscribe or =
unsubscribe, see<br class=3D""> &nbsp;<a =
href=3D"https://www.ietf.org/mailman/listinfo/ietf-announce" =
class=3D"">https://www.ietf.org/mailman/listinfo/ietf-announce</a><br =
class=3D""> &nbsp;<a =
href=3D"https://mailman.rfc-editor.org/mailman/listinfo/rfc-dist" =
class=3D"">https://mailman.rfc-editor.org/mailman/listinfo/rfc-dist</a><br=
 class=3D""><br class=3D"">For searching the RFC series, see <a =
href=3D"https://www.rfc-editor.org/search" =
class=3D"">https://www.rfc-editor.org/search</a><br class=3D"">For =
downloading RFCs, see <a href=3D"https://www.rfc-editor.org/retrieve/bulk"=
 class=3D"">https://www.rfc-editor.org/retrieve/bulk</a><br class=3D""><br=
 class=3D"">Requests for special distribution should be addressed to =
either the<br class=3D"">author of the RFC in question, or to <a =
href=3D"mailto:rfc-editor@rfc-editor.org" =
class=3D"">rfc-editor@rfc-editor.org</a>. &nbsp;Unless<br =
class=3D"">specifically noted otherwise on the RFC itself, all RFCs are =
for<br class=3D"">unlimited distribution.<br class=3D""><br class=3D""><br=
 class=3D"">The RFC Editor Team<br class=3D"">Association Management =
Solutions, LLC<br class=3D""><br class=3D""><br =
class=3D"">_______________________________________________<br =
class=3D"">IETF-Announce mailing list<br class=3D""><a =
href=3D"mailto:IETF-Announce@ietf.org" =
class=3D"">IETF-Announce@ietf.org</a><br =
class=3D"">https://www.ietf.org/mailman/listinfo/ietf-announce<br =
class=3D""></div></div></blockquote></div><br =
class=3D""></div></body></html>=

--Apple-Mail=_AB10C52F-6964-4CDD-A555-E575BA521B6A--


From nobody Thu Feb  3 06:43:52 2022
Return-Path: <internet-drafts@ietf.org>
X-Original-To: tcpm@ietf.org
Delivered-To: tcpm@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id AA7813A1C0E; Thu,  3 Feb 2022 06:43:49 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
From: internet-drafts@ietf.org
To: <i-d-announce@ietf.org>
Cc: tcpm@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 7.44.0
Auto-Submitted: auto-generated
Precedence: bulk
Reply-To: tcpm@ietf.org
Message-ID: <164389942954.13556.7754569279919863072@ietfa.amsl.com>
Date: Thu, 03 Feb 2022 06:43:49 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/Br1GAeb_0c8sA1-Cb9xulugzNmY>
Subject: [tcpm] I-D Action: draft-ietf-tcpm-accurate-ecn-16.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 03 Feb 2022 14:43:50 -0000

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the TCP Maintenance and Minor Extensions WG of the IETF.

        Title           : More Accurate ECN Feedback in TCP
        Authors         : Bob Briscoe
                          Mirja Kühlewind
                          Richard Scheffenegger
	Filename        : draft-ietf-tcpm-accurate-ecn-16.txt
	Pages           : 60
	Date            : 2022-02-03

Abstract:
   Explicit Congestion Notification (ECN) is a mechanism where network
   nodes can mark IP packets instead of dropping them to indicate
   incipient congestion to the end-points.  Receivers with an ECN-
   capable transport protocol feed back this information to the sender.
   ECN was originally specified for TCP in such a way that only one
   feedback signal can be transmitted per Round-Trip Time (RTT).  Recent
   new TCP mechanisms like Congestion Exposure (ConEx), Data Center TCP
   (DCTCP) or Low Latency Low Loss Scalable Throughput (L4S) need more
   accurate ECN feedback information whenever more than one marking is
   received in one RTT.  This document specifies a scheme to provide
   more than one feedback signal per RTT in the TCP header.  Given TCP
   header space is scarce, it allocates a reserved header bit previously
   assigned to the ECN-Nonce.  It also overloads the two existing ECN
   flags in the TCP header.  The resulting extra space is exploited to
   feed back the IP-ECN field received during the 3-way handshake as
   well.  Supplementary feedback information can optionally be provided
   in a new TCP option, which is never used on the TCP SYN.  The
   document also specifies the treatment of this updated TCP wire
   protocol by middleboxes.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/

There is also an htmlized version available at:
https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16

A diff from the previous version is available at:
https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16


Internet-Drafts are also available by rsync at rsync.ietf.org::internet-drafts



From nobody Thu Feb  3 07:24:51 2022
Return-Path: <ietf@bobbriscoe.net>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 318083A08B7 for <tcpm@ietfa.amsl.com>; Thu,  3 Feb 2022 07:24:39 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.812
X-Spam-Level: 
X-Spam-Status: No, score=-2.812 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, NICE_REPLY_A=-0.714, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=bobbriscoe.net
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZFJuKBD0P9BX for <tcpm@ietfa.amsl.com>; Thu,  3 Feb 2022 07:24:34 -0800 (PST)
Received: from mail-ssdrsserver2.hostinginterface.eu (mail-ssdrsserver2.hostinginterface.eu [185.185.85.90]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 895753A088D for <tcpm@ietf.org>; Thu,  3 Feb 2022 07:24:30 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=bobbriscoe.net; s=default; h=Content-Transfer-Encoding:Content-Type: In-Reply-To:Cc:From:References:To:Subject:MIME-Version:Date:Message-ID:Sender :Reply-To:Content-ID:Content-Description:Resent-Date:Resent-From: Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id:List-Help: List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=rO8N+fHbGxpm1eR6qiQKDr6JesldMhy+p+1rr6kJ/BQ=; b=T49ZYcNfpOPmlrMmEElhraTjUI +InjGTdp634o2t0sseA7a3ZaMn0ph+eLO/8UJ8oDvitQX+HJMGF5nvI3wXxWNBaqECU6CVB4Tv3dZ hzkBN10BVtaeIMx8lFhMnQOWURRU/2YBis0MdFW9yb/4ozVgSe6DDf66UCKNFSMLHxtarDBhVzEi4 4m02yL64c/DxWnuqJIrOSpDxjQyR3DL2COWzhqfxI8l2M0KzPHd8Su+/P1SzwXOIPb+0PYFz0xsvn IQKysZKR/PQevdf0p5w/icPBMGRlJig91crTwCMd1nhUiFmVKAlFzK3o0IW9Wn+AFs7KwVZoOvVs9 6PNsrJYA==;
Received: from 67.153.238.178.in-addr.arpa ([178.238.153.67]:55082 helo=[192.168.1.11]) by ssdrsserver2.hostinginterface.eu with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 (Exim 4.94.2) (envelope-from <ietf@bobbriscoe.net>) id 1nFdyN-0002gg-1Y; Thu, 03 Feb 2022 15:24:28 +0000
Message-ID: <c2c5bdf9-6b34-fe19-3296-5f1cd831ed8e@bobbriscoe.net>
Date: Thu, 3 Feb 2022 15:24:26 +0000
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.5.0
Content-Language: en-GB
To: tcpm@ietf.org
References: <164389942954.13556.7754569279919863072@ietfa.amsl.com>
From: Bob Briscoe <ietf@bobbriscoe.net>
Cc: =?UTF-8?Q?Ilpo_J=c3=a4rvinen?= <ilpo.jarvinen@helsinki.fi>, Richard Scheffenegger <rscheff@gmx.at>, Mirja Kuehlewind <ietf@kuehlewind.net>
In-Reply-To: <164389942954.13556.7754569279919863072@ietfa.amsl.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - ssdrsserver2.hostinginterface.eu
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - bobbriscoe.net
X-Get-Message-Sender-Via: ssdrsserver2.hostinginterface.eu: authenticated_id: in@bobbriscoe.net
X-Authenticated-Sender: ssdrsserver2.hostinginterface.eu: in@bobbriscoe.net
X-Source: 
X-Source-Args: 
X-Source-Dir: 
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/xv6WlRSvCy04eKfR_sh40NuUP1w>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-accurate-ecn-16.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 03 Feb 2022 15:24:50 -0000

tcpm folks,

This rev to accurate-ecn is the first of two. The second will hopefully 
follow on its heels in the next couple of days.

Diffs between this first rev (-16) and -15:

1.  switches round two fairly large sections, so I've deferred other 
changes to a second rev so the diffs won't be masked by the switch round 
of sections.
Suggested by Ilpo to match the order in which the tests in these 
sections will be executed:
* Test for mangling the IP-ECN field (now 3.2.2.3),
* Then test for zeroing the ACE field (now 3.2.2.4).

2. Ilpo suggested some clarifications in "3.2.3.2.5.  Consistency 
between AccECN Feedback Fields", which is about the receiver of feedback 
ensuring consistency between the mandatory 3-bit ACE field and the 
optional 24-bit counters. In brief (paraphrasing) it previously only 
said "MUST consider both fields", when it is now clearer that it 
actually meant "MUST reconcile both fields", so that there is always a 
consistent baseline for subsequent ACKs.

3. A minor point is added in an appendix about the details that the 
pseudocode doesn't cover.



Bob

PS. #2 & #3 were added to the XML ages ago (Jul '21), so you will have 
seen them in the HTML. However, prob due to my clumsiness, the posted 
TXT didn't include them whereas the posted XML did (ironic for a section 
about consistency). In turn, inconsistency was only possible because I 
am having to manually post the TXT for this draft, due to an unresolved 
issue with v3 RFC XML tables.


On 03/02/2022 14:43, 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 TCP Maintenance and Minor Extensions WG of the IETF.
>
>          Title           : More Accurate ECN Feedback in TCP
>          Authors         : Bob Briscoe
>                            Mirja Kühlewind
>                            Richard Scheffenegger
> 	Filename        : draft-ietf-tcpm-accurate-ecn-16.txt
> 	Pages           : 60
> 	Date            : 2022-02-03
>
> Abstract:
>     Explicit Congestion Notification (ECN) is a mechanism where network
>     nodes can mark IP packets instead of dropping them to indicate
>     incipient congestion to the end-points.  Receivers with an ECN-
>     capable transport protocol feed back this information to the sender.
>     ECN was originally specified for TCP in such a way that only one
>     feedback signal can be transmitted per Round-Trip Time (RTT).  Recent
>     new TCP mechanisms like Congestion Exposure (ConEx), Data Center TCP
>     (DCTCP) or Low Latency Low Loss Scalable Throughput (L4S) need more
>     accurate ECN feedback information whenever more than one marking is
>     received in one RTT.  This document specifies a scheme to provide
>     more than one feedback signal per RTT in the TCP header.  Given TCP
>     header space is scarce, it allocates a reserved header bit previously
>     assigned to the ECN-Nonce.  It also overloads the two existing ECN
>     flags in the TCP header.  The resulting extra space is exploited to
>     feed back the IP-ECN field received during the 3-way handshake as
>     well.  Supplementary feedback information can optionally be provided
>     in a new TCP option, which is never used on the TCP SYN.  The
>     document also specifies the treatment of this updated TCP wire
>     protocol by middleboxes.
>
>
> The IETF datatracker status page for this draft is:
> https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/
>
> There is also an htmlized version available at:
> https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16
>
> A diff from the previous version is available at:
> https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16
>
>
> Internet-Drafts are also available by rsync at rsync.ietf.org::internet-drafts
>
>
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm

-- 
________________________________________________________________
Bob Briscoe                               http://bobbriscoe.net/


From nobody Thu Feb  3 16:15:50 2022
Return-Path: <internet-drafts@ietf.org>
X-Original-To: tcpm@ietf.org
Delivered-To: tcpm@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 5DB523A0EEE; Thu,  3 Feb 2022 16:15:48 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <i-d-announce@ietf.org>
Cc: tcpm@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 7.44.0
Auto-Submitted: auto-generated
Precedence: bulk
Reply-To: tcpm@ietf.org
Message-ID: <164393374830.18062.17460162529898546742@ietfa.amsl.com>
Date: Thu, 03 Feb 2022 16:15:48 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/ow3SmtRKs_9cxDsPBNUrhir-jQQ>
Subject: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-06.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 04 Feb 2022 00:15:49 -0000

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the TCP Maintenance and Minor Extensions WG of the IETF.

        Title           : A YANG Model for Transmission Control Protocol (TCP) Configuration
        Authors         : Michael Scharf
                          Mahesh Jethanandani
                          Vishal Murgai
	Filename        : draft-ietf-tcpm-yang-tcp-06.txt
	Pages           : 24
	Date            : 2022-02-03

Abstract:
   This document specifies a minimal YANG model for TCP on devices that
   are configured by network management protocols.  The YANG model
   defines a container for all TCP connections and groupings of
   authentication parameters that can be imported and used in TCP
   implementations or by other models that need to configure TCP
   parameters.  The model also includes basic TCP statistics.  The model
   is compliant with Network Management Datastore Architecture (NMDA)
   (RFC 8342).


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/

There is also an htmlized version available at:
https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-yang-tcp-06

A diff from the previous version is available at:
https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-yang-tcp-06


Internet-Drafts are also available by rsync at rsync.ietf.org::internet-drafts



From nobody Thu Feb  3 23:38:05 2022
Return-Path: <vidhi_goel@apple.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 63EF03A0AC2 for <tcpm@ietfa.amsl.com>; Thu,  3 Feb 2022 23:38:02 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.674
X-Spam-Level: 
X-Spam-Status: No, score=-2.674 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.576, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=apple.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GJezX2XMqiQv for <tcpm@ietfa.amsl.com>; Thu,  3 Feb 2022 23:37:56 -0800 (PST)
Received: from ma1-aaemail-dr-lapp01.apple.com (ma1-aaemail-dr-lapp01.apple.com [17.171.2.60]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A78E63A0AC1 for <tcpm@ietf.org>; Thu,  3 Feb 2022 23:37:56 -0800 (PST)
Received: from pps.filterd (ma1-aaemail-dr-lapp01.apple.com [127.0.0.1]) by ma1-aaemail-dr-lapp01.apple.com (8.16.0.42/8.16.0.42) with SMTP id 2147UqmY052359; Thu, 3 Feb 2022 23:37:44 -0800
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=apple.com; h=from : message-id : content-type : mime-version : subject : date : in-reply-to : cc : to : references; s=20180706; bh=Wujp9aQrqfgFq9RQ4BsClinlcZ8asonMov3fU10AnDg=; b=pVj66I6OvwtJiVmsw70wexk3RmLHnst7O4nkgIBi1L5fgjWKOSJy1Ke1EKE2CczCwRB7 Vh6SknmCV9YIY3cKsmBh5w0UCS/hlM8c8+wRYj0VLCFbEFZ0e//e6emn/hoxtRnnC4Vg EB8idOYr8x8ObAPERbXmwlsvfZdMezVvLPKclMfUvM9gaRuJNji7KL2jN/36ubfa1xf9 /5Sg0j1ETDoSIHKOn5Q3UCwOxJ5weD3UxNhnb0pWVlrCafK44Kpw27ctPkUNC/PY53Kf jdFheptK2hRcx3TNnVApZw2BXEiieH8oFk+6Xic8eitaaL6bQrogtB/vHJXfbO7tLNNn 2A== 
Received: from rn-mailsvcp-mta-lapp01.rno.apple.com (rn-mailsvcp-mta-lapp01.rno.apple.com [10.225.203.149]) by ma1-aaemail-dr-lapp01.apple.com with ESMTP id 3e0r284kk2-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NO); Thu, 03 Feb 2022 23:37:44 -0800
Received: from rn-mailsvcp-mmp-lapp04.rno.apple.com (rn-mailsvcp-mmp-lapp04.rno.apple.com [17.179.253.17]) by rn-mailsvcp-mta-lapp01.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPS id <0R6R00CWTRUV1Q30@rn-mailsvcp-mta-lapp01.rno.apple.com>;  Thu, 03 Feb 2022 23:37:43 -0800 (PST)
Received: from process_milters-daemon.rn-mailsvcp-mmp-lapp04.rno.apple.com by rn-mailsvcp-mmp-lapp04.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) id <0R6R00L00RTGWU00@rn-mailsvcp-mmp-lapp04.rno.apple.com>; Thu, 03 Feb 2022 23:37:43 -0800 (PST)
X-Va-A: 
X-Va-T-CD: 5b1df986fc1cf28f03a98c189e99a4df
X-Va-E-CD: d27b0bf1ec5e225c315eb75773194505
X-Va-R-CD: 7f6ba7ce084ea0aad2ff4f136dfa7fc9
X-Va-CD: 0
X-Va-ID: 06c807e5-bc5d-4a66-916c-e7c6d6dc49bf
X-V-A: 
X-V-T-CD: 5b1df986fc1cf28f03a98c189e99a4df
X-V-E-CD: d27b0bf1ec5e225c315eb75773194505
X-V-R-CD: 7f6ba7ce084ea0aad2ff4f136dfa7fc9
X-V-CD: 0
X-V-ID: ad8addac-e818-4c0b-8a4a-c52a7b9ae143
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-04_02:2022-02-03, 2022-02-04 signatures=0
Received: from smtpclient.apple (unknown [17.234.108.119]) by rn-mailsvcp-mmp-lapp04.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPSA id <0R6R00C6IRUU9Z00@rn-mailsvcp-mmp-lapp04.rno.apple.com>; Thu, 03 Feb 2022 23:37:43 -0800 (PST)
From: Vidhi Goel <vidhi_goel@apple.com>
Message-id: <2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com>
Content-type: multipart/signed; boundary="Apple-Mail=_BBEB9370-BBAF-4437-AFCD-7AAA518F2B27"; protocol="application/pkcs7-signature"; micalg=sha-256
MIME-version: 1.0 (Mac OS X Mail 14.0 \(3654.100.0.2.11\))
Date: Thu, 03 Feb 2022 23:37:42 -0800
In-reply-to: <c2c5bdf9-6b34-fe19-3296-5f1cd831ed8e@bobbriscoe.net>
Cc: tcpm@ietf.org, =?utf-8?Q?Ilpo_J=C3=A4rvinen?= <ilpo.jarvinen@helsinki.fi>,  Mirja Kuehlewind <ietf@kuehlewind.net>, Richard Scheffenegger <rscheff@gmx.at>
To: Bob Briscoe <ietf@bobbriscoe.net>
References: <164389942954.13556.7754569279919863072@ietfa.amsl.com> <c2c5bdf9-6b34-fe19-3296-5f1cd831ed8e@bobbriscoe.net>
X-Mailer: Apple Mail (2.3654.100.0.2.11)
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-04_02:2022-02-03, 2022-02-04 signatures=0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/sfD5jr82Q-nD5Jr-_sOxjNBgaw4>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-accurate-ecn-16.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 04 Feb 2022 07:38:03 -0000

--Apple-Mail=_BBEB9370-BBAF-4437-AFCD-7AAA518F2B27
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_98DD90BE-B568-4CC3-B475-76D897915AFB"


--Apple-Mail=_98DD90BE-B568-4CC3-B475-76D897915AFB
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

Hello Bob and authors,

I have read the draft a few times and will send my extensive review at a =
later point. But right now, I want to ask some critical things that an =
implementation could benefit from.

The draft says,
If a TCP client has set the SYN to Not-ECT, but
   receives feedback that the IP-ECN field on the SYN arrived with a
   different codepoint, it can detect such middlebox interference and
   send Not-ECT for the rest of the connection

1. On the forward path from client to server, the client will revert to =
Not-ECT when it sees for example, that it sent a Non-ECT SYN but =
received ACE encoding other than 0 1 0. What does the client do in the =
last ACK of the 3WHS -
 a. does it stay in AccECN mode and still send AccECN encoding based on =
the IP code point of SYN-ACK (Table 4)? This would mean that client =
won=E2=80=99t participate in ECN on the sender half of the connection =
and only provide AccECN feedback to the server as a receiver.
 b.  does it disable AccECN mode and set ACE=3D 0 0 0 so that a server =
in AccECN mode can disable ECN based on Table4?

While writing this, I realized that the intention is probably a. Could =
you confirm? Also, when the sender sets Not-ECT in its data packets, it =
should also disable acting upon any ACE feedback as we could still =
receive false ACE feedback from the server if the network, lets say, =
changed 00 (not ECT) to 01(ECT1). If you agree, we should add some text =
around this. Based on the current text, the sender will always respond =
to the ACE feedback even if it sends Not-ECT.

TBH, this is a complicated scenario, where sender said to the network - =
I don=E2=80=99t trust you so I can=E2=80=99t use ECN. Feel free to drop =
my packets. And the network mangles the IP to ECT1 and then set CE (when =
congested) which would be feedback=E2=80=99ed from the receiver. Now, =
the packet wasn=E2=80=99t dropped which it should have been. So, is it =
better to just ignore this feedback because sender doesn=E2=80=99t trust =
the network or just act on it and reduce cwnd in order to reduce =
congestion in the network somewhere.

2. On the reverse path from server to client, if a server sends a =
Not-ECT SYN-ACK and receives ACE handshake encoding on last ACK other =
than 0 1 0, there is no text like above that says server should send =
Not-ECT for the rest of the connection (or at least I didn=E2=80=99t =
find it). I think the server should also do same as client as the two =
paths could be different. One could make it more complicated by saying, =
if both client and server decide to not use ECN on their corresponding =
sender half, then ECN should be disabled but let=E2=80=99s talk about =
that later.

3. In general, not setting the ECT (ECT0 or ECT1) code point on an =
outgoing packet is different from supporting AccECN right as in the host =
can still provide AccECN feedback on the receive path.
=E2=80=A6.

To be continued if more questions come to my mind.


Thanks,
Vidhi

> On Feb 3, 2022, at 7:24 AM, Bob Briscoe <ietf@bobbriscoe.net> wrote:
>=20
> tcpm folks,
>=20
> This rev to accurate-ecn is the first of two. The second will =
hopefully follow on its heels in the next couple of days.
>=20
> Diffs between this first rev (-16) and -15:
>=20
> 1.  switches round two fairly large sections, so I've deferred other =
changes to a second rev so the diffs won't be masked by the switch round =
of sections.
> Suggested by Ilpo to match the order in which the tests in these =
sections will be executed:
> * Test for mangling the IP-ECN field (now 3.2.2.3),
> * Then test for zeroing the ACE field (now 3.2.2.4).
>=20
> 2. Ilpo suggested some clarifications in "3.2.3.2.5.  Consistency =
between AccECN Feedback Fields", which is about the receiver of feedback =
ensuring consistency between the mandatory 3-bit ACE field and the =
optional 24-bit counters. In brief (paraphrasing) it previously only =
said "MUST consider both fields", when it is now clearer that it =
actually meant "MUST reconcile both fields", so that there is always a =
consistent baseline for subsequent ACKs.
>=20
> 3. A minor point is added in an appendix about the details that the =
pseudocode doesn't cover.
>=20
>=20
>=20
> Bob
>=20
> PS. #2 & #3 were added to the XML ages ago (Jul '21), so you will have =
seen them in the HTML. However, prob due to my clumsiness, the posted =
TXT didn't include them whereas the posted XML did (ironic for a section =
about consistency). In turn, inconsistency was only possible because I =
am having to manually post the TXT for this draft, due to an unresolved =
issue with v3 RFC XML tables.
>=20
>=20
> On 03/02/2022 14:43, 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 TCP Maintenance and Minor Extensions =
WG of the IETF.
>>=20
>>         Title           : More Accurate ECN Feedback in TCP
>>         Authors         : Bob Briscoe
>>                           Mirja K=C3=BChlewind
>>                           Richard Scheffenegger
>> 	Filename        : draft-ietf-tcpm-accurate-ecn-16.txt
>> 	Pages           : 60
>> 	Date            : 2022-02-03
>>=20
>> Abstract:
>>    Explicit Congestion Notification (ECN) is a mechanism where =
network
>>    nodes can mark IP packets instead of dropping them to indicate
>>    incipient congestion to the end-points.  Receivers with an ECN-
>>    capable transport protocol feed back this information to the =
sender.
>>    ECN was originally specified for TCP in such a way that only one
>>    feedback signal can be transmitted per Round-Trip Time (RTT).  =
Recent
>>    new TCP mechanisms like Congestion Exposure (ConEx), Data Center =
TCP
>>    (DCTCP) or Low Latency Low Loss Scalable Throughput (L4S) need =
more
>>    accurate ECN feedback information whenever more than one marking =
is
>>    received in one RTT.  This document specifies a scheme to provide
>>    more than one feedback signal per RTT in the TCP header.  Given =
TCP
>>    header space is scarce, it allocates a reserved header bit =
previously
>>    assigned to the ECN-Nonce.  It also overloads the two existing ECN
>>    flags in the TCP header.  The resulting extra space is exploited =
to
>>    feed back the IP-ECN field received during the 3-way handshake as
>>    well.  Supplementary feedback information can optionally be =
provided
>>    in a new TCP option, which is never used on the TCP SYN.  The
>>    document also specifies the treatment of this updated TCP wire
>>    protocol by middleboxes.
>>=20
>>=20
>> The IETF datatracker status page for this draft is:
>> https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/
>>=20
>> There is also an htmlized version available at:
>> https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16
>>=20
>> A diff from the previous version is available at:
>> https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-accurate-ecn-16
>>=20
>>=20
>> Internet-Drafts are also available by rsync at =
rsync.ietf.org::internet-drafts
>>=20
>>=20
>> _______________________________________________
>> tcpm mailing list
>> tcpm@ietf.org
>> https://www.ietf.org/mailman/listinfo/tcpm
>=20
> --=20
> ________________________________________________________________
> Bob Briscoe                               http://bobbriscoe.net/
>=20
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm


--Apple-Mail=_98DD90BE-B568-4CC3-B475-76D897915AFB
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=utf-8

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3Dutf-8"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: after-white-space;" class=3D"">Hello=
 Bob and authors,<div class=3D""><br class=3D""></div><div class=3D"">I =
have read the draft a few times and will send my extensive review at a =
later point. But right now, I want to ask some critical things that an =
implementation could benefit from.</div><div class=3D""><br =
class=3D""></div><div class=3D"">The draft says,</div><div =
class=3D""><span style=3D"font-family: &quot;Helvetica Neue&quot;; =
font-size: 13px;" class=3D""><i class=3D"">If a TCP client has set the =
SYN to Not-ECT, but</i></span></div><div class=3D""><div style=3D"margin: =
0px; font-stretch: normal; font-size: 13px; line-height: normal; =
font-family: &quot;Helvetica Neue&quot;;" class=3D""><i =
class=3D"">&nbsp;&nbsp; receives feedback that the IP-ECN field on the =
SYN arrived with a</i></div><div style=3D"margin: 0px; font-stretch: =
normal; font-size: 13px; line-height: normal; font-family: =
&quot;Helvetica Neue&quot;;" class=3D""><i class=3D"">&nbsp;&nbsp; =
different codepoint, it can detect such middlebox interference =
and</i></div><div style=3D"margin: 0px; font-stretch: normal; font-size: =
13px; line-height: normal; font-family: &quot;Helvetica Neue&quot;;" =
class=3D""><i class=3D"">&nbsp;&nbsp; send Not-ECT for the rest of the =
connection</i></div><div><br class=3D""></div><div>1. On the forward =
path from client to server, the client will revert to Not-ECT when it =
sees for example, that it sent a Non-ECT SYN but received ACE encoding =
other than&nbsp;<font color=3D"#000000" size=3D"2" class=3D""><span =
style=3D"caret-color: rgb(0, 0, 0);" class=3D"">0   1   0.&nbsp;What =
does the client do in the last ACK of the 3WHS =
-</span></font></div><div><font color=3D"#000000" size=3D"2" =
class=3D"">&nbsp;a. does it stay in AccECN mode and still send AccECN =
encoding based on the IP&nbsp;code point of SYN-ACK (Table 4)? This =
would mean that client won=E2=80=99t participate in ECN on the sender =
half of the connection and only&nbsp;provide AccECN feedback to the =
server as a receiver.</font></div><div><font color=3D"#000000" size=3D"2" =
class=3D""><span style=3D"caret-color: rgb(0, 0, 0);" class=3D"">&nbsp;b. =
&nbsp;does it disable AccECN mode and set ACE=3D 0 0 0 so that a server =
in AccECN mode can disable ECN based on =
Table4?</span></font></div><div><font color=3D"#000000" size=3D"2" =
class=3D""><br class=3D""></font></div><div><font color=3D"#000000" =
size=3D"2" class=3D"">While writing this, I realized that the intention =
is&nbsp;probably a. Could you confirm? Also, when the sender sets =
Not-ECT in its data packets, it should also disable acting upon any ACE =
feedback as we could still receive false ACE feedback from the server =
if&nbsp;the network, lets say, changed 00 (not ECT) to 01(ECT1). If you =
agree, we should add some text around this. Based on the current text, =
the sender will always respond to the ACE feedback even if it sends =
Not-ECT.</font></div><div><font color=3D"#000000" size=3D"2" =
class=3D""><br class=3D""></font></div><div><font color=3D"#000000" =
size=3D"2" class=3D"">TBH, this is a complicated scenario, where sender =
said to the network - I don=E2=80=99t trust you so I can=E2=80=99t use =
ECN. Feel&nbsp;free to drop my packets. And the network mangles the IP =
to ECT1&nbsp;and then set CE (when congested)&nbsp;which would be =
feedback=E2=80=99ed from the receiver. Now, the packet wasn=E2=80=99t =
dropped which it should have been. So, is it better to just ignore this =
feedback because sender doesn=E2=80=99t trust the network or just act on =
it and reduce cwnd in order to reduce congestion in the network =
somewhere.</font></div><div><font color=3D"#000000" size=3D"2" =
class=3D""><br class=3D""></font></div><div><font color=3D"#000000" =
size=3D"2" class=3D"">2. On the reverse path&nbsp;from server to client, =
if a server sends a Not-ECT SYN-ACK and receives ACE handshake encoding =
on last ACK&nbsp;other than 0 1 0, there is no text like above that says =
server should send Not-ECT for the rest of the connection (or&nbsp;at =
least I&nbsp;didn=E2=80=99t find it). I think the server should also do =
same as client as&nbsp;the two paths could be different. One could make =
it more complicated by saying, if both client and server decide to not =
use ECN on their corresponding sender half, then ECN should be disabled =
but let=E2=80=99s talk about that later.</font></div><div><font =
color=3D"#000000" size=3D"2" class=3D""><br =
class=3D""></font></div><div><font color=3D"#000000" size=3D"2" =
class=3D"">3. In general, not setting the ECT (ECT0 or ECT1)&nbsp;code =
point on an outgoing packet is different from supporting AccECN right as =
in the host can still provide AccECN feedback on the&nbsp;receive =
path.</font></div><div>=E2=80=A6.</div><div><br class=3D""></div><div>To =
be continued if more questions come to my mind.</div><div><br =
class=3D""></div><div><br =
class=3D""></div><div>Thanks,</div><div>Vidhi</div><div><br =
class=3D""></div><div><blockquote type=3D"cite" class=3D""><div =
class=3D"">On Feb 3, 2022, at 7:24 AM, Bob Briscoe &lt;<a =
href=3D"mailto:ietf@bobbriscoe.net" class=3D"">ietf@bobbriscoe.net</a>&gt;=
 wrote:</div><br class=3D"Apple-interchange-newline"><div class=3D""><div =
class=3D"">tcpm folks,<br class=3D""><br class=3D"">This rev to =
accurate-ecn is the first of two. The second will hopefully follow on =
its heels in the next couple of days.<br class=3D""><br class=3D"">Diffs =
between this first rev (-16) and -15:<br class=3D""><br =
class=3D"">1.&nbsp; switches round two fairly large sections, so I've =
deferred other changes to a second rev so the diffs won't be masked by =
the switch round of sections.<br class=3D"">Suggested by Ilpo to match =
the order in which the tests in these sections will be executed:<br =
class=3D"">* Test for mangling the IP-ECN field (now 3.2.2.3),<br =
class=3D"">* Then test for zeroing the ACE field (now 3.2.2.4).<br =
class=3D""><br class=3D"">2. Ilpo suggested some clarifications in =
"3.2.3.2.5.&nbsp; Consistency between AccECN Feedback Fields", which is =
about the receiver of feedback ensuring consistency between the =
mandatory 3-bit ACE field and the optional 24-bit counters. In brief =
(paraphrasing) it previously only said "MUST consider both fields", when =
it is now clearer that it actually meant "MUST reconcile both fields", =
so that there is always a consistent baseline for subsequent ACKs.<br =
class=3D""><br class=3D"">3. A minor point is added in an appendix about =
the details that the pseudocode doesn't cover.<br class=3D""><br =
class=3D""><br class=3D""><br class=3D"">Bob<br class=3D""><br =
class=3D"">PS. #2 &amp; #3 were added to the XML ages ago (Jul '21), so =
you will have seen them in the HTML. However, prob due to my clumsiness, =
the posted TXT didn't include them whereas the posted XML did (ironic =
for a section about consistency). In turn, inconsistency was only =
possible because I am having to manually post the TXT for this draft, =
due to an unresolved issue with v3 RFC XML tables.<br class=3D""><br =
class=3D""><br class=3D"">On 03/02/2022 14:43, <a =
href=3D"mailto:internet-drafts@ietf.org" =
class=3D"">internet-drafts@ietf.org</a> wrote:<br class=3D""><blockquote =
type=3D"cite" class=3D"">A New Internet-Draft is available from the =
on-line Internet-Drafts directories.<br class=3D"">This draft is a work =
item of the TCP Maintenance and Minor Extensions WG of the IETF.<br =
class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Title =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: More =
Accurate ECN Feedback in TCP<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Authors =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: Bob Briscoe<br =
class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;Mirja K=C3=BChlewind<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;Richard Scheffenegger<br class=3D""><span class=3D"Apple-tab-span"=
 style=3D"white-space:pre">	</span>Filename =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: =
draft-ietf-tcpm-accurate-ecn-16.txt<br class=3D""><span =
class=3D"Apple-tab-span" style=3D"white-space:pre">	</span>Pages =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: 60<br =
class=3D""><span class=3D"Apple-tab-span" style=3D"white-space:pre">	=
</span>Date =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: =
2022-02-03<br class=3D""><br class=3D"">Abstract:<br class=3D""> =
&nbsp;&nbsp;&nbsp;Explicit Congestion Notification (ECN) is a mechanism =
where network<br class=3D""> &nbsp;&nbsp;&nbsp;nodes can mark IP packets =
instead of dropping them to indicate<br class=3D""> =
&nbsp;&nbsp;&nbsp;incipient congestion to the end-points. =
&nbsp;Receivers with an ECN-<br class=3D""> &nbsp;&nbsp;&nbsp;capable =
transport protocol feed back this information to the sender.<br =
class=3D""> &nbsp;&nbsp;&nbsp;ECN was originally specified for TCP in =
such a way that only one<br class=3D""> &nbsp;&nbsp;&nbsp;feedback =
signal can be transmitted per Round-Trip Time (RTT). &nbsp;Recent<br =
class=3D""> &nbsp;&nbsp;&nbsp;new TCP mechanisms like Congestion =
Exposure (ConEx), Data Center TCP<br class=3D""> =
&nbsp;&nbsp;&nbsp;(DCTCP) or Low Latency Low Loss Scalable Throughput =
(L4S) need more<br class=3D""> &nbsp;&nbsp;&nbsp;accurate ECN feedback =
information whenever more than one marking is<br class=3D""> =
&nbsp;&nbsp;&nbsp;received in one RTT. &nbsp;This document specifies a =
scheme to provide<br class=3D""> &nbsp;&nbsp;&nbsp;more than one =
feedback signal per RTT in the TCP header. &nbsp;Given TCP<br class=3D""> =
&nbsp;&nbsp;&nbsp;header space is scarce, it allocates a reserved header =
bit previously<br class=3D""> &nbsp;&nbsp;&nbsp;assigned to the =
ECN-Nonce. &nbsp;It also overloads the two existing ECN<br class=3D""> =
&nbsp;&nbsp;&nbsp;flags in the TCP header. &nbsp;The resulting extra =
space is exploited to<br class=3D""> &nbsp;&nbsp;&nbsp;feed back the =
IP-ECN field received during the 3-way handshake as<br class=3D""> =
&nbsp;&nbsp;&nbsp;well. &nbsp;Supplementary feedback information can =
optionally be provided<br class=3D""> &nbsp;&nbsp;&nbsp;in a new TCP =
option, which is never used on the TCP SYN. &nbsp;The<br class=3D""> =
&nbsp;&nbsp;&nbsp;document also specifies the treatment of this updated =
TCP wire<br class=3D""> &nbsp;&nbsp;&nbsp;protocol by middleboxes.<br =
class=3D""><br class=3D""><br class=3D"">The IETF datatracker status =
page for this draft is:<br class=3D""><a =
href=3D"https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/" =
class=3D"">https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/<=
/a><br class=3D""><br class=3D"">There is also an htmlized version =
available at:<br =
class=3D"">https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-=
ecn-16<br class=3D""><br class=3D"">A diff from the previous version is =
available at:<br =
class=3D"">https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-accurate-ec=
n-16<br class=3D""><br class=3D""><br class=3D"">Internet-Drafts are =
also available by rsync at rsync.ietf.org::internet-drafts<br =
class=3D""><br class=3D""><br =
class=3D"">_______________________________________________<br =
class=3D"">tcpm mailing list<br class=3D"">tcpm@ietf.org<br =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm<br =
class=3D""></blockquote><br class=3D"">-- <br =
class=3D"">_______________________________________________________________=
_<br class=3D"">Bob Briscoe =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;<a href=3D"http://bobbriscoe.net/" =
class=3D"">http://bobbriscoe.net/</a><br class=3D""><br =
class=3D"">_______________________________________________<br =
class=3D"">tcpm mailing list<br class=3D""><a =
href=3D"mailto:tcpm@ietf.org" class=3D"">tcpm@ietf.org</a><br =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm<br =
class=3D""></div></div></blockquote></div><br =
class=3D""></div></body></html>=

--Apple-Mail=_98DD90BE-B568-4CC3-B475-76D897915AFB--

--Apple-Mail=_BBEB9370-BBAF-4437-AFCD-7AAA518F2B27
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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=
--Apple-Mail=_BBEB9370-BBAF-4437-AFCD-7AAA518F2B27--


From nobody Fri Feb  4 05:44:15 2022
Return-Path: <ietf@bobbriscoe.net>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F084D3A11E3 for <tcpm@ietfa.amsl.com>; Fri,  4 Feb 2022 05:44:13 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.812
X-Spam-Level: 
X-Spam-Status: No, score=-2.812 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, NICE_REPLY_A=-0.714, RCVD_IN_DNSWL_BLOCKED=0.001, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=bobbriscoe.net
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4VD62eiUbUj4 for <tcpm@ietfa.amsl.com>; Fri,  4 Feb 2022 05:44:09 -0800 (PST)
Received: from mail-ssdrsserver2.hostinginterface.eu (mail-ssdrsserver2.hostinginterface.eu [185.185.85.90]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id F16C93A11E0 for <tcpm@ietf.org>; Fri,  4 Feb 2022 05:44:07 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=bobbriscoe.net; s=default; h=In-Reply-To:From:References:Cc:To:Subject: MIME-Version:Date:Message-ID:Content-Type:Sender:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id: List-Help:List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=wrozwDmPFCxI7KtiEt5DruKyqPIZ/y6vKzR4xWavjy4=; b=C+Caum3GcIq2rrfEYNCgwbtLZV 33E8Mr9DsTGyUmYKLSAoe7GVSkH0wrYWDLL0/tE3CuW11WbGrBQVVNel0JVODMK7V1OYihYbPXPD0 MHbIW4WDSC97UouYKOhZynpuEPWslZsx8SsuvBKWLf1jv/K2vEh5aJGmfZpU4c9sRY4PqP00ldfPr D7MHX/VPLbmjqMilu0tp0tAWM/aS7ufBH2nfkONPZyfwnCV3pVniIvLsyICZ8c6lqiPV6cML0+MA5 dlhEFU2H6HzyvzwzJpGxuHoA14ZeTnF5aTThSC0DRN/voT9aYs5hJ3Lj3AW8V+imZ2cDKoR4+F/3C bJ6+litg==;
Received: from 67.153.238.178.in-addr.arpa ([178.238.153.67]:55096 helo=[192.168.1.11]) by ssdrsserver2.hostinginterface.eu with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 (Exim 4.94.2) (envelope-from <ietf@bobbriscoe.net>) id 1nFysf-0007ea-Tp; Fri, 04 Feb 2022 13:44:05 +0000
Content-Type: multipart/alternative; boundary="------------ec4zBqo0WTp2fqwJE0meOgrp"
Message-ID: <e57355ae-5a9d-0a14-9478-29453d25622c@bobbriscoe.net>
Date: Fri, 4 Feb 2022 13:44:03 +0000
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.5.0
Content-Language: en-GB
To: Vidhi Goel <vidhi_goel@apple.com>
Cc: tcpm@ietf.org, Mirja Kuehlewind <ietf@kuehlewind.net>, Richard Scheffenegger <rscheff@gmx.at>
References: <164389942954.13556.7754569279919863072@ietfa.amsl.com> <c2c5bdf9-6b34-fe19-3296-5f1cd831ed8e@bobbriscoe.net> <2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com>
From: Bob Briscoe <ietf@bobbriscoe.net>
In-Reply-To: <2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com>
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - ssdrsserver2.hostinginterface.eu
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - bobbriscoe.net
X-Get-Message-Sender-Via: ssdrsserver2.hostinginterface.eu: authenticated_id: in@bobbriscoe.net
X-Authenticated-Sender: ssdrsserver2.hostinginterface.eu: in@bobbriscoe.net
X-Source: 
X-Source-Args: 
X-Source-Dir: 
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/y6v2yUJMQb0aBLcZBCnhnq0w_wI>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-accurate-ecn-16.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 04 Feb 2022 13:44:14 -0000

This is a multi-part message in MIME format.
--------------ec4zBqo0WTp2fqwJE0meOgrp
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Vidhi,

I've increased the table of contents to tocdepth = "4" which might help 
find the relevant sections better, because a lot of the meat in this 
draft is under sections 3.2.2 (ACE) and 3.2.3 (Option).

I've responded below, but I'd prefer it if you would say these things on 
the list, so we have a better justification for changing the draft. 
Reviewing a draft that has just be re-posted shouldn't imply anything 
about whether Apple is implementing the draft or not.

(BTW, regarding the proposed change to the initial value of r.e1b, I was 
going to send that  to the list once co-authors agreed to it. But it 
would be nice if some of the edits were not self-generated.)

See [BB] inline for responses (which I'll repeat if you send to the list).

On 04/02/2022 07:37, Vidhi Goel wrote:
> Hello Bob and authors,
>
> I have read the draft a few times and will send my extensive review at 
> a later point. But right now, I want to ask some critical things that 
> an implementation could benefit from.
>
> The draft says,
> /If a TCP client has set the SYN to Not-ECT, but/
> /   receives feedback that the IP-ECN field on the SYN arrived with a/
> /   different codepoint, it can detect such middlebox interference and/
> /   send Not-ECT for the rest of the connection/
>
> 1. On the forward path from client to server, the client will revert 
> to Not-ECT when it sees for example, that it sent a Non-ECT SYN but 
> received ACE encoding other than 0 1 0. What does the client do in the 
> last ACK of the 3WHS -
>  a. does it stay in AccECN mode and still send AccECN encoding based 
> on the IP code point of SYN-ACK (Table 4)? This would mean that client 
> won’t participate in ECN on the sender half of the connection and 
> only provide AccECN feedback to the server as a receiver.
>  b.  does it disable AccECN mode and set ACE= 0 0 0 so that a server 
> in AccECN mode can disable ECN based on Table4?
>
> While writing this, I realized that the intention is probably a. Could 
> you confirm? Also, when the sender sets Not-ECT in its data packets, 
> it should also disable acting upon any ACE feedback as we could still 
> receive false ACE feedback from the server if the network, lets say, 
> changed 00 (not ECT) to 01(ECT1). If you agree, we should add some 
> text around this. Based on the current text, the sender will always 
> respond to the ACE feedback even if it sends Not-ECT.
>
> TBH, this is a complicated scenario, where sender said to the network 
> - I don’t trust you so I can’t use ECN. Feel free to drop my packets. 
> And the network mangles the IP to ECT1 and then set CE (when 
> congested) which would be feedback’ed from the receiver. Now, the 
> packet wasn’t dropped which it should have been. So, is it better to 
> just ignore this feedback because sender doesn’t trust the network or 
> just act on it and reduce cwnd in order to reduce congestion in the 
> network somewhere.

[BB] Your points are all good ones.
I'll address this last complex scenario first, because it has 
implications for the earlier questions. First I'll define some terminology:

Simple mangling scenario:

  * some network function, e.g. broken Diffserv ToS-byte remarking,
    illegally remarks a Not-ECT  SYN to CE.

Your complex mangling scenario, repeated here:

  * some network function, e.g. broken Diffserv ToS-byte remarking,
    illegally remarks a Not-ECT SYN to ECT0 or ECT1,
  * then congestion at a subsequent bottleneck is marking some packets CE

Then potentially the client could tailor its behaviour after sending a 
Not-ECT SYN, If the SYN/ACK feedback is

 1. ECT: disable sending ECT, but continue responding to ECN feedback
 2. CE, disable sending ECT, and disable response to ECN feedback

However, this is uncertain, because CE feedback on the SYN/ACK could 
indicate either the simple or the complex mangling scenario.

A simpler alternative would be to always continue responding to ECN 
feedback. Rationale:

  * Whether case #1 or #2, assume that mangling of the SYN might have
    been Not-ECT to ECT, even if the feedback off the SYN is CE.
  * Then as the connection progresses, if /all/ feedback is CE, there's
    already a recommendation to fall-back to disabling congestion response.

If we do this, I think we ought to say "SHOULD continue to respond to 
ECN feedback", not "MUST".

And we'll need to put this all to the WG.

1.a) Now back to the beginning of your point. The text you quote is from 
§2.5 which is in the non-normative "Overview and Rationale" section (§2).
You really need the normative text from:


          3.2.2.3.
          <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3>Testing
          for Mangling of the IP/ECN Field
          <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the>


where it says (end of 1st para):

    " ...for the remainder of the connection the client MUST NOT send
    ECN-capable packets, but it MUST continue to feed back any ECN
    markings on arriving packets."

So, you're right, the answer is (a).
It can't be (b), because, once in AccECN mode, 000 on the ACE field just 
becomes a counter value, and no longer negotiates the feedback mode.

1.b) As per the above, let's conservatively assume complex mangling. So, 
at the end of the first para quoted above I suggest we add:

    " ...for the remainder of the connection the client MUST NOT send
    ECN-capable packets *but**it**SHOULD continue to respond to ECN
    feedback even though **it **is no longer sending ECN-capable packets
    (see reasoning below)**. T**he client**MUST remain in the same
    feedback mode and***it MUST continue to feed back any ECN markings
    on arriving packets."

I'll do the same for next para about the server.
I'll work out some text for the reasoning, but I won't give it here.

There are two other cases where it disables sending ECT, which don't say 
whether it continues to respond to congestion:

  * "Testing for Mangling" section, penultimate para,  where it's
    receiving solid CE:
    "Once a Data Sender has entered AccECN mode it SHOULD check whether
    all feedback received for the first three or four rounds indicated
    that every packet it sent was CE-marked. If so, for the remainder of
    the connection, the Data Sender SHOULD NOT send ECN-capable
    packets*and it SHOULD NOT respond to ECN feedback*, but *it MUST
    remain in the same feedback mode and *it MUST continue to feed back
    any ECN markings on arriving packets*(in its role as Data Receiver)*."
  * Next section "Zeroing of the ACE Field", 2nd para:
    "If the value of this ACE field is zero (0b000), the Data Sender
    disables sending ECN-capable packets for the remainder of the
    half-connection by setting the IP/ECN field in all subsequent
    packets to Not-ECT, *but**it**SHOULD continue to respond to ECN
    feedback even though **it **is no longer sending ECN-capable
    packets.* *It MUST also remain in the same feedback mode and it MUST
    continue to feed back any ECN markings on arriving packets (in its
    role as Data Receiver).*"

I'll also add reasoning (in the zeroing section, it already says that 
ACE=0b000 is not an unambiguous indication of mangling).

And this prompts me to edit the bullet in an earlier section about the 
obligation of a sender to respond to congestion feedback:


        3.1.5.
        <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.1.5>Implications
        of AccECN Mode
        <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-implications-of-accecn-mode>


    "It is still obliged to respond appropriately to AccECN feedback
    that indicates there were ECN marks on *ECN-capable *packets it had
    previously sent."

And I'll add a bullet:

    "*If the sender chooses not to send ECN-capable packets (e.g.
    because path traversal of the ECN field is suspect), it can ignore
    any **ECN **feedback about those packets if it is certain that it
    cannot be valid (see Section 3.2.2, which gives normative
    requirements for certain specific cases).*"

I've said 'can' rather than 'MUST' because it's hard to cover all cases, 
e.g. single packets without ECT when it's not clear whether the feedback 
covered another packet that was ECT.

How does all this sound?

> 2. On the reverse path from server to client, if a server sends a 
> Not-ECT SYN-ACK and receives ACE handshake encoding on last ACK other 
> than 0 1 0, there is no text like above that says server should send 
> Not-ECT for the rest of the connection (or at least I didn’t find it). 
> I think the server should also do same as client as the two paths 
> could be different. One could make it more complicated by saying, if 
> both client and server decide to not use ECN on their corresponding 
> sender half, then ECN should be disabled but let’s talk about that later.

[BB] It's in the second para of:


          3.2.2.3.
          <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3>Testing
          for Mangling of the IP/ECN Field
          <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the>


> 3. In general, not setting the ECT (ECT0 or ECT1) code point on an 
> outgoing packet is different from supporting AccECN right as in the 
> host can still provide AccECN feedback on the receive path.

[BB] Yup.

Do you think any further explanation is needed in the above sections?


Bob

> ….
>
> To be continued if more questions come to my mind.
>
>
> Thanks,
> Vidhi
>
>> On Feb 3, 2022, at 7:24 AM, Bob Briscoe <ietf@bobbriscoe.net> wrote:
>>
>> tcpm folks,
>>
>> This rev to accurate-ecn is the first of two. The second will 
>> hopefully follow on its heels in the next couple of days.
>>
>> Diffs between this first rev (-16) and -15:
>>
>> 1.  switches round two fairly large sections, so I've deferred other 
>> changes to a second rev so the diffs won't be masked by the switch 
>> round of sections.
>> Suggested by Ilpo to match the order in which the tests in these 
>> sections will be executed:
>> * Test for mangling the IP-ECN field (now 3.2.2.3),
>> * Then test for zeroing the ACE field (now 3.2.2.4).
>>
>> 2. Ilpo suggested some clarifications in "3.2.3.2.5. Consistency 
>> between AccECN Feedback Fields", which is about the receiver of 
>> feedback ensuring consistency between the mandatory 3-bit ACE field 
>> and the optional 24-bit counters. In brief (paraphrasing) it 
>> previously only said "MUST consider both fields", when it is now 
>> clearer that it actually meant "MUST reconcile both fields", so that 
>> there is always a consistent baseline for subsequent ACKs.
>>
>> 3. A minor point is added in an appendix about the details that the 
>> pseudocode doesn't cover.
>>
>>
>>
>> Bob
>>
>> PS. #2 & #3 were added to the XML ages ago (Jul '21), so you will 
>> have seen them in the HTML. However, prob due to my clumsiness, the 
>> posted TXT didn't include them whereas the posted XML did (ironic for 
>> a section about consistency). In turn, inconsistency was only 
>> possible because I am having to manually post the TXT for this draft, 
>> due to an unresolved issue with v3 RFC XML tables.
>>
>>
>> On 03/02/2022 14:43, 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 TCP Maintenance and Minor 
>>> Extensions WG of the IETF.
>>>
>>>         Title           : More Accurate ECN Feedback in TCP
>>>         Authors         : Bob Briscoe
>>>                           Mirja Kühlewind
>>>                           Richard Scheffenegger
>>> Filename        : draft-ietf-tcpm-accurate-ecn-16.txt
>>> Pages           : 60
>>> Date            : 2022-02-03
>>>
>>> Abstract:
>>>    Explicit Congestion Notification (ECN) is a mechanism where network
>>>    nodes can mark IP packets instead of dropping them to indicate
>>>    incipient congestion to the end-points.  Receivers with an ECN-
>>>    capable transport protocol feed back this information to the sender.
>>>    ECN was originally specified for TCP in such a way that only one
>>>    feedback signal can be transmitted per Round-Trip Time (RTT).  Recent
>>>    new TCP mechanisms like Congestion Exposure (ConEx), Data Center TCP
>>>    (DCTCP) or Low Latency Low Loss Scalable Throughput (L4S) need more
>>>    accurate ECN feedback information whenever more than one marking is
>>>    received in one RTT.  This document specifies a scheme to provide
>>>    more than one feedback signal per RTT in the TCP header.  Given TCP
>>>    header space is scarce, it allocates a reserved header bit previously
>>>    assigned to the ECN-Nonce.  It also overloads the two existing ECN
>>>    flags in the TCP header.  The resulting extra space is exploited to
>>>    feed back the IP-ECN field received during the 3-way handshake as
>>>    well.  Supplementary feedback information can optionally be provided
>>>    in a new TCP option, which is never used on the TCP SYN.  The
>>>    document also specifies the treatment of this updated TCP wire
>>>    protocol by middleboxes.
>>>
>>>
>>> The IETF datatracker status page for this draft is:
>>> https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/
>>>
>>> There is also an htmlized version available at:
>>> https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16
>>>
>>> A diff from the previous version is available at:
>>> https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16
>>>
>>>
>>> Internet-Drafts are also available by rsync at 
>>> rsync.ietf.org::internet-drafts
>>>
>>>
>>> _______________________________________________
>>> tcpm mailing list
>>> tcpm@ietf.org
>>> https://www.ietf.org/mailman/listinfo/tcpm
>>
>> -- 
>> ________________________________________________________________
>> Bob Briscoe http://bobbriscoe.net/
>>
>> _______________________________________________
>> tcpm mailing list
>> tcpm@ietf.org
>> https://www.ietf.org/mailman/listinfo/tcpm
>

-- 
________________________________________________________________
Bob Briscoehttp://bobbriscoe.net/

--------------ec4zBqo0WTp2fqwJE0meOgrp
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    Vidhi,<br>
    <br>
    I've increased the table of contents to tocdepth = "4" which might
    help find the relevant sections better, because a lot of the meat in
    this draft is under sections 3.2.2 (ACE) and 3.2.3 (Option).<br>
    <br>
    I've responded below, but I'd prefer it if you would say these
    things on the list, so we have a better justification for changing
    the draft. Reviewing a draft that has just be re-posted shouldn't
    imply anything about whether Apple is implementing the draft or not.<br>
    <br>
    (BTW, regarding the proposed change to the initial value of r.e1b, I
    was going to send that  to the list once co-authors agreed to it.
    But it would be nice if some of the edits were not self-generated.)<br>
    <br>
    See [BB] inline for responses (which I'll repeat if you send to the
    list).<br>
    <br>
    <div class="moz-cite-prefix">On 04/02/2022 07:37, Vidhi Goel wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com">
      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
      Hello Bob and authors,
      <div class=""><br class="">
      </div>
      <div class="">I have read the draft a few times and will send my
        extensive review at a later point. But right now, I want to ask
        some critical things that an implementation could benefit from.</div>
      <div class=""><br class="">
      </div>
      <div class="">The draft says,</div>
      <div class=""><span style="font-family: &quot;Helvetica
          Neue&quot;; font-size: 13px;" class=""><i class="">If a TCP
            client has set the SYN to Not-ECT, but</i></span></div>
      <div class="">
        <div style="margin: 0px; font-stretch: normal; font-size: 13px;
          line-height: normal; font-family: &quot;Helvetica Neue&quot;;"
          class=""><i class="">   receives feedback that the IP-ECN
            field on the SYN arrived with a</i></div>
        <div style="margin: 0px; font-stretch: normal; font-size: 13px;
          line-height: normal; font-family: &quot;Helvetica Neue&quot;;"
          class=""><i class="">   different codepoint, it can detect
            such middlebox interference and</i></div>
        <div style="margin: 0px; font-stretch: normal; font-size: 13px;
          line-height: normal; font-family: &quot;Helvetica Neue&quot;;"
          class=""><i class="">   send Not-ECT for the rest of the
            connection</i></div>
        <div><br class="">
        </div>
        <div>1. On the forward path from client to server, the client
          will revert to Not-ECT when it sees for example, that it sent
          a Non-ECT SYN but received ACE encoding other than <font
            class="" size="2" color="#000000"><span style="caret-color:
              rgb(0, 0, 0);" class="">0 1 0. What does the client do in
              the last ACK of the 3WHS -</span></font></div>
        <div><font class="" size="2" color="#000000"> a. does it stay in
            AccECN mode and still send AccECN encoding based on the
            IP code point of SYN-ACK (Table 4)? This would mean that
            client won’t participate in ECN on the sender half of the
            connection and only provide AccECN feedback to the server as
            a receiver.</font></div>
        <div><font class="" size="2" color="#000000"><span
              style="caret-color: rgb(0, 0, 0);" class=""> b.  does it
              disable AccECN mode and set ACE= 0 0 0 so that a server in
              AccECN mode can disable ECN based on Table4?</span></font></div>
        <div><font class="" size="2" color="#000000"><br class="">
          </font></div>
        <div><font class="" size="2" color="#000000">While writing this,
            I realized that the intention is probably a. Could you
            confirm? Also, when the sender sets Not-ECT in its data
            packets, it should also disable acting upon any ACE feedback
            as we could still receive false ACE feedback from the server
            if the network, lets say, changed 00 (not ECT) to 01(ECT1).
            If you agree, we should add some text around this. Based on
            the current text, the sender will always respond to the ACE
            feedback even if it sends Not-ECT.</font></div>
        <div><font class="" size="2" color="#000000"><br class="">
          </font></div>
        <div><font class="" size="2" color="#000000">TBH, this is a
            complicated scenario, where sender said to the network - I
            don’t trust you so I can’t use ECN. Feel free to drop my
            packets. And the network mangles the IP to ECT1 and then set
            CE (when congested) which would be feedback’ed from the
            receiver. Now, the packet wasn’t dropped which it should
            have been. So, is it better to just ignore this feedback
            because sender doesn’t trust the network or just act on it
            and reduce cwnd in order to reduce congestion in the network
            somewhere.</font></div>
      </div>
    </blockquote>
    <br>
    [BB] Your points are all good ones.<br>
    I'll address this last complex scenario first, because it has
    implications for the earlier questions. First I'll define some
    terminology:<br>
    <br>
    Simple mangling scenario: <br>
    <ul>
      <li>some network function, e.g. broken Diffserv ToS-byte
        remarking, illegally remarks a Not-ECT  SYN
        to CE. </li>
    </ul>
    Your complex mangling scenario, repeated here:<br>
    <ul>
      <li>some network function, e.g. broken Diffserv ToS-byte
        remarking, illegally remarks a Not-ECT SYN to ECT0 or ECT1, <br>
      </li>
      <li>then congestion at a subsequent bottleneck is marking some
        packets CE<br>
      </li>
    </ul>
    Then potentially the client could tailor its behaviour after sending
    a Not-ECT SYN, If the SYN/ACK feedback is<br>
    <ol>
      <li>ECT: disable sending ECT, but continue responding to ECN
        feedback</li>
      <li>CE, disable sending ECT, and disable response to ECN feedback</li>
    </ol>
    However, this is uncertain, because CE feedback on the SYN/ACK could
    indicate either the simple or the complex mangling scenario.<br>
    <br>
    A simpler alternative would be to always continue responding to ECN
    feedback. Rationale: <br>
    <ul>
      <li>Whether case #1 or #2, assume that mangling of the SYN might
        have been Not-ECT to ECT, even if the feedback off the SYN is
        CE. </li>
      <li>Then as the connection progresses, if /all/ feedback is CE,
        there's already a recommendation to fall-back to disabling
        congestion response.</li>
    </ul>
    If we do this, I think we ought to say "SHOULD continue to respond
    to ECN feedback", not "MUST".<br>
    <br>
    And we'll need to put this all to the WG. <br>
    <br>
    1.a) Now back to the beginning of your point. The text you quote is
    from §2.5 which is in the non-normative "Overview and Rationale"
    section (§2). <br>
    You really need the normative text from: <br>
    <h5 id="name-testing-for-mangling-of-the">
      <a
href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3"
        class="section-number selfRef">3.2.2.3. </a><a
href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the"
        class="section-name selfRef">Testing for Mangling of the IP/ECN
        Field</a> </h5>
    where it says (end of 1st para): <br>
    <blockquote>" ...for the remainder of the connection the client MUST
      NOT send ECN-capable packets, but it MUST continue to feed back
      any ECN markings on arriving packets."<br>
    </blockquote>
    So, you're right, the answer is (a).<br>
    It can't be (b), because, once in AccECN mode, 000 on the ACE field
    just becomes a counter value, and no longer negotiates the feedback
    mode.<br>
    <br>
    1.b) As per the above, let's conservatively assume complex mangling.
    So, at the end of the first para quoted above I suggest we add:<br>
    <blockquote>" ...for the remainder of the connection the client MUST
      NOT send ECN-capable packets <b>but</b><b> it</b><b> SHOULD
        continue to respond to ECN feedback even though </b><b>it </b><b>is
        no longer sending ECN-capable packets (see reasoning below)</b><b>.
        T</b><b>he client</b><b> MUST remain in the same feedback mode
        and</b><b> </b>it MUST continue to feed back any ECN markings
      on arriving packets."<br>
    </blockquote>
    I'll do the same for next para about the server. <br>
    I'll work out some text for the reasoning, but I won't give it here.<br>
    <br>
    There are two other cases where it disables sending ECT, which don't
    say whether it continues to respond to congestion:<br>
    <ul>
      <li>"Testing for Mangling" section, penultimate para,  where it's
        receiving solid CE:<br>
        "Once a Data Sender has entered AccECN mode it SHOULD check
        whether all feedback received for the first three or four rounds
        indicated that every packet it sent was CE-marked. If so, for
        the remainder of the connection, the Data Sender SHOULD NOT send
        ECN-capable packets<b> and it SHOULD NOT respond to ECN feedback</b>,
        but <b>it MUST remain in the same feedback mode and </b>it
        MUST continue to feed back any ECN markings on arriving packets<b>
          (in its role as Data Receiver)</b>."</li>
      <li>Next section "Zeroing of the ACE Field", 2nd para:<br>
        "If the value of this ACE field is zero (0b000), the Data Sender
        disables sending ECN-capable packets for the remainder of the
        half-connection by setting the IP/ECN field in all subsequent
        packets to Not-ECT, <b>but</b><b> it</b><b> SHOULD continue to
          respond to ECN feedback even though </b><b>it </b><b>is no
          longer sending ECN-capable packets.</b> <b>It MUST also
          remain in the same feedback mode and it MUST continue to feed
          back any ECN markings on arriving packets (in its role as Data
          Receiver).</b>"<br>
      </li>
    </ul>
    I'll also add reasoning (in the zeroing section, it already says
    that ACE=0b000 is not an unambiguous indication of mangling).<br>
    <br>
    And this prompts me to edit the bullet in an earlier section about
    the obligation of a sender to respond to congestion feedback:<br>
    <h4 id="name-implications-of-accecn-mode">
      <a
href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.1.5"
        class="section-number selfRef">3.1.5. </a><a
href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-implications-of-accecn-mode"
        class="section-name selfRef">Implications of AccECN Mode</a> </h4>
    <blockquote>"It is still obliged to respond appropriately to AccECN
      feedback that indicates there were ECN marks on <b>ECN-capable </b>packets
      it had previously sent."<br>
    </blockquote>
    And I'll add a bullet:<br>
    <blockquote>"<b>If the sender chooses not to send ECN-capable
        packets (e.g. because path traversal of the ECN field is
        suspect), it can ignore any </b><b>ECN </b><b>feedback about
        those packets if it is certain that it cannot be valid (see
        Section 3.2.2, which gives normative requirements for certain
        specific cases).</b>"<br>
    </blockquote>
    I've said 'can' rather than 'MUST' because it's hard to cover all
    cases, e.g. single packets without ECT when it's not clear whether
    the feedback covered another packet that was ECT. <br>
    <br>
    How does all this sound?<br>
    <br>
    <blockquote type="cite"
      cite="mid:2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com">
      <div class="">
        <div><font class="" size="2" color="#000000">2. On the reverse
            path from server to client, if a server sends a Not-ECT
            SYN-ACK and receives ACE handshake encoding on last
            ACK other than 0 1 0, there is no text like above that says
            server should send Not-ECT for the rest of the connection
            (or at least I didn’t find it). I think the server should
            also do same as client as the two paths could be different.
            One could make it more complicated by saying, if both client
            and server decide to not use ECN on their corresponding
            sender half, then ECN should be disabled but let’s talk
            about that later.</font></div>
      </div>
    </blockquote>
    <br>
    [BB] It's in the second para of:<br>
    <h5 id="name-testing-for-mangling-of-the">
      <a
href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3"
        class="section-number selfRef">3.2.2.3. </a><a
href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the"
        class="section-name selfRef">Testing for Mangling of the IP/ECN
        Field</a> </h5>
    <blockquote type="cite"
      cite="mid:2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com">
      <div class="">
        <div><font class="" size="2" color="#000000">3. In general, not
            setting the ECT (ECT0 or ECT1) code point on an outgoing
            packet is different from supporting AccECN right as in the
            host can still provide AccECN feedback on the receive path.</font></div>
      </div>
    </blockquote>
    <br>
    [BB] Yup.<br>
    <br>
    Do you think any further explanation is needed in the above
    sections?<br>
    <br>
    <br>
    Bob<br>
    <br>
    <blockquote type="cite"
      cite="mid:2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com">
      <div class="">
        <div>….</div>
        <div><br class="">
        </div>
        <div>To be continued if more questions come to my mind.</div>
        <div><br class="">
        </div>
        <div><br class="">
        </div>
        <div>Thanks,</div>
        <div>Vidhi</div>
        <div><br class="">
        </div>
        <div>
          <blockquote type="cite" class="">
            <div class="">On Feb 3, 2022, at 7:24 AM, Bob Briscoe &lt;<a
                href="mailto:ietf@bobbriscoe.net"
                class="moz-txt-link-freetext" moz-do-not-send="true">ietf@bobbriscoe.net</a>&gt;
              wrote:</div>
            <br class="Apple-interchange-newline">
            <div class="">
              <div class="">tcpm folks,<br class="">
                <br class="">
                This rev to accurate-ecn is the first of two. The second
                will hopefully follow on its heels in the next couple of
                days.<br class="">
                <br class="">
                Diffs between this first rev (-16) and -15:<br class="">
                <br class="">
                1.  switches round two fairly large sections, so I've
                deferred other changes to a second rev so the diffs
                won't be masked by the switch round of sections.<br
                  class="">
                Suggested by Ilpo to match the order in which the tests
                in these sections will be executed:<br class="">
                * Test for mangling the IP-ECN field (now 3.2.2.3),<br
                  class="">
                * Then test for zeroing the ACE field (now 3.2.2.4).<br
                  class="">
                <br class="">
                2. Ilpo suggested some clarifications in "3.2.3.2.5. 
                Consistency between AccECN Feedback Fields", which is
                about the receiver of feedback ensuring consistency
                between the mandatory 3-bit ACE field and the optional
                24-bit counters. In brief (paraphrasing) it previously
                only said "MUST consider both fields", when it is now
                clearer that it actually meant "MUST reconcile both
                fields", so that there is always a consistent baseline
                for subsequent ACKs.<br class="">
                <br class="">
                3. A minor point is added in an appendix about the
                details that the pseudocode doesn't cover.<br class="">
                <br class="">
                <br class="">
                <br class="">
                Bob<br class="">
                <br class="">
                PS. #2 &amp; #3 were added to the XML ages ago (Jul
                '21), so you will have seen them in the HTML. However,
                prob due to my clumsiness, the posted TXT didn't include
                them whereas the posted XML did (ironic for a section
                about consistency). In turn, inconsistency was only
                possible because I am having to manually post the TXT
                for this draft, due to an unresolved issue with v3 RFC
                XML tables.<br class="">
                <br class="">
                <br class="">
                On 03/02/2022 14:43, <a
                  href="mailto:internet-drafts@ietf.org"
                  class="moz-txt-link-freetext" moz-do-not-send="true">internet-drafts@ietf.org</a>
                wrote:<br class="">
                <blockquote type="cite" class="">A New Internet-Draft is
                  available from the on-line Internet-Drafts
                  directories.<br class="">
                  This draft is a work item of the TCP Maintenance and
                  Minor Extensions WG of the IETF.<br class="">
                  <br class="">
                          Title           : More Accurate ECN Feedback
                  in TCP<br class="">
                          Authors         : Bob Briscoe<br class="">
                                            Mirja Kühlewind<br class="">
                                            Richard Scheffenegger<br
                    class="">
                  <span class="Apple-tab-span" style="white-space:pre">	</span>Filename
                         : draft-ietf-tcpm-accurate-ecn-16.txt<br
                    class="">
                  <span class="Apple-tab-span" style="white-space:pre">	</span>Pages
                            : 60<br class="">
                  <span class="Apple-tab-span" style="white-space:pre">	</span>Date
                             : 2022-02-03<br class="">
                  <br class="">
                  Abstract:<br class="">
                     Explicit Congestion Notification (ECN) is a
                  mechanism where network<br class="">
                     nodes can mark IP packets instead of dropping them
                  to indicate<br class="">
                     incipient congestion to the end-points.  Receivers
                  with an ECN-<br class="">
                     capable transport protocol feed back this
                  information to the sender.<br class="">
                     ECN was originally specified for TCP in such a way
                  that only one<br class="">
                     feedback signal can be transmitted per Round-Trip
                  Time (RTT).  Recent<br class="">
                     new TCP mechanisms like Congestion Exposure
                  (ConEx), Data Center TCP<br class="">
                     (DCTCP) or Low Latency Low Loss Scalable Throughput
                  (L4S) need more<br class="">
                     accurate ECN feedback information whenever more
                  than one marking is<br class="">
                     received in one RTT.  This document specifies a
                  scheme to provide<br class="">
                     more than one feedback signal per RTT in the TCP
                  header.  Given TCP<br class="">
                     header space is scarce, it allocates a reserved
                  header bit previously<br class="">
                     assigned to the ECN-Nonce.  It also overloads the
                  two existing ECN<br class="">
                     flags in the TCP header.  The resulting extra space
                  is exploited to<br class="">
                     feed back the IP-ECN field received during the
                  3-way handshake as<br class="">
                     well.  Supplementary feedback information can
                  optionally be provided<br class="">
                     in a new TCP option, which is never used on the TCP
                  SYN.  The<br class="">
                     document also specifies the treatment of this
                  updated TCP wire<br class="">
                     protocol by middleboxes.<br class="">
                  <br class="">
                  <br class="">
                  The IETF datatracker status page for this draft is:<br
                    class="">
                  <a
                    href="https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/"
                    class="moz-txt-link-freetext" moz-do-not-send="true">https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/</a><br
                    class="">
                  <br class="">
                  There is also an htmlized version available at:<br
                    class="">
<a class="moz-txt-link-freetext" href="https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16">https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16</a><br
                    class="">
                  <br class="">
                  A diff from the previous version is available at:<br
                    class="">
<a class="moz-txt-link-freetext" href="https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16">https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16</a><br
                    class="">
                  <br class="">
                  <br class="">
                  Internet-Drafts are also available by rsync at
                  rsync.ietf.org::internet-drafts<br class="">
                  <br class="">
                  <br class="">
                  _______________________________________________<br
                    class="">
                  tcpm mailing list<br class="">
                  <a class="moz-txt-link-abbreviated" href="mailto:tcpm@ietf.org">tcpm@ietf.org</a><br class="">
                  <a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/tcpm">https://www.ietf.org/mailman/listinfo/tcpm</a><br class="">
                </blockquote>
                <br class="">
                -- <br class="">
________________________________________________________________<br
                  class="">
                Bob Briscoe                               <a
                  href="http://bobbriscoe.net/"
                  class="moz-txt-link-freetext" moz-do-not-send="true">http://bobbriscoe.net/</a><br
                  class="">
                <br class="">
                _______________________________________________<br
                  class="">
                tcpm mailing list<br class="">
                <a href="mailto:tcpm@ietf.org"
                  class="moz-txt-link-freetext" moz-do-not-send="true">tcpm@ietf.org</a><br
                  class="">
                <a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/tcpm">https://www.ietf.org/mailman/listinfo/tcpm</a><br class="">
              </div>
            </div>
          </blockquote>
        </div>
        <br class="">
      </div>
    </blockquote>
    <br>
    <pre class="moz-signature" cols="72">-- 
________________________________________________________________
Bob Briscoe                               <a class="moz-txt-link-freetext" href="http://bobbriscoe.net/">http://bobbriscoe.net/</a></pre>
  </body>
</html>

--------------ec4zBqo0WTp2fqwJE0meOgrp--


From nobody Fri Feb  4 13:11:50 2022
Return-Path: <Michael.Scharf@hs-esslingen.de>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5D75B3A237D for <tcpm@ietfa.amsl.com>; Fri,  4 Feb 2022 13:11:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.097
X-Spam-Level: 
X-Spam-Status: No, score=-2.097 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, SPF_HELO_NONE=0.001, SPF_NONE=0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=hs-esslingen.de
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3HWQ6nJgAzpE for <tcpm@ietfa.amsl.com>; Fri,  4 Feb 2022 13:11:43 -0800 (PST)
Received: from mail.hs-esslingen.de (mail.hs-esslingen.de [134.108.32.78]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C71683A237A for <tcpm@ietf.org>; Fri,  4 Feb 2022 13:11:42 -0800 (PST)
Received: from localhost (localhost.localdomain [127.0.0.1]) by mail.hs-esslingen.de (Postfix) with ESMTP id A60C625A18; Fri,  4 Feb 2022 22:11:39 +0100 (CET)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=hs-esslingen.de; s=mail; t=1644009099; bh=yYtDXEhR7BMkJNVocwkJOmGsn7PQQnNfYT6w7raBwS8=; h=From:To:CC:Subject:Date:References:In-Reply-To:From; b=HelBQ+zDOYyiotE0zD/u8gA34eOCTDOEvLb/N5eTDXqCHkOJMjUZTKZ2aIQFGonLQ gYRqJMimND/Mk6hlaAZOKcLdLIPguwXQIjNkUy6o+ysRFY9T6Jkm+hgB9SCZSVxEOZ xiGsw8/0DFLsKne8guN4khm0UdhoDDBs8z0m1E3A=
X-Virus-Scanned: by amavisd-new-2.7.1 (20120429) (Debian) at hs-esslingen.de
Received: from mail.hs-esslingen.de ([127.0.0.1]) by localhost (hs-esslingen.de [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yivmb5fUEZ4T; Fri,  4 Feb 2022 22:11:38 +0100 (CET)
Received: from rznt8201.rznt.rzdir.fht-esslingen.de (rznt8201.hs-esslingen.de [134.108.48.164]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.hs-esslingen.de (Postfix) with ESMTPS; Fri,  4 Feb 2022 22:11:38 +0100 (CET)
Received: from rznt8202.rznt.rzdir.fht-esslingen.de (134.108.48.165) by rznt8201.rznt.rzdir.fht-esslingen.de (134.108.48.164) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.18; Fri, 4 Feb 2022 22:11:37 +0100
Received: from rznt8202.rznt.rzdir.fht-esslingen.de ([fe80::aca4:171a:3ee1:57e0]) by rznt8202.rznt.rzdir.fht-esslingen.de ([fe80::aca4:171a:3ee1:57e0%3]) with mapi id 15.01.2375.018; Fri, 4 Feb 2022 22:11:37 +0100
From: "Scharf, Michael" <Michael.Scharf@hs-esslingen.de>
To: "mohamed.boucadair@orange.com" <mohamed.boucadair@orange.com>
CC: "tcpm@ietf.org" <tcpm@ietf.org>
Thread-Topic: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
Thread-Index: AQHXyc25qfTSowdLSEa6dKhesfqRYKv/uF0AgEvBUuCABenxUIAYa72QgBqz19A=
Date: Fri, 4 Feb 2022 21:11:37 +0000
Message-ID: <bdc3831619dd442dbc5e2914fff7a2b5@hs-esslingen.de>
References: <163518613944.7645.2494692847367626249@ietfa.amsl.com> <18622_1636712195_618E3F03_18622_35_1_787AE7BB302AE849A7480A190F8B933035450859@OPEXCAUBMA2.corporate.adroot.infra.ftgroup> <bde902389b8a4c76aae776ea2d68d9ce@hs-esslingen.de> <3337_1641201204_61D2BE34_3337_132_1_787AE7BB302AE849A7480A190F8B93303546BDC5@OPEXCAUBMA2.corporate.adroot.infra.ftgroup> <706848a62fd74d18a34566046369d049@hs-esslingen.de>
In-Reply-To: <706848a62fd74d18a34566046369d049@hs-esslingen.de>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
msip_labels: MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_Enabled=true; MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_SetDate=2022-01-03T08:26:45Z;  MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_Method=Privileged; MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_Name=unrestricted_parent.2; MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_SiteId=90c7a20a-f34b-40bf-bc48-b9253b6f5d20; MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_ActionId=49c250bb-949e-4569-b0b4-01c9d91c8976; MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_ContentBits=0
x-originating-ip: [134.108.140.249]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/8KWF_UWhEtvvKZrFxf-MTpDWW94>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 04 Feb 2022 21:11:49 -0000

For what it is worth, -06 with the mentioned changes has been published.

The diff can be found at: https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tc=
pm-yang-tcp-06.txt

I hope that this update fixes all issues.

Michael

> -----Original Message-----
> From: tcpm <tcpm-bounces@ietf.org> On Behalf Of Scharf, Michael
> Sent: Tuesday, January 18, 2022 10:32 PM
> To: mohamed.boucadair@orange.com
> Cc: tcpm@ietf.org
> Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
>=20
> Hi Med,
>=20
> Thanks a lot for this further review with good catches.
>=20
> We have addressed all comments in the current working copy for -06, excep=
t
> for the disclaimer text regarding line wrap in B.1. This disclaimer is cu=
rrently
> automatically introduced by the build tools no matter whether it is neede=
d.
> That detail can be sorted out later by the RFC editor.
>=20
> Details can be found at the repository with the working copy:
> https://github.com/mjethanandani/ietf-tcp
>=20
> Our suggestion is to change the TCP-AO example so that it illustrates the=
 use
> of send-id and recv-id with parameters documented in draft-ietf-tcpm-ao-
> test-vectors.
>=20
> As the WGLC is ongoing, we would wait for further comments before
> finalizing -06.
>=20
> Thanks
>=20
> Michael (as author)
>=20
>=20
> > -----Original Message-----
> > From: mohamed.boucadair@orange.com
> > <mohamed.boucadair@orange.com>
> > Sent: Monday, January 3, 2022 10:13 AM
> > To: Scharf, Michael <Michael.Scharf@hs-esslingen.de>
> > Cc: tcpm@ietf.org
> > Subject: RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
> >
> > Hi Michael,
> >
> > The changes look good. Thank you.
> >
> > Please find below some comments on the new version:
> > - Section 4
> >
> > (1)
> > OLD:
> >        reference
> >          "RFC XXXX, YANG Model for Transmission Control Protocol (TCP)
> >                     Configuration.";
> >
> > NEW:
> >        reference
> >          "RFC XXXX: A YANG Model for Transmission Control Protocol (TCP=
)
> >                     Configuration";
> >
> > (2)
> >
> > Add "when set to true" in the description of almost all the Boolean nod=
es.
> >
> > (3)
> >
> > I still don't think "forms the connection identifier" is accurate. The
> connection
> > is identified by the 4-uple (which is captured in the key statement). Y=
ou
> may
> > consider simply: s/forms the connection identifier/of the connection.
> > Another better approach would be, e.g.:
> >
> > OLD:
> >  "Local address that forms the connection identifier.";
> > NEW:
> >  "Identifies the address that is used locally by an endpoint to bind th=
e
> > connection.";
> >
> > OLD:
> >  "Remote address that forms the connection identifier.";
> > NEW:
> >  "Identifies the address that is used by the peer endpoint to bind the
> > connection.";
> >
> > etc.
> >
> > - Make this change in 5.2: s/registrations are requested/registration i=
s
> > requested
> > - Update 5.2 to make it explicit that the module is not maintained by I=
ANA:
> >
> > OLD:
> >       name:         ietf-tcp
> >       namespace:    urn:ietf:params:xml:ns:yang:ietf-tcp
> >       prefix:       tcp
> >       reference:    RFC XXXX
> >
> > NEW:
> >       name:         ietf-tcp
> >       namespace:    urn:ietf:params:xml:ns:yang:ietf-tcp
> >       prefix:       tcp
> >       maintained by IANA: N
> >       reference:    RFC XXXX
> >
> > - I think you can remove the note in B.1
> > - remove "\" right after "</description>" in B.2
> > - As you are referring to RFC 8792, you need to add the 8792 header to
> > adhere with the following (B.2):
> >
> > =3D=3D=3D=3D=3D=3D
> > ...
> >    Text content that has been folded as specified by this strategy MUST
> >    adhere to the following structure.
> >
> > 7.1.1.  Header
> >
> >    The header is two lines long.
> >
> >    The first line is the following 36-character string; this string MAY
> >    be surrounded by any number of printable characters.  This first lin=
e
> >    cannot itself be folded.
> >
> >    NOTE: '\' line wrapping per RFC 8792
> >
> >    The second line is an empty line, containing only the end-of-line
> >    character sequence.  This line provides visual separation for
> >    readability.
> > =3D=3D=3D=3D=3D
> >
> > - B.2:
> >
> > <!--
> > This example sets TCP-AO configuration parameters as
> > demonstrated by examples in draft-touch-tcpm-ao-test-vectors.
> > -->
> >
> > Not sure why this I-D is cited here given the example does not echo an
> > example from that I-D (e.g., same local/remote port, local address, etc=
.).
> > Also, I was expecting to see an example that illustrates the use of sen=
d-id
> > and recv-id.
> >
> > If the citation is maintained, please make this change:
> >
> > s/demonstrated by examples in draft-touch-tcpm-ao-test-
> > vectors/demonstrated by examples in I-D.ietf-tcpm-ao-test-vectors
> >
> > Cheers,
> > Med
> >
> > > -----Message d'origine-----
> > > De=A0: Scharf, Michael <Michael.Scharf@hs-esslingen.de>
> > > Envoy=E9=A0: jeudi 30 d=E9cembre 2021 15:19
> > > =C0=A0: BOUCADAIR Mohamed INNOV/NET
> > <mohamed.boucadair@orange.com>
> > > Cc=A0: tcpm@ietf.org
> > > Objet=A0: RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
> > >
> > > Hi Med,
> > >
> > > Thanks a lot for the detailed review. I went through all your comment=
s.
> > > Most were straightforward to address.
> > >
> > > Our suggestion is to clearly state that MPTCP is outside the scope of=
 this
> > > model. MPTCP may need a different YANG model. Yet, the most obvious
> > use
> > > case of the YANG model are TCP-based control plane protocols on
> routers
> > > (BGP, LDP, ...). At least I am not aware of widespread use of BGP ove=
r
> > > MPTCP.
> > >
> > > Can you please have a look at -05 and let me know if -05 works for yo=
u?
> > >
> > > The full diff is available at: https://www.ietf.org/rfcdiff?url2=3Ddr=
aft-
> > > ietf-tcpm-yang-tcp-05
> > >
> > > Thanks
> > >
> > > Michael
> > >
> > > > -----Original Message-----
> > > > From: mohamed.boucadair@orange.com
> > > > <mohamed.boucadair@orange.com>
> > > > Sent: Friday, November 12, 2021 11:17 AM
> > > > To: Scharf, Michael <Michael.Scharf@hs-esslingen.de>
> > > > Cc: tcpm@ietf.org
> > > > Subject: RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
> > > >
> > > > Hi Michael,
> > > >
> > > > FWIW, please find below some comments to the latest version of the
> > > draft:
> > > >
> > > > * pdf: https://raw.githubusercontent.com/boucadair/IETF-Drafts-
> > > > Reviews/master/draft-ietf-tcpm-yang-tcp-04-rev%20Med.pdf
> > > > * doc: https://github.com/boucadair/IETF-Drafts-
> > > > Reviews/raw/master/draft-ietf-tcpm-yang-tcp-04-rev%20Med.doc
> > > >
> > > > I think that the document is almost stable. Thank you.
> > > >
> > > > Cheers,
> > > > Med
> > > >
> > > > > -----Message d'origine-----
> > > > > De=A0: tcpm <tcpm-bounces@ietf.org> De la part de
> > > > > internet-drafts@ietf.org Envoy=E9=A0: lundi 25 octobre 2021 20:22=
 =C0=A0:
> > > > > i-d-announce@ietf.org Cc=A0: tcpm@ietf.org Objet=A0: [tcpm] I-D A=
ction:
> > > > > draft-ietf-tcpm-yang-tcp-04.txt
> > > > >
> > > > >
> > > > > A New Internet-Draft is available from the on-line Internet-Draft=
s
> > > > > directories.
> > > > > This draft is a work item of the TCP Maintenance and Minor
> > > > > Extensions WG of the IETF.
> > > > >
> > > > >         Title           : YANG Model for Transmission Control Pro=
tocol
> > > > > (TCP) Configuration
> > > > >         Authors         : Michael Scharf
> > > > >                           Mahesh Jethanandani
> > > > >                           Vishal Murgai
> > > > > 	Filename        : draft-ietf-tcpm-yang-tcp-04.txt
> > > > > 	Pages           : 23
> > > > > 	Date            : 2021-10-25
> > > > >
> > > > > Abstract:
> > > > >    This document specifies a minimal YANG model for TCP on device=
s
> > > that
> > > > >    are configured by network management protocols.  The YANG
> model
> > > > >    defines a container for all TCP connections and groupings of
> > > > >    authentication parameters that can be imported and used in TCP
> > > > >    implementations or by other models that need to configure TCP
> > > > >    parameters.  The model also includes basic TCP statistics.  Th=
e
> > > model
> > > > >    is NMDA (RFC 8342) compliant.
> > > > >
> > > > >
> > > > > The IETF datatracker status page for this draft is:
> > > > > https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/
> > > > >
> > > > > There is also an htmlized version available at:
> > > > > https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-yang-tcp-04
> > > > >
> > > > > A diff from the previous version is available at:
> > > > > https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-yang-tcp-04
> > > > >
> > > > >
> > > > > Internet-Drafts are also available by anonymous FTP at:
> > > > > ftp://ftp.ietf.org/internet-drafts/
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > tcpm mailing list
> > > > > tcpm@ietf.org
> > > > > https://www.ietf.org/mailman/listinfo/tcpm
> > > >
> > > >
> >
> __________________________________________________________
> > > >
> >
> __________________________________________________________
> > > > _____
> > > >
> > > > Ce message et ses pieces jointes peuvent contenir des informations
> > > > confidentielles ou privilegiees et ne doivent donc pas etre diffuse=
s,
> > > > exploites ou copies sans autorisation. Si vous avez recu ce message
> > > > par erreur, veuillez le signaler a l'expediteur et le detruire ains=
i
> > > > que les pieces jointes. Les messages electroniques etant susceptibl=
es
> > > > d'alteration, Orange decline toute responsabilite si ce message a e=
te
> > > > altere, deforme ou falsifie. Merci.
> > > >
> > > > This message and its attachments may contain confidential or
> > > > privileged information that may be protected by law; they should no=
t
> > > > be distributed, used or copied without authorisation.
> > > > If you have received this email in error, please notify the sender =
and
> > > > delete this message and its attachments.
> > > > As emails may be altered, Orange is not liable for messages that ha=
ve
> > > > been modified, changed or falsified.
> > > > Thank you.
> >
> >
> >
> __________________________________________________________
> >
> __________________________________________________________
> > _____
> >
> > Ce message et ses pieces jointes peuvent contenir des informations
> > confidentielles ou privilegiees et ne doivent donc
> > pas etre diffuses, exploites ou copies sans autorisation. Si vous avez =
recu ce
> > message par erreur, veuillez le signaler
> > a l'expediteur et le detruire ainsi que les pieces jointes. Les message=
s
> > electroniques etant susceptibles d'alteration,
> > Orange decline toute responsabilite si ce message a ete altere, deforme=
 ou
> > falsifie. Merci.
> >
> > This message and its attachments may contain confidential or privileged
> > information that may be protected by law;
> > they should not be distributed, used or copied without authorisation.
> > If you have received this email in error, please notify the sender and =
delete
> > this message and its attachments.
> > As emails may be altered, Orange is not liable for messages that have b=
een
> > modified, changed or falsified.
> > Thank you.
>=20
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm


From nobody Fri Feb  4 17:15:00 2022
Return-Path: <vidhi_goel@apple.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B23A33A2BBE for <tcpm@ietfa.amsl.com>; Fri,  4 Feb 2022 17:14:57 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.674
X-Spam-Level: 
X-Spam-Status: No, score=-2.674 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.576, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=apple.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WmGfQiWuFeuq for <tcpm@ietfa.amsl.com>; Fri,  4 Feb 2022 17:14:53 -0800 (PST)
Received: from rn-mailsvcp-ppex-lapp24.apple.com (rn-mailsvcp-ppex-lapp24.rno.apple.com [17.179.253.38]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1E4A23A2BBC for <tcpm@ietf.org>; Fri,  4 Feb 2022 17:14:53 -0800 (PST)
Received: from pps.filterd (rn-mailsvcp-ppex-lapp24.rno.apple.com [127.0.0.1]) by rn-mailsvcp-ppex-lapp24.rno.apple.com (8.16.1.2/8.16.1.2) with SMTP id 2151AwsN026437; Fri, 4 Feb 2022 17:14:42 -0800
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=apple.com; h=from : message-id : content-type : mime-version : subject : date : in-reply-to : cc : to : references; s=20180706; bh=FL+E1KryqV7/uimTz4nIJB9aarK8aAra+CEdfMrtjdA=; b=rSfUa8CbGvV/IJw/yOwSQMqxH49HSBNUUGZRm3K6YoUbHJ/vDirMkHsGn9tIIWkeJ4Lk LmwKLVSufppvJWD4uHpga+XGd0g1O3zO8EbDsCqc6oAoWYh9ee5htiW9Wml0UrMvrAWm qu3eQnMLDVDTfChWGJp1akS3oRwI7adlU9skUARB+/rEczjJY2tHTcneOzHWoQ3wmI/e xMt2TIF1VzYFmRO69UfBKKf8ISDyZM5TePKZXKRV4rGo9gqvA6fOE/9clPqfwiBK2QOW w6d1bX+vJsiVo7fb0fOMHZo6N4i6ZYd/f0pZMY/ydeDO+53qL8AWrUCci7KaPuOVV+Jq rg== 
Received: from rn-mailsvcp-mta-lapp03.rno.apple.com (rn-mailsvcp-mta-lapp03.rno.apple.com [10.225.203.151]) by rn-mailsvcp-ppex-lapp24.rno.apple.com with ESMTP id 3e0rd26nqq-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NO); Fri, 04 Feb 2022 17:14:42 -0800
Received: from rn-mailsvcp-mmp-lapp04.rno.apple.com (rn-mailsvcp-mmp-lapp04.rno.apple.com [17.179.253.17]) by rn-mailsvcp-mta-lapp03.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPS id <0R6T00QOC4SH2560@rn-mailsvcp-mta-lapp03.rno.apple.com>;  Fri, 04 Feb 2022 17:14:41 -0800 (PST)
Received: from process_milters-daemon.rn-mailsvcp-mmp-lapp04.rno.apple.com by rn-mailsvcp-mmp-lapp04.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) id <0R6T00I004KYN100@rn-mailsvcp-mmp-lapp04.rno.apple.com>; Fri, 04 Feb 2022 17:14:41 -0800 (PST)
X-Va-A: 
X-Va-T-CD: 5b1df986fc1cf28f03a98c189e99a4df
X-Va-E-CD: d27b0bf1ec5e225c315eb75773194505
X-Va-R-CD: 7f6ba7ce084ea0aad2ff4f136dfa7fc9
X-Va-CD: 0
X-Va-ID: feb9054c-618e-4993-91e5-10ef028f8eb5
X-V-A: 
X-V-T-CD: 5b1df986fc1cf28f03a98c189e99a4df
X-V-E-CD: d27b0bf1ec5e225c315eb75773194505
X-V-R-CD: 7f6ba7ce084ea0aad2ff4f136dfa7fc9
X-V-CD: 0
X-V-ID: 38f9f43c-65e7-4bc6-88e9-0b9798ccefd9
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-04_07:2022-02-03, 2022-02-04 signatures=0
Received: from smtpclient.apple (unknown [17.234.96.218]) by rn-mailsvcp-mmp-lapp04.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPSA id <0R6T00AIU4SGB100@rn-mailsvcp-mmp-lapp04.rno.apple.com>; Fri, 04 Feb 2022 17:14:41 -0800 (PST)
From: Vidhi Goel <vidhi_goel@apple.com>
Message-id: <BB23E172-C995-41FE-A47E-E02D5586B67D@apple.com>
Content-type: multipart/signed; boundary="Apple-Mail=_4844375E-DDC1-4B78-B59B-E77932053680"; protocol="application/pkcs7-signature"; micalg=sha-256
MIME-version: 1.0 (Mac OS X Mail 14.0 \(3654.100.0.2.11\))
Date: Fri, 04 Feb 2022 17:14:39 -0800
In-reply-to: <e57355ae-5a9d-0a14-9478-29453d25622c@bobbriscoe.net>
Cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, Mirja Kuehlewind <ietf@kuehlewind.net>, Richard Scheffenegger <rscheff@gmx.at>
To: Bob Briscoe <ietf@bobbriscoe.net>
References: <164389942954.13556.7754569279919863072@ietfa.amsl.com> <c2c5bdf9-6b34-fe19-3296-5f1cd831ed8e@bobbriscoe.net> <2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com> <e57355ae-5a9d-0a14-9478-29453d25622c@bobbriscoe.net>
X-Mailer: Apple Mail (2.3654.100.0.2.11)
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-04_07:2022-02-03, 2022-02-04 signatures=0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/suOSD615PBvsTRVnR_4BeMzkXS8>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-accurate-ecn-16.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 05 Feb 2022 01:14:58 -0000

--Apple-Mail=_4844375E-DDC1-4B78-B59B-E77932053680
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_2310DDF5-C279-419F-90E2-6C486952D0E4"


--Apple-Mail=_2310DDF5-C279-419F-90E2-6C486952D0E4
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

I like your response for all the points. I can review the diff for the =
proposed changes, if you=E2=80=99d like before committing them to the =
draft.

Sorry, I want to add a new comment for this text,
If a TCP client has set the SYN to Not-ECT, but
   receives feedback that the IP-ECN field on the SYN arrived with a
   different codepoint, it can detect such middlebox interference and
   send Not-ECT for the rest of the connection.

This statement holds good for Not-ECT but doesn=E2=80=99t if lets say =
the SYN was ECT0 or ECT1 as those could be marked CE and still be valid. =
Should we add a statement for ECT marked SYN as well?

Thanks,
Vidhi

> On Feb 4, 2022, at 5:44 AM, Bob Briscoe <ietf@bobbriscoe.net> wrote:
>=20
> Vidhi,
>=20
> I've increased the table of contents to tocdepth =3D "4" which might =
help find the relevant sections better, because a lot of the meat in =
this draft is under sections 3.2.2 (ACE) and 3.2.3 (Option).
>=20
> I've responded below, but I'd prefer it if you would say these things =
on the list, so we have a better justification for changing the draft. =
Reviewing a draft that has just be re-posted shouldn't imply anything =
about whether Apple is implementing the draft or not.
>=20
> (BTW, regarding the proposed change to the initial value of r.e1b, I =
was going to send that  to the list once co-authors agreed to it. But it =
would be nice if some of the edits were not self-generated.)
>=20
> See [BB] inline for responses (which I'll repeat if you send to the =
list).
>=20
> On 04/02/2022 07:37, Vidhi Goel wrote:
>> Hello Bob and authors,
>>=20
>> I have read the draft a few times and will send my extensive review =
at a later point. But right now, I want to ask some critical things that =
an implementation could benefit from.
>>=20
>> The draft says,
>> If a TCP client has set the SYN to Not-ECT, but
>>    receives feedback that the IP-ECN field on the SYN arrived with a
>>    different codepoint, it can detect such middlebox interference and
>>    send Not-ECT for the rest of the connection
>>=20
>> 1. On the forward path from client to server, the client will revert =
to Not-ECT when it sees for example, that it sent a Non-ECT SYN but =
received ACE encoding other than 0 1 0. What does the client do in the =
last ACK of the 3WHS -
>>  a. does it stay in AccECN mode and still send AccECN encoding based =
on the IP code point of SYN-ACK (Table 4)? This would mean that client =
won=E2=80=99t participate in ECN on the sender half of the connection =
and only provide AccECN feedback to the server as a receiver.
>>  b.  does it disable AccECN mode and set ACE=3D 0 0 0 so that a =
server in AccECN mode can disable ECN based on Table4?
>>=20
>> While writing this, I realized that the intention is probably a. =
Could you confirm? Also, when the sender sets Not-ECT in its data =
packets, it should also disable acting upon any ACE feedback as we could =
still receive false ACE feedback from the server if the network, lets =
say, changed 00 (not ECT) to 01(ECT1). If you agree, we should add some =
text around this. Based on the current text, the sender will always =
respond to the ACE feedback even if it sends Not-ECT.
>>=20
>> TBH, this is a complicated scenario, where sender said to the network =
- I don=E2=80=99t trust you so I can=E2=80=99t use ECN. Feel free to =
drop my packets. And the network mangles the IP to ECT1 and then set CE =
(when congested) which would be feedback=E2=80=99ed from the receiver. =
Now, the packet wasn=E2=80=99t dropped which it should have been. So, is =
it better to just ignore this feedback because sender doesn=E2=80=99t =
trust the network or just act on it and reduce cwnd in order to reduce =
congestion in the network somewhere.
>=20
> [BB] Your points are all good ones.
> I'll address this last complex scenario first, because it has =
implications for the earlier questions. First I'll define some =
terminology:
>=20
> Simple mangling scenario:=20
> some network function, e.g. broken Diffserv ToS-byte remarking, =
illegally remarks a Not-ECT  SYN to CE.
> Your complex mangling scenario, repeated here:
> some network function, e.g. broken Diffserv ToS-byte remarking, =
illegally remarks a Not-ECT SYN to ECT0 or ECT1,=20
> then congestion at a subsequent bottleneck is marking some packets CE
> Then potentially the client could tailor its behaviour after sending a =
Not-ECT SYN, If the SYN/ACK feedback is
> ECT: disable sending ECT, but continue responding to ECN feedback
> CE, disable sending ECT, and disable response to ECN feedback
> However, this is uncertain, because CE feedback on the SYN/ACK could =
indicate either the simple or the complex mangling scenario.
>=20
> A simpler alternative would be to always continue responding to ECN =
feedback. Rationale:=20
> Whether case #1 or #2, assume that mangling of the SYN might have been =
Not-ECT to ECT, even if the feedback off the SYN is CE.
> Then as the connection progresses, if /all/ feedback is CE, there's =
already a recommendation to fall-back to disabling congestion response.
> If we do this, I think we ought to say "SHOULD continue to respond to =
ECN feedback", not "MUST".
>=20
> And we'll need to put this all to the WG.=20
>=20
> 1.a) Now back to the beginning of your point. The text you quote is =
from =C2=A72.5 which is in the non-normative "Overview and Rationale" =
section (=C2=A72).=20
> You really need the normative text from:=20
> 3.2.2.3.  =
<https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#sect=
ion-3.2.2.3>Testing for Mangling of the IP/ECN Field =
<https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name=
-testing-for-mangling-of-the>where it says (end of 1st para):=20
> " ...for the remainder of the connection the client MUST NOT send =
ECN-capable packets, but it MUST continue to feed back any ECN markings =
on arriving packets."
> So, you're right, the answer is (a).
> It can't be (b), because, once in AccECN mode, 000 on the ACE field =
just becomes a counter value, and no longer negotiates the feedback =
mode.
>=20
> 1.b) As per the above, let's conservatively assume complex mangling. =
So, at the end of the first para quoted above I suggest we add:
> " ...for the remainder of the connection the client MUST NOT send =
ECN-capable packets but it SHOULD continue to respond to ECN feedback =
even though it is no longer sending ECN-capable packets (see reasoning =
below). The client MUST remain in the same feedback mode and it MUST =
continue to feed back any ECN markings on arriving packets."
> I'll do the same for next para about the server.=20
> I'll work out some text for the reasoning, but I won't give it here.
>=20
> There are two other cases where it disables sending ECT, which don't =
say whether it continues to respond to congestion:
> "Testing for Mangling" section, penultimate para,  where it's =
receiving solid CE:
> "Once a Data Sender has entered AccECN mode it SHOULD check whether =
all feedback received for the first three or four rounds indicated that =
every packet it sent was CE-marked. If so, for the remainder of the =
connection, the Data Sender SHOULD NOT send ECN-capable packets and it =
SHOULD NOT respond to ECN feedback, but it MUST remain in the same =
feedback mode and it MUST continue to feed back any ECN markings on =
arriving packets (in its role as Data Receiver)."
> Next section "Zeroing of the ACE Field", 2nd para:
> "If the value of this ACE field is zero (0b000), the Data Sender =
disables sending ECN-capable packets for the remainder of the =
half-connection by setting the IP/ECN field in all subsequent packets to =
Not-ECT, but it SHOULD continue to respond to ECN feedback even though =
it is no longer sending ECN-capable packets. It MUST also remain in the =
same feedback mode and it MUST continue to feed back any ECN markings on =
arriving packets (in its role as Data Receiver)."
> I'll also add reasoning (in the zeroing section, it already says that =
ACE=3D0b000 is not an unambiguous indication of mangling).
>=20
> And this prompts me to edit the bullet in an earlier section about the =
obligation of a sender to respond to congestion feedback:
> 3.1.5.  =
<https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#sect=
ion-3.1.5>Implications of AccECN Mode =
<https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name=
-implications-of-accecn-mode>
> "It is still obliged to respond appropriately to AccECN feedback that =
indicates there were ECN marks on ECN-capable packets it had previously =
sent."
> And I'll add a bullet:
> "If the sender chooses not to send ECN-capable packets (e.g. because =
path traversal of the ECN field is suspect), it can ignore any ECN =
feedback about those packets if it is certain that it cannot be valid =
(see Section 3.2.2, which gives normative requirements for certain =
specific cases)."
> I've said 'can' rather than 'MUST' because it's hard to cover all =
cases, e.g. single packets without ECT when it's not clear whether the =
feedback covered another packet that was ECT.=20
>=20
> How does all this sound?
>=20
>> 2. On the reverse path from server to client, if a server sends a =
Not-ECT SYN-ACK and receives ACE handshake encoding on last ACK other =
than 0 1 0, there is no text like above that says server should send =
Not-ECT for the rest of the connection (or at least I didn=E2=80=99t =
find it). I think the server should also do same as client as the two =
paths could be different. One could make it more complicated by saying, =
if both client and server decide to not use ECN on their corresponding =
sender half, then ECN should be disabled but let=E2=80=99s talk about =
that later.
>=20
> [BB] It's in the second para of:
> 3.2.2.3.  =
<https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#sect=
ion-3.2.2.3>Testing for Mangling of the IP/ECN Field =
<https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name=
-testing-for-mangling-of-the>3. In general, not setting the ECT (ECT0 or =
ECT1) code point on an outgoing packet is different from supporting =
AccECN right as in the host can still provide AccECN feedback on the =
receive path.
>=20
> [BB] Yup.
>=20
> Do you think any further explanation is needed in the above sections?
>=20
>=20
> Bob
>=20
>> =E2=80=A6.
>>=20
>> To be continued if more questions come to my mind.
>>=20
>>=20
>> Thanks,
>> Vidhi
>>=20
>>> On Feb 3, 2022, at 7:24 AM, Bob Briscoe <ietf@bobbriscoe.net =
<mailto:ietf@bobbriscoe.net>> wrote:
>>>=20
>>> tcpm folks,
>>>=20
>>> This rev to accurate-ecn is the first of two. The second will =
hopefully follow on its heels in the next couple of days.
>>>=20
>>> Diffs between this first rev (-16) and -15:
>>>=20
>>> 1.  switches round two fairly large sections, so I've deferred other =
changes to a second rev so the diffs won't be masked by the switch round =
of sections.
>>> Suggested by Ilpo to match the order in which the tests in these =
sections will be executed:
>>> * Test for mangling the IP-ECN field (now 3.2.2.3),
>>> * Then test for zeroing the ACE field (now 3.2.2.4).
>>>=20
>>> 2. Ilpo suggested some clarifications in "3.2.3.2.5.  Consistency =
between AccECN Feedback Fields", which is about the receiver of feedback =
ensuring consistency between the mandatory 3-bit ACE field and the =
optional 24-bit counters. In brief (paraphrasing) it previously only =
said "MUST consider both fields", when it is now clearer that it =
actually meant "MUST reconcile both fields", so that there is always a =
consistent baseline for subsequent ACKs.
>>>=20
>>> 3. A minor point is added in an appendix about the details that the =
pseudocode doesn't cover.
>>>=20
>>>=20
>>>=20
>>> Bob
>>>=20
>>> PS. #2 & #3 were added to the XML ages ago (Jul '21), so you will =
have seen them in the HTML. However, prob due to my clumsiness, the =
posted TXT didn't include them whereas the posted XML did (ironic for a =
section about consistency). In turn, inconsistency was only possible =
because I am having to manually post the TXT for this draft, due to an =
unresolved issue with v3 RFC XML tables.
>>>=20
>>>=20
>>> On 03/02/2022 14:43, internet-drafts@ietf.org =
<mailto: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 TCP Maintenance and Minor =
Extensions WG of the IETF.
>>>>=20
>>>>         Title           : More Accurate ECN Feedback in TCP
>>>>         Authors         : Bob Briscoe
>>>>                           Mirja K=C3=BChlewind
>>>>                           Richard Scheffenegger
>>>> 	Filename        : draft-ietf-tcpm-accurate-ecn-16.txt
>>>> 	Pages           : 60
>>>> 	Date            : 2022-02-03
>>>>=20
>>>> Abstract:
>>>>    Explicit Congestion Notification (ECN) is a mechanism where =
network
>>>>    nodes can mark IP packets instead of dropping them to indicate
>>>>    incipient congestion to the end-points.  Receivers with an ECN-
>>>>    capable transport protocol feed back this information to the =
sender.
>>>>    ECN was originally specified for TCP in such a way that only one
>>>>    feedback signal can be transmitted per Round-Trip Time (RTT).  =
Recent
>>>>    new TCP mechanisms like Congestion Exposure (ConEx), Data Center =
TCP
>>>>    (DCTCP) or Low Latency Low Loss Scalable Throughput (L4S) need =
more
>>>>    accurate ECN feedback information whenever more than one marking =
is
>>>>    received in one RTT.  This document specifies a scheme to =
provide
>>>>    more than one feedback signal per RTT in the TCP header.  Given =
TCP
>>>>    header space is scarce, it allocates a reserved header bit =
previously
>>>>    assigned to the ECN-Nonce.  It also overloads the two existing =
ECN
>>>>    flags in the TCP header.  The resulting extra space is exploited =
to
>>>>    feed back the IP-ECN field received during the 3-way handshake =
as
>>>>    well.  Supplementary feedback information can optionally be =
provided
>>>>    in a new TCP option, which is never used on the TCP SYN.  The
>>>>    document also specifies the treatment of this updated TCP wire
>>>>    protocol by middleboxes.
>>>>=20
>>>>=20
>>>> The IETF datatracker status page for this draft is:
>>>> https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/ =
<https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/>
>>>>=20
>>>> There is also an htmlized version available at:
>>>> =
https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16 =
<https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16>
>>>>=20
>>>> A diff from the previous version is available at:
>>>> https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-accurate-ecn-16 =
<https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-accurate-ecn-16>
>>>>=20
>>>>=20
>>>> Internet-Drafts are also available by rsync at =
rsync.ietf.org::internet-drafts
>>>>=20
>>>>=20
>>>> _______________________________________________
>>>> tcpm mailing list
>>>> tcpm@ietf.org <mailto:tcpm@ietf.org>
>>>> https://www.ietf.org/mailman/listinfo/tcpm =
<https://www.ietf.org/mailman/listinfo/tcpm>
>>>=20
>>> --=20
>>> ________________________________________________________________
>>> Bob Briscoe                               http://bobbriscoe.net/ =
<http://bobbriscoe.net/>
>>>=20
>>> _______________________________________________
>>> tcpm mailing list
>>> tcpm@ietf.org <mailto:tcpm@ietf.org>
>>> https://www.ietf.org/mailman/listinfo/tcpm =
<https://www.ietf.org/mailman/listinfo/tcpm>
>>=20
>=20
> --=20
> ________________________________________________________________
> Bob Briscoe                               http://bobbriscoe.net/ =
<http://bobbriscoe.net/>

--Apple-Mail=_2310DDF5-C279-419F-90E2-6C486952D0E4
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=utf-8

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3Dutf-8"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: after-white-space;" class=3D"">I =
like your response for all the points. I can review the diff for the =
proposed changes, if you=E2=80=99d like before committing them to the =
draft.<div class=3D""><br class=3D""></div><div class=3D"">Sorry, I want =
to add a new comment for this text,</div><div class=3D""><pre =
class=3D"newpage" style=3D"font-size: 13.333333015441895px; margin-top: =
0px; margin-bottom: 0px; break-before: page; caret-color: rgb(0, 0, 0); =
color: rgb(0, 0, 0);">If a TCP client has set the SYN to Not-ECT, but
   receives feedback that the IP-ECN field on the SYN arrived with a
   different codepoint, it can detect such middlebox interference and
   send Not-ECT for the rest of the connection.</pre><div class=3D""><br =
class=3D""></div></div><div class=3D""><font color=3D"#000000" =
class=3D"">This statement holds good for Not-ECT but doesn=E2=80=99t if =
lets say the SYN was ECT0 or ECT1&nbsp;as those could be marked CE and =
still be valid.&nbsp;Should we add a statement for ECT marked SYN as =
well?</font></div><div class=3D""><font color=3D"#000000" class=3D""><br =
class=3D""></font></div><div class=3D"">Thanks,</div><div =
class=3D"">Vidhi<br class=3D""><div><br class=3D""><blockquote =
type=3D"cite" class=3D""><div class=3D"">On Feb 4, 2022, at 5:44 AM, Bob =
Briscoe &lt;<a href=3D"mailto:ietf@bobbriscoe.net" =
class=3D"">ietf@bobbriscoe.net</a>&gt; wrote:</div><br =
class=3D"Apple-interchange-newline"><div class=3D"">
 =20
    <meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3DUTF-8" class=3D"">
 =20
  <div class=3D"">
    Vidhi,<br class=3D"">
    <br class=3D"">
    I've increased the table of contents to tocdepth =3D "4" which might
    help find the relevant sections better, because a lot of the meat in
    this draft is under sections 3.2.2 (ACE) and 3.2.3 (Option).<br =
class=3D"">
    <br class=3D"">
    I've responded below, but I'd prefer it if you would say these
    things on the list, so we have a better justification for changing
    the draft. Reviewing a draft that has just be re-posted shouldn't
    imply anything about whether Apple is implementing the draft or =
not.<br class=3D"">
    <br class=3D"">
    (BTW, regarding the proposed change to the initial value of r.e1b, I
    was going to send that&nbsp; to the list once co-authors agreed to =
it.
    But it would be nice if some of the edits were not =
self-generated.)<br class=3D"">
    <br class=3D"">
    See [BB] inline for responses (which I'll repeat if you send to the
    list).<br class=3D"">
    <br class=3D"">
    <div class=3D"moz-cite-prefix">On 04/02/2022 07:37, Vidhi Goel =
wrote:<br class=3D"">
    </div>
    <blockquote type=3D"cite" =
cite=3D"mid:2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com" class=3D"">
      <meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3DUTF-8" class=3D"">
      Hello Bob and authors,
      <div class=3D""><br class=3D"">
      </div>
      <div class=3D"">I have read the draft a few times and will send my
        extensive review at a later point. But right now, I want to ask
        some critical things that an implementation could benefit =
from.</div>
      <div class=3D""><br class=3D"">
      </div>
      <div class=3D"">The draft says,</div>
      <div class=3D""><span style=3D"font-family: &quot;Helvetica
          Neue&quot;; font-size: 13px;" class=3D""><i class=3D"">If a =
TCP
            client has set the SYN to Not-ECT, but</i></span></div>
      <div class=3D"">
        <div style=3D"margin: 0px; font-stretch: normal; font-size: =
13px;
          line-height: normal; font-family: &quot;Helvetica Neue&quot;;" =
class=3D""><i class=3D"">&nbsp;&nbsp; receives feedback that the IP-ECN
            field on the SYN arrived with a</i></div>
        <div style=3D"margin: 0px; font-stretch: normal; font-size: =
13px;
          line-height: normal; font-family: &quot;Helvetica Neue&quot;;" =
class=3D""><i class=3D"">&nbsp;&nbsp; different codepoint, it can detect
            such middlebox interference and</i></div>
        <div style=3D"margin: 0px; font-stretch: normal; font-size: =
13px;
          line-height: normal; font-family: &quot;Helvetica Neue&quot;;" =
class=3D""><i class=3D"">&nbsp;&nbsp; send Not-ECT for the rest of the
            connection</i></div>
        <div class=3D""><br class=3D"">
        </div>
        <div class=3D"">1. On the forward path from client to server, =
the client
          will revert to Not-ECT when it sees for example, that it sent
          a Non-ECT SYN but received ACE encoding other than&nbsp;<font =
class=3D"" size=3D"2"><span style=3D"caret-color:
              rgb(0, 0, 0);" class=3D"">0 1 0.&nbsp;What does the client =
do in
              the last ACK of the 3WHS -</span></font></div>
        <div class=3D""><font class=3D"" size=3D"2">&nbsp;a. does it =
stay in
            AccECN mode and still send AccECN encoding based on the
            IP&nbsp;code point of SYN-ACK (Table 4)? This would mean =
that
            client won=E2=80=99t participate in ECN on the sender half =
of the
            connection and only&nbsp;provide AccECN feedback to the =
server as
            a receiver.</font></div>
        <div class=3D""><font class=3D"" size=3D"2"><span =
style=3D"caret-color: rgb(0, 0, 0);" class=3D"">&nbsp;b. &nbsp;does it
              disable AccECN mode and set ACE=3D 0 0 0 so that a server =
in
              AccECN mode can disable ECN based on =
Table4?</span></font></div>
        <div class=3D""><font class=3D"" size=3D"2"><br class=3D"">
          </font></div>
        <div class=3D""><font class=3D"" size=3D"2">While writing this,
            I realized that the intention is&nbsp;probably a. Could you
            confirm? Also, when the sender sets Not-ECT in its data
            packets, it should also disable acting upon any ACE feedback
            as we could still receive false ACE feedback from the server
            if&nbsp;the network, lets say, changed 00 (not ECT) to =
01(ECT1).
            If you agree, we should add some text around this. Based on
            the current text, the sender will always respond to the ACE
            feedback even if it sends Not-ECT.</font></div>
        <div class=3D""><font class=3D"" size=3D"2"><br class=3D"">
          </font></div>
        <div class=3D""><font class=3D"" size=3D"2">TBH, this is a
            complicated scenario, where sender said to the network - I
            don=E2=80=99t trust you so I can=E2=80=99t use ECN. =
Feel&nbsp;free to drop my
            packets. And the network mangles the IP to ECT1&nbsp;and =
then set
            CE (when congested)&nbsp;which would be feedback=E2=80=99ed =
from the
            receiver. Now, the packet wasn=E2=80=99t dropped which it =
should
            have been. So, is it better to just ignore this feedback
            because sender doesn=E2=80=99t trust the network or just act =
on it
            and reduce cwnd in order to reduce congestion in the network
            somewhere.</font></div>
      </div>
    </blockquote>
    <br class=3D"">
    [BB] Your points are all good ones.<br class=3D"">
    I'll address this last complex scenario first, because it has
    implications for the earlier questions. First I'll define some
    terminology:<br class=3D"">
    <br class=3D"">
    Simple mangling scenario: <br class=3D"">
    <ul class=3D"">
      <li class=3D"">some network function, e.g. broken Diffserv =
ToS-byte
        remarking, illegally remarks a Not-ECT&nbsp; SYN
        to CE. </li>
    </ul>
    Your complex mangling scenario, repeated here:<br class=3D"">
    <ul class=3D"">
      <li class=3D"">some network function, e.g. broken Diffserv =
ToS-byte
        remarking, illegally remarks a Not-ECT SYN to ECT0 or ECT1, <br =
class=3D"">
      </li>
      <li class=3D"">then congestion at a subsequent bottleneck is =
marking some
        packets CE<br class=3D"">
      </li>
    </ul>
    Then potentially the client could tailor its behaviour after sending
    a Not-ECT SYN, If the SYN/ACK feedback is<br class=3D"">
    <ol class=3D"">
      <li class=3D"">ECT: disable sending ECT, but continue responding =
to ECN
        feedback</li>
      <li class=3D"">CE, disable sending ECT, and disable response to =
ECN feedback</li>
    </ol>
    However, this is uncertain, because CE feedback on the SYN/ACK could
    indicate either the simple or the complex mangling scenario.<br =
class=3D"">
    <br class=3D"">
    A simpler alternative would be to always continue responding to ECN
    feedback. Rationale: <br class=3D"">
    <ul class=3D"">
      <li class=3D"">Whether case #1 or #2, assume that mangling of the =
SYN might
        have been Not-ECT to ECT, even if the feedback off the SYN is
        CE. </li>
      <li class=3D"">Then as the connection progresses, if /all/ =
feedback is CE,
        there's already a recommendation to fall-back to disabling
        congestion response.</li>
    </ul>
    If we do this, I think we ought to say "SHOULD continue to respond
    to ECN feedback", not "MUST".<br class=3D"">
    <br class=3D"">
    And we'll need to put this all to the WG. <br class=3D"">
    <br class=3D"">
    1.a) Now back to the beginning of your point. The text you quote is
    from =C2=A72.5 which is in the non-normative "Overview and =
Rationale"
    section (=C2=A72). <br class=3D"">
    You really need the normative text from: <br class=3D"">
    <h5 id=3D"name-testing-for-mangling-of-the" class=3D"">
      <a =
href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.ht=
ml#section-3.2.2.3" class=3D"section-number selfRef">3.2.2.3. </a><a =
href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.ht=
ml#name-testing-for-mangling-of-the" class=3D"selfRef =
section-name">Testing for Mangling of the IP/ECN
        Field</a> </h5>
    where it says (end of 1st para): <br class=3D"">
    <blockquote class=3D"">" ...for the remainder of the connection the =
client MUST
      NOT send ECN-capable packets, but it MUST continue to feed back
      any ECN markings on arriving packets."<br class=3D"">
    </blockquote>
    So, you're right, the answer is (a).<br class=3D"">
    It can't be (b), because, once in AccECN mode, 000 on the ACE field
    just becomes a counter value, and no longer negotiates the feedback
    mode.<br class=3D"">
    <br class=3D"">
    1.b) As per the above, let's conservatively assume complex mangling.
    So, at the end of the first para quoted above I suggest we add:<br =
class=3D"">
    <blockquote class=3D"">" ...for the remainder of the connection the =
client MUST
      NOT send ECN-capable packets <b class=3D"">but</b><b class=3D""> =
it</b><b class=3D""> SHOULD
        continue to respond to ECN feedback even though </b><b =
class=3D"">it </b><b class=3D"">is
        no longer sending ECN-capable packets (see reasoning =
below)</b><b class=3D"">.
        T</b><b class=3D"">he client</b><b class=3D""> MUST remain in =
the same feedback mode
        and</b><b class=3D""> </b>it MUST continue to feed back any ECN =
markings
      on arriving packets."<br class=3D"">
    </blockquote>
    I'll do the same for next para about the server. <br class=3D"">
    I'll work out some text for the reasoning, but I won't give it =
here.<br class=3D"">
    <br class=3D"">
    There are two other cases where it disables sending ECT, which don't
    say whether it continues to respond to congestion:<br class=3D"">
    <ul class=3D"">
      <li class=3D"">"Testing for Mangling" section, penultimate =
para,&nbsp; where it's
        receiving solid CE:<br class=3D"">
        "Once a Data Sender has entered AccECN mode it SHOULD check
        whether all feedback received for the first three or four rounds
        indicated that every packet it sent was CE-marked. If so, for
        the remainder of the connection, the Data Sender SHOULD NOT send
        ECN-capable packets<b class=3D""> and it SHOULD NOT respond to =
ECN feedback</b>,
        but <b class=3D"">it MUST remain in the same feedback mode and =
</b>it
        MUST continue to feed back any ECN markings on arriving =
packets<b class=3D"">
          (in its role as Data Receiver)</b>."</li>
      <li class=3D"">Next section "Zeroing of the ACE Field", 2nd =
para:<br class=3D"">
        "If the value of this ACE field is zero (0b000), the Data Sender
        disables sending ECN-capable packets for the remainder of the
        half-connection by setting the IP/ECN field in all subsequent
        packets to Not-ECT, <b class=3D"">but</b><b class=3D""> it</b><b =
class=3D""> SHOULD continue to
          respond to ECN feedback even though </b><b class=3D"">it =
</b><b class=3D"">is no
          longer sending ECN-capable packets.</b> <b class=3D"">It MUST =
also
          remain in the same feedback mode and it MUST continue to feed
          back any ECN markings on arriving packets (in its role as Data
          Receiver).</b>"<br class=3D"">
      </li>
    </ul>
    I'll also add reasoning (in the zeroing section, it already says
    that ACE=3D0b000 is not an unambiguous indication of mangling).<br =
class=3D"">
    <br class=3D"">
    And this prompts me to edit the bullet in an earlier section about
    the obligation of a sender to respond to congestion feedback:<br =
class=3D"">
    <h4 id=3D"name-implications-of-accecn-mode" class=3D"">
      <a =
href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.ht=
ml#section-3.1.5" class=3D"section-number selfRef">3.1.5. </a><a =
href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.ht=
ml#name-implications-of-accecn-mode" class=3D"selfRef =
section-name">Implications of AccECN Mode</a> </h4>
    <blockquote class=3D"">"It is still obliged to respond appropriately =
to AccECN
      feedback that indicates there were ECN marks on <b =
class=3D"">ECN-capable </b>packets
      it had previously sent."<br class=3D"">
    </blockquote>
    And I'll add a bullet:<br class=3D"">
    <blockquote class=3D"">"<b class=3D"">If the sender chooses not to =
send ECN-capable
        packets (e.g. because path traversal of the ECN field is
        suspect), it can ignore any </b><b class=3D"">ECN </b><b =
class=3D"">feedback about
        those packets if it is certain that it cannot be valid (see
        Section 3.2.2, which gives normative requirements for certain
        specific cases).</b>"<br class=3D"">
    </blockquote>
    I've said 'can' rather than 'MUST' because it's hard to cover all
    cases, e.g. single packets without ECT when it's not clear whether
    the feedback covered another packet that was ECT. <br class=3D"">
    <br class=3D"">
    How does all this sound?<br class=3D"">
    <br class=3D"">
    <blockquote type=3D"cite" =
cite=3D"mid:2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com" class=3D"">
      <div class=3D"">
        <div class=3D""><font class=3D"" size=3D"2">2. On the reverse
            path&nbsp;from server to client, if a server sends a Not-ECT
            SYN-ACK and receives ACE handshake encoding on last
            ACK&nbsp;other than 0 1 0, there is no text like above that =
says
            server should send Not-ECT for the rest of the connection
            (or&nbsp;at least I&nbsp;didn=E2=80=99t find it). I think =
the server should
            also do same as client as&nbsp;the two paths could be =
different.
            One could make it more complicated by saying, if both client
            and server decide to not use ECN on their corresponding
            sender half, then ECN should be disabled but let=E2=80=99s =
talk
            about that later.</font></div>
      </div>
    </blockquote>
    <br class=3D"">
    [BB] It's in the second para of:<br class=3D"">
    <h5 id=3D"name-testing-for-mangling-of-the" class=3D"">
      <a =
href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.ht=
ml#section-3.2.2.3" class=3D"section-number selfRef">3.2.2.3. </a><a =
href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.ht=
ml#name-testing-for-mangling-of-the" class=3D"selfRef =
section-name">Testing for Mangling of the IP/ECN
        Field</a> </h5>
    <blockquote type=3D"cite" =
cite=3D"mid:2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com" class=3D"">
      <div class=3D"">
        <div class=3D""><font class=3D"" size=3D"2">3. In general, not
            setting the ECT (ECT0 or ECT1)&nbsp;code point on an =
outgoing
            packet is different from supporting AccECN right as in the
            host can still provide AccECN feedback on the&nbsp;receive =
path.</font></div>
      </div>
    </blockquote>
    <br class=3D"">
    [BB] Yup.<br class=3D"">
    <br class=3D"">
    Do you think any further explanation is needed in the above
    sections?<br class=3D"">
    <br class=3D"">
    <br class=3D"">
    Bob<br class=3D"">
    <br class=3D"">
    <blockquote type=3D"cite" =
cite=3D"mid:2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com" class=3D"">
      <div class=3D"">
        <div class=3D"">=E2=80=A6.</div>
        <div class=3D""><br class=3D"">
        </div>
        <div class=3D"">To be continued if more questions come to my =
mind.</div>
        <div class=3D""><br class=3D"">
        </div>
        <div class=3D""><br class=3D"">
        </div>
        <div class=3D"">Thanks,</div>
        <div class=3D"">Vidhi</div>
        <div class=3D""><br class=3D"">
        </div>
        <div class=3D"">
          <blockquote type=3D"cite" class=3D"">
            <div class=3D"">On Feb 3, 2022, at 7:24 AM, Bob Briscoe =
&lt;<a href=3D"mailto:ietf@bobbriscoe.net" class=3D"moz-txt-link-freetext"=
 moz-do-not-send=3D"true">ietf@bobbriscoe.net</a>&gt;
              wrote:</div>
            <br class=3D"Apple-interchange-newline">
            <div class=3D"">
              <div class=3D"">tcpm folks,<br class=3D"">
                <br class=3D"">
                This rev to accurate-ecn is the first of two. The second
                will hopefully follow on its heels in the next couple of
                days.<br class=3D"">
                <br class=3D"">
                Diffs between this first rev (-16) and -15:<br class=3D"">=

                <br class=3D"">
                1.&nbsp; switches round two fairly large sections, so =
I've
                deferred other changes to a second rev so the diffs
                won't be masked by the switch round of sections.<br =
class=3D"">
                Suggested by Ilpo to match the order in which the tests
                in these sections will be executed:<br class=3D"">
                * Test for mangling the IP-ECN field (now 3.2.2.3),<br =
class=3D"">
                * Then test for zeroing the ACE field (now 3.2.2.4).<br =
class=3D"">
                <br class=3D"">
                2. Ilpo suggested some clarifications in =
"3.2.3.2.5.&nbsp;
                Consistency between AccECN Feedback Fields", which is
                about the receiver of feedback ensuring consistency
                between the mandatory 3-bit ACE field and the optional
                24-bit counters. In brief (paraphrasing) it previously
                only said "MUST consider both fields", when it is now
                clearer that it actually meant "MUST reconcile both
                fields", so that there is always a consistent baseline
                for subsequent ACKs.<br class=3D"">
                <br class=3D"">
                3. A minor point is added in an appendix about the
                details that the pseudocode doesn't cover.<br class=3D"">
                <br class=3D"">
                <br class=3D"">
                <br class=3D"">
                Bob<br class=3D"">
                <br class=3D"">
                PS. #2 &amp; #3 were added to the XML ages ago (Jul
                '21), so you will have seen them in the HTML. However,
                prob due to my clumsiness, the posted TXT didn't include
                them whereas the posted XML did (ironic for a section
                about consistency). In turn, inconsistency was only
                possible because I am having to manually post the TXT
                for this draft, due to an unresolved issue with v3 RFC
                XML tables.<br class=3D"">
                <br class=3D"">
                <br class=3D"">
                On 03/02/2022 14:43, <a =
href=3D"mailto:internet-drafts@ietf.org" class=3D"moz-txt-link-freetext" =
moz-do-not-send=3D"true">internet-drafts@ietf.org</a>
                wrote:<br class=3D"">
                <blockquote type=3D"cite" class=3D"">A New =
Internet-Draft is
                  available from the on-line Internet-Drafts
                  directories.<br class=3D"">
                  This draft is a work item of the TCP Maintenance and
                  Minor Extensions WG of the IETF.<br class=3D"">
                  <br class=3D"">
                  &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Title =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: More =
Accurate ECN Feedback
                  in TCP<br class=3D"">
                  =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Authors =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: Bob Briscoe<br =
class=3D"">
                  =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;Mirja K=C3=BChlewind<br class=3D"">
                  =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;Richard Scheffenegger<br class=3D"">
                  <span class=3D"Apple-tab-span" =
style=3D"white-space:pre">	</span>Filename
                  &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: =
draft-ietf-tcpm-accurate-ecn-16.txt<br class=3D"">
                  <span class=3D"Apple-tab-span" =
style=3D"white-space:pre">	</span>Pages
                  =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: 60<br =
class=3D"">
                  <span class=3D"Apple-tab-span" =
style=3D"white-space:pre">	</span>Date
                  =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: =
2022-02-03<br class=3D"">
                  <br class=3D"">
                  Abstract:<br class=3D"">
                  &nbsp;&nbsp;&nbsp;Explicit Congestion Notification =
(ECN) is a
                  mechanism where network<br class=3D"">
                  &nbsp;&nbsp;&nbsp;nodes can mark IP packets instead of =
dropping them
                  to indicate<br class=3D"">
                  &nbsp;&nbsp;&nbsp;incipient congestion to the =
end-points. &nbsp;Receivers
                  with an ECN-<br class=3D"">
                  &nbsp;&nbsp;&nbsp;capable transport protocol feed back =
this
                  information to the sender.<br class=3D"">
                  &nbsp;&nbsp;&nbsp;ECN was originally specified for TCP =
in such a way
                  that only one<br class=3D"">
                  &nbsp;&nbsp;&nbsp;feedback signal can be transmitted =
per Round-Trip
                  Time (RTT). &nbsp;Recent<br class=3D"">
                  &nbsp;&nbsp;&nbsp;new TCP mechanisms like Congestion =
Exposure
                  (ConEx), Data Center TCP<br class=3D"">
                  &nbsp;&nbsp;&nbsp;(DCTCP) or Low Latency Low Loss =
Scalable Throughput
                  (L4S) need more<br class=3D"">
                  &nbsp;&nbsp;&nbsp;accurate ECN feedback information =
whenever more
                  than one marking is<br class=3D"">
                  &nbsp;&nbsp;&nbsp;received in one RTT. &nbsp;This =
document specifies a
                  scheme to provide<br class=3D"">
                  &nbsp;&nbsp;&nbsp;more than one feedback signal per =
RTT in the TCP
                  header. &nbsp;Given TCP<br class=3D"">
                  &nbsp;&nbsp;&nbsp;header space is scarce, it allocates =
a reserved
                  header bit previously<br class=3D"">
                  &nbsp;&nbsp;&nbsp;assigned to the ECN-Nonce. &nbsp;It =
also overloads the
                  two existing ECN<br class=3D"">
                  &nbsp;&nbsp;&nbsp;flags in the TCP header. &nbsp;The =
resulting extra space
                  is exploited to<br class=3D"">
                  &nbsp;&nbsp;&nbsp;feed back the IP-ECN field received =
during the
                  3-way handshake as<br class=3D"">
                  &nbsp;&nbsp;&nbsp;well. &nbsp;Supplementary feedback =
information can
                  optionally be provided<br class=3D"">
                  &nbsp;&nbsp;&nbsp;in a new TCP option, which is never =
used on the TCP
                  SYN. &nbsp;The<br class=3D"">
                  &nbsp;&nbsp;&nbsp;document also specifies the =
treatment of this
                  updated TCP wire<br class=3D"">
                  &nbsp;&nbsp;&nbsp;protocol by middleboxes.<br =
class=3D"">
                  <br class=3D"">
                  <br class=3D"">
                  The IETF datatracker status page for this draft is:<br =
class=3D"">
                  <a =
href=3D"https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/" =
class=3D"moz-txt-link-freetext" =
moz-do-not-send=3D"true">https://datatracker.ietf.org/doc/draft-ietf-tcpm-=
accurate-ecn/</a><br class=3D"">
                  <br class=3D"">
                  There is also an htmlized version available at:<br =
class=3D"">
<a class=3D"moz-txt-link-freetext" =
href=3D"https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn=
-16">https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16=
</a><br class=3D"">
                  <br class=3D"">
                  A diff from the previous version is available at:<br =
class=3D"">
<a class=3D"moz-txt-link-freetext" =
href=3D"https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-accurate-ecn-1=
6">https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-accurate-ecn-16</a>=
<br class=3D"">
                  <br class=3D"">
                  <br class=3D"">
                  Internet-Drafts are also available by rsync at
                  <a href=3D"http://rsync.ietf.org" =
class=3D"">rsync.ietf.org</a>::internet-drafts<br class=3D"">
                  <br class=3D"">
                  <br class=3D"">
                  _______________________________________________<br =
class=3D"">
                  tcpm mailing list<br class=3D"">
                  <a class=3D"moz-txt-link-abbreviated" =
href=3D"mailto:tcpm@ietf.org">tcpm@ietf.org</a><br class=3D"">
                  <a class=3D"moz-txt-link-freetext" =
href=3D"https://www.ietf.org/mailman/listinfo/tcpm">https://www.ietf.org/m=
ailman/listinfo/tcpm</a><br class=3D"">
                </blockquote>
                <br class=3D"">
                -- <br class=3D"">
________________________________________________________________<br =
class=3D"">
                Bob Briscoe =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;<a href=3D"http://bobbriscoe.net/" =
class=3D"moz-txt-link-freetext" =
moz-do-not-send=3D"true">http://bobbriscoe.net/</a><br class=3D"">
                <br class=3D"">
                _______________________________________________<br =
class=3D"">
                tcpm mailing list<br class=3D"">
                <a href=3D"mailto:tcpm@ietf.org" =
class=3D"moz-txt-link-freetext" =
moz-do-not-send=3D"true">tcpm@ietf.org</a><br class=3D"">
                <a class=3D"moz-txt-link-freetext" =
href=3D"https://www.ietf.org/mailman/listinfo/tcpm">https://www.ietf.org/m=
ailman/listinfo/tcpm</a><br class=3D"">
              </div>
            </div>
          </blockquote>
        </div>
        <br class=3D"">
      </div>
    </blockquote>
    <br class=3D"">
    <pre class=3D"moz-signature" cols=3D"72">--=20
________________________________________________________________
Bob Briscoe                               <a =
class=3D"moz-txt-link-freetext" =
href=3D"http://bobbriscoe.net/">http://bobbriscoe.net/</a></pre>
  </div>

</div></blockquote></div><br class=3D""></div></body></html>=

--Apple-Mail=_2310DDF5-C279-419F-90E2-6C486952D0E4--

--Apple-Mail=_4844375E-DDC1-4B78-B59B-E77932053680
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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=
--Apple-Mail=_4844375E-DDC1-4B78-B59B-E77932053680--


From nobody Sat Feb  5 04:02:16 2022
Return-Path: <rs.ietf@gmx.at>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EA71D3A18D2; Sat,  5 Feb 2022 04:02:13 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.612
X-Spam-Level: 
X-Spam-Status: No, score=-2.612 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, FREEMAIL_FROM=0.001, NICE_REPLY_A=-0.714, RCVD_IN_DNSWL_BLOCKED=0.001, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=gmx.net
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5eZVCKIPoX2l; Sat,  5 Feb 2022 04:02:09 -0800 (PST)
Received: from mout.gmx.net (mout.gmx.net [212.227.15.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 35DB53A18CA; Sat,  5 Feb 2022 04:02:09 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=gmx.net; s=badeba3b8450; t=1644062481; bh=tldLGdS2h58KMciA3AH6zsnITrvuhsWNtLm4zl0MgMs=; h=X-UI-Sender-Class:Date:Subject:To:Cc:References:From:In-Reply-To; b=E2RJfgn7lR3bfHE7/BW/lHnsV5unSLUd/fmkfN5GN49DAtpGqCns58QKNrDFNhTWO LQcb3Sy6AVKx53olNoGp0w3pUnbYhUiJb6R33qJSfQcAyb3Rlw3YmCku/ol6cTVM59 hI4dZ6Y9SSGSx+S597DVXwjBv/owu8A/FbdUQ66Y=
X-UI-Sender-Class: 01bb95c1-4bf8-414a-932a-4f6e2808ef9c
Received: from [192.168.233.112] ([185.236.167.136]) by mail.gmx.net (mrgmx004 [212.227.17.190]) with ESMTPSA (Nemesis) id 1N4hzZ-1mFNHL3DaJ-011kXa; Sat, 05 Feb 2022 13:01:20 +0100
Message-ID: <1cbe3dbe-241b-3f24-096f-f1a2ffe17d98@gmx.at>
Date: Sat, 5 Feb 2022 13:01:19 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Thunderbird/91.5.0
To: Vidhi Goel <vidhi_goel=40apple.com@dmarc.ietf.org>, Bob Briscoe <ietf@bobbriscoe.net>
Cc: Richard Scheffenegger <rscheff@gmx.at>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, Mirja Kuehlewind <ietf@kuehlewind.net>
References: <164389942954.13556.7754569279919863072@ietfa.amsl.com> <c2c5bdf9-6b34-fe19-3296-5f1cd831ed8e@bobbriscoe.net> <2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com> <e57355ae-5a9d-0a14-9478-29453d25622c@bobbriscoe.net> <BB23E172-C995-41FE-A47E-E02D5586B67D@apple.com>
From: "Scheffenegger, Richard" <rs.ietf@gmx.at>
In-Reply-To: <BB23E172-C995-41FE-A47E-E02D5586B67D@apple.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: quoted-printable
X-Provags-ID: V03:K1:XFrxiI5ZqWFkAo7fMY94iRM/yffkHzLWkCsFCXSdoY/0kedvBLv /e6afGdBbegJLC9QMHTWvc/0dSBgMI49d5ZkCaoUoMRcPRIIxjsqvWg58dmUSbDujTt+evp SzYyDuaN9IIafZM9UnfJor9HO33jyx1lSAPqxPTCAKRElYHa5HorNmQsvRjwL1rQzD1CkkB OTdqMDWSInpJiPmmEOvPA==
X-UI-Out-Filterresults: notjunk:1;V03:K0:aVP2p13wsnY=:IfawQZGkf76uJWabyAcgqb x3cd7LCD0lmN7B2Wl2WeC3mmTqRkU3DhWULZpu8/Aa3zcZBTNBR3nsKqWKlQgt6eYHnS8OYB+ mVAhvdPLnvLwoxv+SbiZVLVaVfFSUib7qxf2CMHSX57JOkkgb7gLiZzhC768JMF9Zpof5y/Uo E0ak4oPUiZpojTWVLCoT6PHl2Pp09y6xEC6LVcr80s1a/mbMq/KbIkFf5t1daMDi+cH8cyAMs 9P7Jil/O5suvqrbjwq8C2Utfv9MphcZEc85Dl/htyFGmr8Dypvx0I2FWQV4KLEZzrc9YErlvD C1EFQVDf0CfZvy+kOzm0kuFnieaeyLbp9JthATyQbcOV+bPo+dvNHDNYqdNaliVN/7isSqu1E SOUzlI7E9xskqiWII5eMrRx9v4N5pj2W8D2cALalOCwiU/3GXAbH7Ab2ysc0agR/pRB1cIbu4 0WdI8cVqHBDgcQppYfkJ2+JxbUZIjMAVdUxlQuaOt0VLU7NcHrUFzz/BYMafZ2AWFtHWtead+ Dhl7UpOY/7pkIeJGdF4UjkCuH/RSTKUCdQWFAP4c9lYlI5PjfNoGl0k4No+thkCcZF8DlIdC/ jjY3Wz2hY3NLKH8P1IcFopi2z7Fa5CokWp3SnxMojLG1nQ9O3qAwnOjWX86sLuEDNmr51Qxji dMMbbNxTy/J7ff/7u6Ck6Zx5BvBfGloslbKI+SUIcWTTWAK/6tNuaB3sHma8LFSSt8+LhYaW1 gCxC0pNPRt6HQOxZyi1oD6wqgz4pNxHdPUOPJjek7B9RVl25AoE44YaHqEPEFYHBBW1qnWuRe LzUv8FaaKop3pJmF5vqeWYSRAMfgXAly9Wmq3dd+yVcQVPRwGV1VguBdySSPPJOFKfyi1/djI Y/Tq0raK2rqmeiUBZb3H8VacquRJi07dhsm6pLgjk/DdN+EMlg+E9hM/0p9SmQLTDSsCULyDy 2iTy0PSTUHaH1HrnB0oBxCp++4Lk2pbWYDd3F4yso5Qbthv0SaTDt5q8tmf6n2Ztr//9yLSZN RLXa/1xViQ0CWtWyvwk8aPIBlZFfB2mTS120W0+48r5Kv5hOWrqHAhxahHxFJTkRCFk7v7gOn 5HfHMbrOARzvYA=
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/lIWhGhE3JEF9OVphHn7ibQ1G90c>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-accurate-ecn-16.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 05 Feb 2022 12:02:14 -0000

Vidhi,

Thanks for reading the draft very closely!

Well, in the non-ECT SYN case, where the SYNACK indicates an ECT (ect0,
ect1, ce) is the obvious - and fully semantically correct - use case of
that possible use.

For the set of three codepoints (ECT0, ECT1, CE) the only other possible
valid detection scenario would be if a SYN+CE is sent (deliberately,
althought one can argue that a sender should not originate CE-marked
packets), but a SYNACK without CE is returned (full or partial
bleaching). At least under the current RFC3168 definition of IP ECN
codepoints.

allowed transitions:
nonECT -> nonECT

ECT0 -> ECT0
ECT0 -> ECT1
ECT0 -> CE

ECT1 -> ECT1
ECT1 -> ECT0 (possibly)
ECT1 -> CE

CE -> CE


Thinking about this aspect more, I am now unsure, if the higher level
functions, such as detection of IP ECN codepoint mangling, or validating
across multiple RTTs, should be in this document - or not rather be in a
separate draft, in order to facilitate independent implementation and
improvement of mechanisms in that space.

However, the above table of allowed IP ECN transistions may be updated
independently of AccECN - and performing extensive checks as part of a
signal protocol ossifies this IMHO.

As a middle ground, perhaps having these descriptions in a non-normative
section, clearly stating these to be examples of what an impemented
could choose to do?

Richard

Am 05.02.2022 um 02:14 schrieb Vidhi Goel:
> I like your response for all the points. I can review the diff for the
> proposed changes, if you=E2=80=99d like before committing them to the dr=
aft.
>
> Sorry, I want to add a new comment for this text,
>
> If a TCP client has set the SYN to Not-ECT, but
>     receives feedback that the IP-ECN field on the SYN arrived with a
>     different codepoint, it can detect such middlebox interference and
>     send Not-ECT for the rest of the connection.
>
>
> This statement holds good for Not-ECT but doesn=E2=80=99t if lets say th=
e SYN
> was ECT0 or ECT1=C2=A0as those could be marked CE and still be valid.=C2=
=A0Should
> we add a statement for ECT marked SYN as well?
>
> Thanks,
> Vidhi
>
>> On Feb 4, 2022, at 5:44 AM, Bob Briscoe <ietf@bobbriscoe.net
>> <mailto:ietf@bobbriscoe.net>> wrote:
>>
>> Vidhi,
>>
>> I've increased the table of contents to tocdepth =3D "4" which might
>> help find the relevant sections better, because a lot of the meat in
>> this draft is under sections 3.2.2 (ACE) and 3.2.3 (Option).
>>
>> I've responded below, but I'd prefer it if you would say these things
>> on the list, so we have a better justification for changing the draft.
>> Reviewing a draft that has just be re-posted shouldn't imply anything
>> about whether Apple is implementing the draft or not.
>>
>> (BTW, regarding the proposed change to the initial value of r.e1b, I
>> was going to send that=C2=A0 to the list once co-authors agreed to it. =
But
>> it would be nice if some of the edits were not self-generated.)
>>
>> See [BB] inline for responses (which I'll repeat if you send to the lis=
t).
>>
>> On 04/02/2022 07:37, Vidhi Goel wrote:
>>> Hello Bob and authors,
>>>
>>> I have read the draft a few times and will send my extensive review
>>> at a later point. But right now, I want to ask some critical things
>>> that an implementation could benefit from.
>>>
>>> The draft says,
>>> /If a TCP client has set the SYN to Not-ECT, but/
>>> /=C2=A0=C2=A0 receives feedback that the IP-ECN field on the SYN arriv=
ed with a/
>>> /=C2=A0=C2=A0 different codepoint, it can detect such middlebox interf=
erence and/
>>> /=C2=A0=C2=A0 send Not-ECT for the rest of the connection/
>>>
>>> 1. On the forward path from client to server, the client will revert
>>> to Not-ECT when it sees for example, that it sent a Non-ECT SYN but
>>> received ACE encoding other than 0 1 0.=C2=A0What does the client do i=
n
>>> the last ACK of the 3WHS -
>>> =C2=A0a. does it stay in AccECN mode and still send AccECN encoding ba=
sed
>>> on the IP=C2=A0code point of SYN-ACK (Table 4)? This would mean that
>>> client won=E2=80=99t participate in ECN on the sender half of the conn=
ection
>>> and only=C2=A0provide AccECN feedback to the server as a receiver.
>>> =C2=A0b. =C2=A0does it disable AccECN mode and set ACE=3D 0 0 0 so tha=
t a server
>>> in AccECN mode can disable ECN based on Table4?
>>>
>>> While writing this, I realized that the intention is=C2=A0probably a.
>>> Could you confirm? Also, when the sender sets Not-ECT in its data
>>> packets, it should also disable acting upon any ACE feedback as we
>>> could still receive false ACE feedback from the server if=C2=A0the
>>> network, lets say, changed 00 (not ECT) to 01(ECT1). If you agree, we
>>> should add some text around this. Based on the current text, the
>>> sender will always respond to the ACE feedback even if it sends Not-EC=
T.
>>>
>>> TBH, this is a complicated scenario, where sender said to the network
>>> - I don=E2=80=99t trust you so I can=E2=80=99t use ECN. Feel=C2=A0free=
 to drop my packets.
>>> And the network mangles the IP to ECT1=C2=A0and then set CE (when
>>> congested)=C2=A0which would be feedback=E2=80=99ed from the receiver. =
Now, the
>>> packet wasn=E2=80=99t dropped which it should have been. So, is it bet=
ter to
>>> just ignore this feedback because sender doesn=E2=80=99t trust the net=
work or
>>> just act on it and reduce cwnd in order to reduce congestion in the
>>> network somewhere.
>>
>> [BB] Your points are all good ones.
>> I'll address this last complex scenario first, because it has
>> implications for the earlier questions. First I'll define some
>> terminology:
>>
>> Simple mangling scenario:
>>
>>   * some network function, e.g. broken Diffserv ToS-byte remarking,
>>     illegally remarks a Not-ECT=C2=A0 SYN to CE.
>>
>> Your complex mangling scenario, repeated here:
>>
>>   * some network function, e.g. broken Diffserv ToS-byte remarking,
>>     illegally remarks a Not-ECT SYN to ECT0 or ECT1,
>>   * then congestion at a subsequent bottleneck is marking some packets =
CE
>>
>> Then potentially the client could tailor its behaviour after sending a
>> Not-ECT SYN, If the SYN/ACK feedback is
>>
>>  1. ECT: disable sending ECT, but continue responding to ECN feedback
>>  2. CE, disable sending ECT, and disable response to ECN feedback
>>
>> However, this is uncertain, because CE feedback on the SYN/ACK could
>> indicate either the simple or the complex mangling scenario.
>>
>> A simpler alternative would be to always continue responding to ECN
>> feedback. Rationale:
>>
>>   * Whether case #1 or #2, assume that mangling of the SYN might have
>>     been Not-ECT to ECT, even if the feedback off the SYN is CE.
>>   * Then as the connection progresses, if /all/ feedback is CE,
>>     there's already a recommendation to fall-back to disabling
>>     congestion response.
>>
>> If we do this, I think we ought to say "SHOULD continue to respond to
>> ECN feedback", not "MUST".
>>
>> And we'll need to put this all to the WG.
>>
>> 1.a) Now back to the beginning of your point. The text you quote is
>> from =C2=A72.5 which is in the non-normative "Overview and Rationale"
>> section (=C2=A72).
>> You really need the normative text from:
>>
>>
>>           3.2.2.3.
>>           <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn=
-16.html#section-3.2.2.3>Testing
>>           for Mangling of the IP/ECN Field
>>           <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn=
-16.html#name-testing-for-mangling-of-the>
>>
>>
>> where it says (end of 1st para):
>>
>>     " ...for the remainder of the connection the client MUST NOT send
>>     ECN-capable packets, but it MUST continue to feed back any ECN
>>     markings on arriving packets."
>>
>> So, you're right, the answer is (a).
>> It can't be (b), because, once in AccECN mode, 000 on the ACE field
>> just becomes a counter value, and no longer negotiates the feedback mod=
e.
>>
>> 1.b) As per the above, let's conservatively assume complex mangling.
>> So, at the end of the first para quoted above I suggest we add:
>>
>>     " ...for the remainder of the connection the client MUST NOT send
>>     ECN-capable packets *but**it**SHOULD continue to respond to ECN
>>     feedback even though **it **is no longer sending ECN-capable
>>     packets (see reasoning below)**. T**he client**MUST remain in the
>>     same feedback mode and***it MUST continue to feed back any ECN
>>     markings on arriving packets."
>>
>> I'll do the same for next para about the server.
>> I'll work out some text for the reasoning, but I won't give it here.
>>
>> There are two other cases where it disables sending ECT, which don't
>> say whether it continues to respond to congestion:
>>
>>   * "Testing for Mangling" section, penultimate para,=C2=A0 where it's
>>     receiving solid CE:
>>     "Once a Data Sender has entered AccECN mode it SHOULD check
>>     whether all feedback received for the first three or four rounds
>>     indicated that every packet it sent was CE-marked. If so, for the
>>     remainder of the connection, the Data Sender SHOULD NOT send
>>     ECN-capable packets*and it SHOULD NOT respond to ECN feedback*,
>>     but *it MUST remain in the same feedback mode and *it MUST
>>     continue to feed back any ECN markings on arriving packets*(in its
>>     role as Data Receiver)*."
>>   * Next section "Zeroing of the ACE Field", 2nd para:
>>     "If the value of this ACE field is zero (0b000), the Data Sender
>>     disables sending ECN-capable packets for the remainder of the
>>     half-connection by setting the IP/ECN field in all subsequent
>>     packets to Not-ECT, *but**it**SHOULD continue to respond to ECN
>>     feedback even though **it **is no longer sending ECN-capable
>>     packets.* *It MUST also remain in the same feedback mode and it
>>     MUST continue to feed back any ECN markings on arriving packets
>>     (in its role as Data Receiver).*"
>>
>> I'll also add reasoning (in the zeroing section, it already says that
>> ACE=3D0b000 is not an unambiguous indication of mangling).
>>
>> And this prompts me to edit the bullet in an earlier section about the
>> obligation of a sender to respond to congestion feedback:
>>
>>
>>         3.1.5.
>>         <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-1=
6.html#section-3.1.5>Implications
>>         of AccECN Mode
>>         <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-1=
6.html#name-implications-of-accecn-mode>
>>
>>
>>     "It is still obliged to respond appropriately to AccECN feedback
>>     that indicates there were ECN marks on *ECN-capable *packets it
>>     had previously sent."
>>
>> And I'll add a bullet:
>>
>>     "*If the sender chooses not to send ECN-capable packets (e.g.
>>     because path traversal of the ECN field is suspect), it can ignore
>>     any **ECN **feedback about those packets if it is certain that it
>>     cannot be valid (see Section 3.2.2, which gives normative
>>     requirements for certain specific cases).*"
>>
>> I've said 'can' rather than 'MUST' because it's hard to cover all
>> cases, e.g. single packets without ECT when it's not clear whether the
>> feedback covered another packet that was ECT.
>>
>> How does all this sound?
>>
>>> 2. On the reverse path=C2=A0from server to client, if a server sends a
>>> Not-ECT SYN-ACK and receives ACE handshake encoding on last ACK=C2=A0o=
ther
>>> than 0 1 0, there is no text like above that says server should send
>>> Not-ECT for the rest of the connection (or=C2=A0at least I=C2=A0didn=
=E2=80=99t find
>>> it). I think the server should also do same as client as=C2=A0the two
>>> paths could be different. One could make it more complicated by
>>> saying, if both client and server decide to not use ECN on their
>>> corresponding sender half, then ECN should be disabled but let=E2=80=
=99s talk
>>> about that later.
>>
>> [BB] It's in the second para of:
>>
>>
>>           3.2.2.3.
>>           <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn=
-16.html#section-3.2.2.3>Testing
>>           for Mangling of the IP/ECN Field
>>           <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn=
-16.html#name-testing-for-mangling-of-the>
>>
>>
>>> 3. In general, not setting the ECT (ECT0 or ECT1)=C2=A0code point on a=
n
>>> outgoing packet is different from supporting AccECN right as in the
>>> host can still provide AccECN feedback on the=C2=A0receive path.
>>
>> [BB] Yup.
>>
>> Do you think any further explanation is needed in the above sections?
>>
>>
>> Bob
>>
>>> =E2=80=A6.
>>>
>>> To be continued if more questions come to my mind.
>>>
>>>
>>> Thanks,
>>> Vidhi
>>>
>>>> On Feb 3, 2022, at 7:24 AM, Bob Briscoe <ietf@bobbriscoe.net> wrote:
>>>>
>>>> tcpm folks,
>>>>
>>>> This rev to accurate-ecn is the first of two. The second will
>>>> hopefully follow on its heels in the next couple of days.
>>>>
>>>> Diffs between this first rev (-16) and -15:
>>>>
>>>> 1.=C2=A0 switches round two fairly large sections, so I've deferred o=
ther
>>>> changes to a second rev so the diffs won't be masked by the switch
>>>> round of sections.
>>>> Suggested by Ilpo to match the order in which the tests in these
>>>> sections will be executed:
>>>> * Test for mangling the IP-ECN field (now 3.2.2.3),
>>>> * Then test for zeroing the ACE field (now 3.2.2.4).
>>>>
>>>> 2. Ilpo suggested some clarifications in "3.2.3.2.5. Consistency
>>>> between AccECN Feedback Fields", which is about the receiver of
>>>> feedback ensuring consistency between the mandatory 3-bit ACE field
>>>> and the optional 24-bit counters. In brief (paraphrasing) it
>>>> previously only said "MUST consider both fields", when it is now
>>>> clearer that it actually meant "MUST reconcile both fields", so that
>>>> there is always a consistent baseline for subsequent ACKs.
>>>>
>>>> 3. A minor point is added in an appendix about the details that the
>>>> pseudocode doesn't cover.
>>>>
>>>>
>>>>
>>>> Bob
>>>>
>>>> PS. #2 & #3 were added to the XML ages ago (Jul '21), so you will
>>>> have seen them in the HTML. However, prob due to my clumsiness, the
>>>> posted TXT didn't include them whereas the posted XML did (ironic
>>>> for a section about consistency). In turn, inconsistency was only
>>>> possible because I am having to manually post the TXT for this
>>>> draft, due to an unresolved issue with v3 RFC XML tables.
>>>>
>>>>
>>>> On 03/02/2022 14:43, 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 TCP Maintenance and Minor
>>>>> Extensions WG of the IETF.
>>>>>
>>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Title =C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0: More Accurate ECN Feedba=
ck in TCP
>>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Authors =C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0: Bob Briscoe
>>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0Mirja K=C3=BChlewind
>>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0Richard Scheffenegger
>>>>> Filename =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0: draft-ietf-tcpm=
-accurate-ecn-16.txt
>>>>> Pages =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0: =
60
>>>>> Date =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0: 2022-02-03
>>>>>
>>>>> Abstract:
>>>>> =C2=A0=C2=A0=C2=A0Explicit Congestion Notification (ECN) is a mechan=
ism where network
>>>>> =C2=A0=C2=A0=C2=A0nodes can mark IP packets instead of dropping them=
 to indicate
>>>>> =C2=A0=C2=A0=C2=A0incipient congestion to the end-points. =C2=A0Rece=
ivers with an ECN-
>>>>> =C2=A0=C2=A0=C2=A0capable transport protocol feed back this informat=
ion to the sender.
>>>>> =C2=A0=C2=A0=C2=A0ECN was originally specified for TCP in such a way=
 that only one
>>>>> =C2=A0=C2=A0=C2=A0feedback signal can be transmitted per Round-Trip =
Time (RTT).
>>>>> =C2=A0Recent
>>>>> =C2=A0=C2=A0=C2=A0new TCP mechanisms like Congestion Exposure (ConEx=
), Data Center TCP
>>>>> =C2=A0=C2=A0=C2=A0(DCTCP) or Low Latency Low Loss Scalable Throughpu=
t (L4S) need more
>>>>> =C2=A0=C2=A0=C2=A0accurate ECN feedback information whenever more th=
an one marking is
>>>>> =C2=A0=C2=A0=C2=A0received in one RTT. =C2=A0This document specifies=
 a scheme to provide
>>>>> =C2=A0=C2=A0=C2=A0more than one feedback signal per RTT in the TCP h=
eader. =C2=A0Given TCP
>>>>> =C2=A0=C2=A0=C2=A0header space is scarce, it allocates a reserved he=
ader bit
>>>>> previously
>>>>> =C2=A0=C2=A0=C2=A0assigned to the ECN-Nonce. =C2=A0It also overloads=
 the two existing ECN
>>>>> =C2=A0=C2=A0=C2=A0flags in the TCP header. =C2=A0The resulting extra=
 space is exploited to
>>>>> =C2=A0=C2=A0=C2=A0feed back the IP-ECN field received during the 3-w=
ay handshake as
>>>>> =C2=A0=C2=A0=C2=A0well. =C2=A0Supplementary feedback information can=
 optionally be provided
>>>>> =C2=A0=C2=A0=C2=A0in a new TCP option, which is never used on the TC=
P SYN. =C2=A0The
>>>>> =C2=A0=C2=A0=C2=A0document also specifies the treatment of this upda=
ted TCP wire
>>>>> =C2=A0=C2=A0=C2=A0protocol by middleboxes.
>>>>>
>>>>>
>>>>> The IETF datatracker status page for this draft is:
>>>>> https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/
>>>>>
>>>>> There is also an htmlized version available at:
>>>>> https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-1=
6
>>>>>
>>>>> A diff from the previous version is available at:
>>>>> https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-accurate-ecn-16
>>>>>
>>>>>
>>>>> Internet-Drafts are also available by rsync at rsync.ietf.org
>>>>> <http://rsync.ietf.org>::internet-drafts
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> tcpm mailing list
>>>>> tcpm@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>
>>>> --
>>>> ________________________________________________________________
>>>> Bob Briscoe http://bobbriscoe.net/
>>>>
>>>> _______________________________________________
>>>> tcpm mailing list
>>>> tcpm@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>
>>
>> --
>> ________________________________________________________________
>> Bob Briscoehttp://bobbriscoe.net/
>
>
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm


From nobody Sat Feb  5 14:59:45 2022
Return-Path: <vidhi_goel@apple.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CF2B43A1E83 for <tcpm@ietfa.amsl.com>; Sat,  5 Feb 2022 14:59:42 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.675
X-Spam-Level: 
X-Spam-Status: No, score=-2.675 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.576, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=apple.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id urNfLm9v1o03 for <tcpm@ietfa.amsl.com>; Sat,  5 Feb 2022 14:59:37 -0800 (PST)
Received: from rn-mailsvcp-ppex-lapp14.apple.com (rn-mailsvcp-ppex-lapp14.rno.apple.com [17.179.253.33]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 69B683A1E82 for <tcpm@ietf.org>; Sat,  5 Feb 2022 14:59:37 -0800 (PST)
Received: from pps.filterd (rn-mailsvcp-ppex-lapp14.rno.apple.com [127.0.0.1]) by rn-mailsvcp-ppex-lapp14.rno.apple.com (8.16.1.2/8.16.1.2) with SMTP id 215MpKTL005793; Sat, 5 Feb 2022 14:59:33 -0800
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=apple.com; h=content-type : content-transfer-encoding : from : mime-version : subject : date : message-id : references : cc : in-reply-to : to; s=20180706; bh=vH9owh7pC0d4sY3hC9qrbjz6jn0qJNGu8rysPOZHFwk=; b=muNB0khFm5yYU3hDMShzzPWVjYSeSm+XjbkIgvYckPw21XmQITuDsOEjy9fzOahyDLh1 TNpIWvnsEhQct2LPPUzYKwyB9lwp7O4v4z3dva6Bcpidr/6xD/8Jpt2sng8I5w2d+OH4 b9+IHys9HhTvMXYtrpO4MxsUOcQKku7maoBQq7lxPo0MEu6y8UIpeqR5nHodCMdo8qKB grdopO8RuNrHfDp/4pJXdohHY2l1VtVCxZHyPIDsnF9gIFTpTp9AAJHSHAdy7kL9ECaL N1YfVa0n1k7wSSV7/aQLpHzbKddSx6G+9VcqKpLBIivks/EFsIi2UzZYc7CaL1sXbLdg sQ== 
Received: from ma-mailsvcp-mta-lapp04.corp.apple.com (ma-mailsvcp-mta-lapp04.corp.apple.com [10.226.18.136]) by rn-mailsvcp-ppex-lapp14.rno.apple.com with ESMTP id 3e1q0dntsx-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NO); Sat, 05 Feb 2022 14:59:33 -0800
Received: from ma-mailsvcp-mmp-lapp01.apple.com (ma-mailsvcp-mmp-lapp01.apple.com [17.32.222.14]) by ma-mailsvcp-mta-lapp04.corp.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPS id <0R6U002CFT78C000@ma-mailsvcp-mta-lapp04.corp.apple.com>; Sat, 05 Feb 2022 14:59:32 -0800 (PST)
Received: from process_milters-daemon.ma-mailsvcp-mmp-lapp01.apple.com by ma-mailsvcp-mmp-lapp01.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) id <0R6U00900SI9O900@ma-mailsvcp-mmp-lapp01.apple.com>; Sat, 05 Feb 2022 14:59:32 -0800 (PST)
X-Va-A: 
X-Va-T-CD: 81ca60fce39c2560b6c4a7e5841f9b8f
X-Va-E-CD: d27b0bf1ec5e225c315eb75773194505
X-Va-R-CD: 7f6ba7ce084ea0aad2ff4f136dfa7fc9
X-Va-CD: 0
X-Va-ID: 9a0b082f-20cb-4682-929d-fd5e1b1c4790
X-V-A: 
X-V-T-CD: 81ca60fce39c2560b6c4a7e5841f9b8f
X-V-E-CD: d27b0bf1ec5e225c315eb75773194505
X-V-R-CD: 7f6ba7ce084ea0aad2ff4f136dfa7fc9
X-V-CD: 0
X-V-ID: 9b210405-7100-4f09-8b7c-6694241a158e
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-05_12:2022-02-03, 2022-02-05 signatures=0
Received: from smtpclient.apple (unknown [17.10.30.3]) by ma-mailsvcp-mmp-lapp01.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPSA id <0R6U00P7FT77LI00@ma-mailsvcp-mmp-lapp01.apple.com>; Sat, 05 Feb 2022 14:59:32 -0800 (PST)
Content-type: text/plain; charset=utf-8
Content-transfer-encoding: quoted-printable
From: Vidhi Goel <vidhi_goel@apple.com>
MIME-version: 1.0 (1.0)
Date: Sat, 05 Feb 2022 14:59:31 -0800
Message-id: <BB2835FB-58E6-409A-BEBE-273D874D0B55@apple.com>
References: <1cbe3dbe-241b-3f24-096f-f1a2ffe17d98@gmx.at>
Cc: Richard Scheffenegger <rscheff@gmx.at>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, Mirja Kuehlewind <ietf@kuehlewind.net>
In-reply-to: <1cbe3dbe-241b-3f24-096f-f1a2ffe17d98@gmx.at>
To: "Scheffenegger, Richard" <rs.ietf@gmx.at>
X-Mailer: iPhone Mail (19B5060d)
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-05_12:2022-02-03, 2022-02-05 signatures=0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/hfLqWzSraTkgsmHn0eoJU2QyH7s>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-accurate-ecn-16.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 05 Feb 2022 22:59:43 -0000

> As a middle ground, perhaps having these descriptions in a non-normative
> section, clearly stating these to be examples of what an impemented
> could choose to do?

This draft already covers some of the path mangling detection so it would be=
 good to cover all of it. I agree with your middle ground suggestion.

Vidhi

> On Feb 5, 2022, at 4:02 AM, Scheffenegger, Richard <rs.ietf@gmx.at> wrote:=

>=20
> =EF=BB=BFVidhi,
>=20
> Thanks for reading the draft very closely!
>=20
> Well, in the non-ECT SYN case, where the SYNACK indicates an ECT (ect0,
> ect1, ce) is the obvious - and fully semantically correct - use case of
> that possible use.
>=20
> For the set of three codepoints (ECT0, ECT1, CE) the only other possible
> valid detection scenario would be if a SYN+CE is sent (deliberately,
> althought one can argue that a sender should not originate CE-marked
> packets), but a SYNACK without CE is returned (full or partial
> bleaching). At least under the current RFC3168 definition of IP ECN
> codepoints.
>=20
> allowed transitions:
> nonECT -> nonECT
>=20
> ECT0 -> ECT0
> ECT0 -> ECT1
> ECT0 -> CE
>=20
> ECT1 -> ECT1
> ECT1 -> ECT0 (possibly)
> ECT1 -> CE
>=20
> CE -> CE
>=20
>=20
> Thinking about this aspect more, I am now unsure, if the higher level
> functions, such as detection of IP ECN codepoint mangling, or validating
> across multiple RTTs, should be in this document - or not rather be in a
> separate draft, in order to facilitate independent implementation and
> improvement of mechanisms in that space.
>=20
> However, the above table of allowed IP ECN transistions may be updated
> independently of AccECN - and performing extensive checks as part of a
> signal protocol ossifies this IMHO.
>=20
> As a middle ground, perhaps having these descriptions in a non-normative
> section, clearly stating these to be examples of what an impemented
> could choose to do?
>=20
> Richard
>=20
>> Am 05.02.2022 um 02:14 schrieb Vidhi Goel:
>> I like your response for all the points. I can review the diff for the
>> proposed changes, if you=E2=80=99d like before committing them to the dra=
ft.
>>=20
>> Sorry, I want to add a new comment for this text,
>>=20
>> If a TCP client has set the SYN to Not-ECT, but
>>    receives feedback that the IP-ECN field on the SYN arrived with a
>>    different codepoint, it can detect such middlebox interference and
>>    send Not-ECT for the rest of the connection.
>>=20
>>=20
>> This statement holds good for Not-ECT but doesn=E2=80=99t if lets say the=
 SYN
>> was ECT0 or ECT1 as those could be marked CE and still be valid. Should
>> we add a statement for ECT marked SYN as well?
>>=20
>> Thanks,
>> Vidhi
>>=20
>>> On Feb 4, 2022, at 5:44 AM, Bob Briscoe <ietf@bobbriscoe.net
>>> <mailto:ietf@bobbriscoe.net>> wrote:
>>>=20
>>> Vidhi,
>>>=20
>>> I've increased the table of contents to tocdepth =3D "4" which might
>>> help find the relevant sections better, because a lot of the meat in
>>> this draft is under sections 3.2.2 (ACE) and 3.2.3 (Option).
>>>=20
>>> I've responded below, but I'd prefer it if you would say these things
>>> on the list, so we have a better justification for changing the draft.
>>> Reviewing a draft that has just be re-posted shouldn't imply anything
>>> about whether Apple is implementing the draft or not.
>>>=20
>>> (BTW, regarding the proposed change to the initial value of r.e1b, I
>>> was going to send that  to the list once co-authors agreed to it. But
>>> it would be nice if some of the edits were not self-generated.)
>>>=20
>>> See [BB] inline for responses (which I'll repeat if you send to the list=
).
>>>=20
>>> On 04/02/2022 07:37, Vidhi Goel wrote:
>>>> Hello Bob and authors,
>>>>=20
>>>> I have read the draft a few times and will send my extensive review
>>>> at a later point. But right now, I want to ask some critical things
>>>> that an implementation could benefit from.
>>>>=20
>>>> The draft says,
>>>> /If a TCP client has set the SYN to Not-ECT, but/
>>>> /   receives feedback that the IP-ECN field on the SYN arrived with a/
>>>> /   different codepoint, it can detect such middlebox interference and/=

>>>> /   send Not-ECT for the rest of the connection/
>>>>=20
>>>> 1. On the forward path from client to server, the client will revert
>>>> to Not-ECT when it sees for example, that it sent a Non-ECT SYN but
>>>> received ACE encoding other than 0 1 0. What does the client do in
>>>> the last ACK of the 3WHS -
>>>>  a. does it stay in AccECN mode and still send AccECN encoding based
>>>> on the IP code point of SYN-ACK (Table 4)? This would mean that
>>>> client won=E2=80=99t participate in ECN on the sender half of the conne=
ction
>>>> and only provide AccECN feedback to the server as a receiver.
>>>>  b.  does it disable AccECN mode and set ACE=3D 0 0 0 so that a server
>>>> in AccECN mode can disable ECN based on Table4?
>>>>=20
>>>> While writing this, I realized that the intention is probably a.
>>>> Could you confirm? Also, when the sender sets Not-ECT in its data
>>>> packets, it should also disable acting upon any ACE feedback as we
>>>> could still receive false ACE feedback from the server if the
>>>> network, lets say, changed 00 (not ECT) to 01(ECT1). If you agree, we
>>>> should add some text around this. Based on the current text, the
>>>> sender will always respond to the ACE feedback even if it sends Not-ECT=
.
>>>>=20
>>>> TBH, this is a complicated scenario, where sender said to the network
>>>> - I don=E2=80=99t trust you so I can=E2=80=99t use ECN. Feel free to dr=
op my packets.
>>>> And the network mangles the IP to ECT1 and then set CE (when
>>>> congested) which would be feedback=E2=80=99ed from the receiver. Now, t=
he
>>>> packet wasn=E2=80=99t dropped which it should have been. So, is it bett=
er to
>>>> just ignore this feedback because sender doesn=E2=80=99t trust the netw=
ork or
>>>> just act on it and reduce cwnd in order to reduce congestion in the
>>>> network somewhere.
>>>=20
>>> [BB] Your points are all good ones.
>>> I'll address this last complex scenario first, because it has
>>> implications for the earlier questions. First I'll define some
>>> terminology:
>>>=20
>>> Simple mangling scenario:
>>>=20
>>>  * some network function, e.g. broken Diffserv ToS-byte remarking,
>>>    illegally remarks a Not-ECT  SYN to CE.
>>>=20
>>> Your complex mangling scenario, repeated here:
>>>=20
>>>  * some network function, e.g. broken Diffserv ToS-byte remarking,
>>>    illegally remarks a Not-ECT SYN to ECT0 or ECT1,
>>>  * then congestion at a subsequent bottleneck is marking some packets CE=

>>>=20
>>> Then potentially the client could tailor its behaviour after sending a
>>> Not-ECT SYN, If the SYN/ACK feedback is
>>>=20
>>> 1. ECT: disable sending ECT, but continue responding to ECN feedback
>>> 2. CE, disable sending ECT, and disable response to ECN feedback
>>>=20
>>> However, this is uncertain, because CE feedback on the SYN/ACK could
>>> indicate either the simple or the complex mangling scenario.
>>>=20
>>> A simpler alternative would be to always continue responding to ECN
>>> feedback. Rationale:
>>>=20
>>>  * Whether case #1 or #2, assume that mangling of the SYN might have
>>>    been Not-ECT to ECT, even if the feedback off the SYN is CE.
>>>  * Then as the connection progresses, if /all/ feedback is CE,
>>>    there's already a recommendation to fall-back to disabling
>>>    congestion response.
>>>=20
>>> If we do this, I think we ought to say "SHOULD continue to respond to
>>> ECN feedback", not "MUST".
>>>=20
>>> And we'll need to put this all to the WG.
>>>=20
>>> 1.a) Now back to the beginning of your point. The text you quote is
>>> from =C2=A72.5 which is in the non-normative "Overview and Rationale"
>>> section (=C2=A72).
>>> You really need the normative text from:
>>>=20
>>>=20
>>>          3.2.2.3.
>>>          <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-1=
6.html#section-3.2.2.3>Testing
>>>          for Mangling of the IP/ECN Field
>>>          <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-1=
6.html#name-testing-for-mangling-of-the>
>>>=20
>>>=20
>>> where it says (end of 1st para):
>>>=20
>>>    " ...for the remainder of the connection the client MUST NOT send
>>>    ECN-capable packets, but it MUST continue to feed back any ECN
>>>    markings on arriving packets."
>>>=20
>>> So, you're right, the answer is (a).
>>> It can't be (b), because, once in AccECN mode, 000 on the ACE field
>>> just becomes a counter value, and no longer negotiates the feedback mode=
.
>>>=20
>>> 1.b) As per the above, let's conservatively assume complex mangling.
>>> So, at the end of the first para quoted above I suggest we add:
>>>=20
>>>    " ...for the remainder of the connection the client MUST NOT send
>>>    ECN-capable packets *but**it**SHOULD continue to respond to ECN
>>>    feedback even though **it **is no longer sending ECN-capable
>>>    packets (see reasoning below)**. T**he client**MUST remain in the
>>>    same feedback mode and***it MUST continue to feed back any ECN
>>>    markings on arriving packets."
>>>=20
>>> I'll do the same for next para about the server.
>>> I'll work out some text for the reasoning, but I won't give it here.
>>>=20
>>> There are two other cases where it disables sending ECT, which don't
>>> say whether it continues to respond to congestion:
>>>=20
>>>  * "Testing for Mangling" section, penultimate para,  where it's
>>>    receiving solid CE:
>>>    "Once a Data Sender has entered AccECN mode it SHOULD check
>>>    whether all feedback received for the first three or four rounds
>>>    indicated that every packet it sent was CE-marked. If so, for the
>>>    remainder of the connection, the Data Sender SHOULD NOT send
>>>    ECN-capable packets*and it SHOULD NOT respond to ECN feedback*,
>>>    but *it MUST remain in the same feedback mode and *it MUST
>>>    continue to feed back any ECN markings on arriving packets*(in its
>>>    role as Data Receiver)*."
>>>  * Next section "Zeroing of the ACE Field", 2nd para:
>>>    "If the value of this ACE field is zero (0b000), the Data Sender
>>>    disables sending ECN-capable packets for the remainder of the
>>>    half-connection by setting the IP/ECN field in all subsequent
>>>    packets to Not-ECT, *but**it**SHOULD continue to respond to ECN
>>>    feedback even though **it **is no longer sending ECN-capable
>>>    packets.* *It MUST also remain in the same feedback mode and it
>>>    MUST continue to feed back any ECN markings on arriving packets
>>>    (in its role as Data Receiver).*"
>>>=20
>>> I'll also add reasoning (in the zeroing section, it already says that
>>> ACE=3D0b000 is not an unambiguous indication of mangling).
>>>=20
>>> And this prompts me to edit the bullet in an earlier section about the
>>> obligation of a sender to respond to congestion feedback:
>>>=20
>>>=20
>>>        3.1.5.
>>>        <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.=
html#section-3.1.5>Implications
>>>        of AccECN Mode
>>>        <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.=
html#name-implications-of-accecn-mode>
>>>=20
>>>=20
>>>    "It is still obliged to respond appropriately to AccECN feedback
>>>    that indicates there were ECN marks on *ECN-capable *packets it
>>>    had previously sent."
>>>=20
>>> And I'll add a bullet:
>>>=20
>>>    "*If the sender chooses not to send ECN-capable packets (e.g.
>>>    because path traversal of the ECN field is suspect), it can ignore
>>>    any **ECN **feedback about those packets if it is certain that it
>>>    cannot be valid (see Section 3.2.2, which gives normative
>>>    requirements for certain specific cases).*"
>>>=20
>>> I've said 'can' rather than 'MUST' because it's hard to cover all
>>> cases, e.g. single packets without ECT when it's not clear whether the
>>> feedback covered another packet that was ECT.
>>>=20
>>> How does all this sound?
>>>=20
>>>> 2. On the reverse path from server to client, if a server sends a
>>>> Not-ECT SYN-ACK and receives ACE handshake encoding on last ACK other
>>>> than 0 1 0, there is no text like above that says server should send
>>>> Not-ECT for the rest of the connection (or at least I didn=E2=80=99t fi=
nd
>>>> it). I think the server should also do same as client as the two
>>>> paths could be different. One could make it more complicated by
>>>> saying, if both client and server decide to not use ECN on their
>>>> corresponding sender half, then ECN should be disabled but let=E2=80=99=
s talk
>>>> about that later.
>>>=20
>>> [BB] It's in the second para of:
>>>=20
>>>=20
>>>          3.2.2.3.
>>>          <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-1=
6.html#section-3.2.2.3>Testing
>>>          for Mangling of the IP/ECN Field
>>>          <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-1=
6.html#name-testing-for-mangling-of-the>
>>>=20
>>>=20
>>>> 3. In general, not setting the ECT (ECT0 or ECT1) code point on an
>>>> outgoing packet is different from supporting AccECN right as in the
>>>> host can still provide AccECN feedback on the receive path.
>>>=20
>>> [BB] Yup.
>>>=20
>>> Do you think any further explanation is needed in the above sections?
>>>=20
>>>=20
>>> Bob
>>>=20
>>>> =E2=80=A6.
>>>>=20
>>>> To be continued if more questions come to my mind.
>>>>=20
>>>>=20
>>>> Thanks,
>>>> Vidhi
>>>>=20
>>>>> On Feb 3, 2022, at 7:24 AM, Bob Briscoe <ietf@bobbriscoe.net> wrote:
>>>>>=20
>>>>> tcpm folks,
>>>>>=20
>>>>> This rev to accurate-ecn is the first of two. The second will
>>>>> hopefully follow on its heels in the next couple of days.
>>>>>=20
>>>>> Diffs between this first rev (-16) and -15:
>>>>>=20
>>>>> 1.  switches round two fairly large sections, so I've deferred other
>>>>> changes to a second rev so the diffs won't be masked by the switch
>>>>> round of sections.
>>>>> Suggested by Ilpo to match the order in which the tests in these
>>>>> sections will be executed:
>>>>> * Test for mangling the IP-ECN field (now 3.2.2.3),
>>>>> * Then test for zeroing the ACE field (now 3.2.2.4).
>>>>>=20
>>>>> 2. Ilpo suggested some clarifications in "3.2.3.2.5. Consistency
>>>>> between AccECN Feedback Fields", which is about the receiver of
>>>>> feedback ensuring consistency between the mandatory 3-bit ACE field
>>>>> and the optional 24-bit counters. In brief (paraphrasing) it
>>>>> previously only said "MUST consider both fields", when it is now
>>>>> clearer that it actually meant "MUST reconcile both fields", so that
>>>>> there is always a consistent baseline for subsequent ACKs.
>>>>>=20
>>>>> 3. A minor point is added in an appendix about the details that the
>>>>> pseudocode doesn't cover.
>>>>>=20
>>>>>=20
>>>>>=20
>>>>> Bob
>>>>>=20
>>>>> PS. #2 & #3 were added to the XML ages ago (Jul '21), so you will
>>>>> have seen them in the HTML. However, prob due to my clumsiness, the
>>>>> posted TXT didn't include them whereas the posted XML did (ironic
>>>>> for a section about consistency). In turn, inconsistency was only
>>>>> possible because I am having to manually post the TXT for this
>>>>> draft, due to an unresolved issue with v3 RFC XML tables.
>>>>>=20
>>>>>=20
>>>>> On 03/02/2022 14:43, 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 TCP Maintenance and Minor
>>>>>> Extensions WG of the IETF.
>>>>>>=20
>>>>>>         Title           : More Accurate ECN Feedback in TCP
>>>>>>         Authors         : Bob Briscoe
>>>>>>                           Mirja K=C3=BChlewind
>>>>>>                           Richard Scheffenegger
>>>>>> Filename        : draft-ietf-tcpm-accurate-ecn-16.txt
>>>>>> Pages           : 60
>>>>>> Date            : 2022-02-03
>>>>>>=20
>>>>>> Abstract:
>>>>>>    Explicit Congestion Notification (ECN) is a mechanism where networ=
k
>>>>>>    nodes can mark IP packets instead of dropping them to indicate
>>>>>>    incipient congestion to the end-points.  Receivers with an ECN-
>>>>>>    capable transport protocol feed back this information to the sende=
r.
>>>>>>    ECN was originally specified for TCP in such a way that only one
>>>>>>    feedback signal can be transmitted per Round-Trip Time (RTT).
>>>>>>  Recent
>>>>>>    new TCP mechanisms like Congestion Exposure (ConEx), Data Center T=
CP
>>>>>>    (DCTCP) or Low Latency Low Loss Scalable Throughput (L4S) need mor=
e
>>>>>>    accurate ECN feedback information whenever more than one marking i=
s
>>>>>>    received in one RTT.  This document specifies a scheme to provide
>>>>>>    more than one feedback signal per RTT in the TCP header.  Given TC=
P
>>>>>>    header space is scarce, it allocates a reserved header bit
>>>>>> previously
>>>>>>    assigned to the ECN-Nonce.  It also overloads the two existing ECN=

>>>>>>    flags in the TCP header.  The resulting extra space is exploited t=
o
>>>>>>    feed back the IP-ECN field received during the 3-way handshake as
>>>>>>    well.  Supplementary feedback information can optionally be provid=
ed
>>>>>>    in a new TCP option, which is never used on the TCP SYN.  The
>>>>>>    document also specifies the treatment of this updated TCP wire
>>>>>>    protocol by middleboxes.
>>>>>>=20
>>>>>>=20
>>>>>> The IETF datatracker status page for this draft is:
>>>>>> https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/
>>>>>>=20
>>>>>> There is also an htmlized version available at:
>>>>>> https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16=

>>>>>>=20
>>>>>> A diff from the previous version is available at:
>>>>>> https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-accurate-ecn-16
>>>>>>=20
>>>>>>=20
>>>>>> Internet-Drafts are also available by rsync at rsync.ietf.org
>>>>>> <http://rsync.ietf.org>::internet-drafts
>>>>>>=20
>>>>>>=20
>>>>>> _______________________________________________
>>>>>> tcpm mailing list
>>>>>> tcpm@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>>=20
>>>>> --
>>>>> ________________________________________________________________
>>>>> Bob Briscoe http://bobbriscoe.net/
>>>>>=20
>>>>> _______________________________________________
>>>>> tcpm mailing list
>>>>> tcpm@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>=20
>>>=20
>>> --
>>> ________________________________________________________________
>>> Bob Briscoehttp://bobbriscoe.net/
>>=20
>>=20
>> _______________________________________________
>> tcpm mailing list
>> tcpm@ietf.org
>> https://www.ietf.org/mailman/listinfo/tcpm


From nobody Sun Feb  6 03:01:58 2022
Return-Path: <nsd.ietf@gmail.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1D2703A232A for <tcpm@ietfa.amsl.com>; Sun,  6 Feb 2022 03:01:43 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.097
X-Spam-Level: 
X-Spam-Status: No, score=-2.097 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4cVPbIzB1ox4 for <tcpm@ietfa.amsl.com>; Sun,  6 Feb 2022 03:01:36 -0800 (PST)
Received: from mail-qk1-x72c.google.com (mail-qk1-x72c.google.com [IPv6:2607:f8b0:4864:20::72c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 24C7C3A232B for <tcpm@ietf.org>; Sun,  6 Feb 2022 03:01:36 -0800 (PST)
Received: by mail-qk1-x72c.google.com with SMTP id 200so8714334qki.2 for <tcpm@ietf.org>; Sun, 06 Feb 2022 03:01:36 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112;  h=mime-version:references:in-reply-to:from:date:message-id:subject:to;  bh=s2HQQD8bW1LBR8UMKNEZze98RAsH6tJnqYaDm56M/zA=; b=g+tR5f2lLYz5eTiZW053jy3WoC4uFUyg60Iy1v3mH22mDJsDSoa8vIDo/l65T77ooO HVis0lpw67hK+EtbitlT5XpmnTs44Mhneb/GOdRk7DVCL+VppAunM7hvVRydtZPjbxyE UXPkeEXIApTOZwHvLKCgsOyIYlAItzO6V1WWHNOvPYKp6CL1MX9mrf+Ff72iIIOwU5i1 u04LITn+LMSjVLqPwIe1bZC+2GEbIRq2G8EpEq5s/LMJzhT6pD5/XGY8YB7W9Ir4ONQQ q/JXrMmMx2U94clwUzIkLv7fhXbfX0eGz0p3uqa8X826yutG6DSKGtHi8aSJbom1q4Pn My4w==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to; bh=s2HQQD8bW1LBR8UMKNEZze98RAsH6tJnqYaDm56M/zA=; b=NMESLmwAgEdtRZuBoh8jFwUhO8ixcVV5HUCpdb5d6X2vJfEdFoG71tJBkBeMWT7p6+ F1HUGKpttHADyQs7TmOEHm6/k2QwAtOPtoiqy/QSjq5fGYoCNd9eNRnv6jYZMbvPAhSk xQw5KtsRWAuO31ozTx0reOsfqs4n2XZEPX85YcbxenTngUv7KrTOMq12Txn+hDnEW7yA BJdQQzOPtTeBoXQBlrx45UXlGbGUfElrGESsxneyIDzXzpDlMew0VM0Nt7c6MJigmo4c BZ3Fe4RLE2bpPZxK35bB9Y3jI4kkGdWMjjmVR8r+dxYKhgy3d/V9lDmkBdJkZZ5lJcjq ehfw==
X-Gm-Message-State: AOAM530oezJO0vQFvZdX7rSmvPJsVRdlN7YZTbr1Mgwwa4CXgfslfQRB r7kYqxjTQ0wZdudeYILYpNRsgQaA1QdmEdM8wCzoPl+O
X-Google-Smtp-Source: ABdhPJx9tY3RGEwejHled3y9tKXjvbPInj08u6paavuo49SegVGb/JAaE983RF2q8lB0OQ5VHIFt7NybwXrPY8efiU0=
X-Received: by 2002:a05:620a:2441:: with SMTP id h1mr3906624qkn.353.1644145294153;  Sun, 06 Feb 2022 03:01:34 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com>
In-Reply-To: <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com>
From: Yoshifumi Nishida <nsd.ietf@gmail.com>
Date: Sun, 6 Feb 2022 03:01:23 -0800
Message-ID: <CAAK044R6B7HT0mdSDwHU=FhvN=A=M-8pxYb8wGfwxczwtCT-yw@mail.gmail.com>
To: "tcpm@ietf.org Extensions" <tcpm@ietf.org>
Content-Type: multipart/alternative; boundary="000000000000379b1405d75768bb"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/9fSZqABNNT4-r-zGDS0kHBUivqo>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 06 Feb 2022 11:01:51 -0000

--000000000000379b1405d75768bb
Content-Type: text/plain; charset="UTF-8"

Hi,
I just have one more comment on the draft.
In RFC5681, it mentions in Section 3:
''

   Implementation Note: An easy mistake to make is to simply use cwnd,
   rather than FlightSize, which in some implementations may
   incidentally increase well beyond rwnd.

"

while this draft mentions in Section 4.6:
"

   To avoid suboptimal performance with such applications,

   the mechanisms described in [RFC7661
<https://datatracker.ietf.org/doc/html/rfc7661>] can be used to
mitigate this
   issue as it would allow using a value between _cwnd_ and
   _flight_size_ to calculate the new _ssthresh_ in Figure 5.

"

So, it seems one PS doc mentions not to use cwnd, while another PS doc
mentions using cwnd is fine (with RFC7661).

I am not sure which can cause this situation yet, but in any case, does
this mean there is no concern on
"some implementations may incidentally increase well beyond rwnd" in cubic?
Is this because this is an old implementation bug or are there other
reasons?
I am wondering if we might want some clarification on this point.

Also, I am wondering if we might want to use "MAY be used" here instead of
"can be used"

Thanks,
--
Yoshi

On Mon, Jan 31, 2022 at 11:43 PM Yoshifumi Nishida <nsd.ietf@gmail.com>
wrote:

> Hello,
>
> After some discussions among chairs, we decided to run the 2nd WGLC on
> draft-ietf-tcpm-rfc8312bis in consideration of the importance of the draft.
> We'll be grateful if you could send your feedback to the ML. The WGLC runs
> until *Feb 11*.
>
> If interested, you can check in-depth past discussions in the following
> URL.
> https://github.com/NTAP/rfc8312bis/
>
> Thank you so much!
> --
> tcpm co-chairs
>
>
> On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <lars@eggert.org> wrote:
>
>> Hi,
>>
>> this -06 version rolls in all the changes requested during (and after)
>> WGLC ended.
>>
>> I'll leave it up to the chairs to decide if another WGLC is warranted or
>> the document can progress as-is.
>>
>> Thanks,
>> Lars
>>
>>
>> > On 2022-1-26, at 11:12, 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 TCP Maintenance and Minor Extensions
>> WG of the IETF.
>> >
>> >        Title           : CUBIC for Fast and Long-Distance Networks
>> >        Authors         : Lisong Xu
>> >                          Sangtae Ha
>> >                          Injong Rhee
>> >                          Vidhi Goel
>> >                          Lars Eggert
>> >       Filename        : draft-ietf-tcpm-rfc8312bis-06.txt
>> >       Pages           : 35
>> >       Date            : 2022-01-26
>> >
>> > Abstract:
>> >   CUBIC is a standard TCP congestion control algorithm that uses a
>> >   cubic function instead of a linear congestion window increase
>> >   function to improve scalability and stability over fast and long-
>> >   distance networks.  CUBIC has been adopted as the default TCP
>> >   congestion control algorithm by the Linux, Windows, and Apple stacks.
>> >
>> >   This document updates the specification of CUBIC to include
>> >   algorithmic improvements based on these implementations and recent
>> >   academic work.  Based on the extensive deployment experience with
>> >   CUBIC, it also moves the specification to the Standards Track,
>> >   obsoleting RFC 8312.  This also requires updating RFC 5681, to allow
>> >   for CUBIC's occasionally more aggressive sending behavior.
>> >
>> >
>> > The IETF datatracker status page for this draft is:
>> > https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/
>> >
>> > There is also an HTML version available at:
>> > https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html
>> >
>> > A diff from the previous version is available at:
>> > https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rfc8312bis-06
>> >
>> >
>> > Internet-Drafts are also available by rsync at rsync.ietf.org:
>> :internet-drafts
>> >
>> >
>> > _______________________________________________
>> > tcpm mailing list
>> > tcpm@ietf.org
>> > https://www.ietf.org/mailman/listinfo/tcpm
>>
>> _______________________________________________
>> tcpm mailing list
>> tcpm@ietf.org
>> https://www.ietf.org/mailman/listinfo/tcpm
>>
>

--000000000000379b1405d75768bb
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi,</div><div>I just have one more comment on the dra=
ft.</div><div>In RFC5681, it mentions in Section 3:</div><div>&#39;&#39;</d=
iv><div><pre style=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0px;=
break-before:page;color:rgb(0,0,0)">   Implementation Note: An easy mistake=
 to make is to simply use cwnd,
   rather than FlightSize, which in some implementations may
   incidentally increase well beyond rwnd.</pre></div><div>&quot;</div><div=
><br></div><div>while this draft mentions in Section 4.6:</div><div>&quot;<=
/div><div><pre style=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0p=
x;break-before:page;color:rgb(0,0,0)">   To avoid suboptimal performance wi=
th such applications,</pre><pre style=3D"font-size:13.3333px;margin-top:0px=
;margin-bottom:0px;break-before:page;color:rgb(0,0,0)">   the mechanisms de=
scribed in [<a href=3D"https://datatracker.ietf.org/doc/html/rfc7661" title=
=3D"&quot;Updating TCP to Support Rate-Limited Traffic&quot;" target=3D"_bl=
ank">RFC7661</a>] can be used to mitigate this
   issue as it would allow using a value between _cwnd_ and
   _flight_size_ to calculate the new _ssthresh_ in Figure 5. </pre><pre st=
yle=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0px;break-before:pa=
ge;color:rgb(0,0,0)">&quot;</pre></div><div>So, it seems one PS doc mention=
s not to use cwnd, while another PS doc mentions using cwnd is fine (with R=
FC7661).</div><div><br></div><div><div>I am not sure which can cause this s=
ituation yet, but in any case, does this mean there is no concern on=C2=A0<=
/div><div>&quot;some implementations may incidentally increase well beyond =
rwnd&quot; in cubic?</div></div><div>Is this because this is an old impleme=
ntation bug or are there other reasons?<br></div><div>I am wondering if we =
might want some clarification on this point.<br></div><div>=C2=A0</div><div=
>Also, I am wondering if we might want to use &quot;MAY be used&quot; here =
instead of &quot;can be used&quot;=C2=A0=C2=A0</div><div><br></div><div>Tha=
nks,</div><div>--</div><div>Yoshi</div><div><br></div><div class=3D"gmail_q=
uote"><div dir=3D"ltr" class=3D"gmail_attr">On Mon, Jan 31, 2022 at 11:43 P=
M Yoshifumi Nishida &lt;<a href=3D"mailto:nsd.ietf@gmail.com" target=3D"_bl=
ank">nsd.ietf@gmail.com</a>&gt; wrote:<br></div><blockquote class=3D"gmail_=
quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,=
204);padding-left:1ex"><div dir=3D"ltr"><div>Hello,</div><div><br></div><di=
v>After some discussions among chairs, we decided to run the 2nd WGLC on dr=
aft-ietf-tcpm-rfc8312bis in consideration of the importance of the draft.=
=C2=A0</div><div>We&#39;ll be grateful if you could send your feedback to t=
he ML. The WGLC runs until *Feb 11*.</div><div><br></div><div>If interested=
, you can check in-depth past discussions in the following URL.</div><div><=
a href=3D"https://github.com/NTAP/rfc8312bis/" target=3D"_blank">https://gi=
thub.com/NTAP/rfc8312bis/</a><br></div><div><br></div><div>Thank you so muc=
h!</div><div>--</div><div>tcpm co-chairs</div><div><br></div><br><div class=
=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Wed, Jan 26, 2022=
 at 2:50 AM Lars Eggert &lt;<a href=3D"mailto:lars@eggert.org" target=3D"_b=
lank">lars@eggert.org</a>&gt; wrote:<br></div><blockquote class=3D"gmail_qu=
ote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,20=
4);padding-left:1ex">Hi,<br>
<br>
this -06 version rolls in all the changes requested during (and after) WGLC=
 ended.<br>
<br>
I&#39;ll leave it up to the chairs to decide if another WGLC is warranted o=
r the document can progress as-is.<br>
<br>
Thanks,<br>
Lars<br>
<br>
<br>
&gt; On 2022-1-26, at 11:12, <a href=3D"mailto:internet-drafts@ietf.org" ta=
rget=3D"_blank">internet-drafts@ietf.org</a> wrote:<br>
&gt; <br>
&gt; <br>
&gt; A New Internet-Draft is available from the on-line Internet-Drafts dir=
ectories.<br>
&gt; This draft is a work item of the TCP Maintenance and Minor Extensions =
WG of the IETF.<br>
&gt; <br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Title=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0: CUBIC for Fast and Long-Distance Networks<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Authors=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: =
Lisong Xu<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Sangtae Ha<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Injong Rhee<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Vidhi Goel<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Lars Eggert<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Filename=C2=A0 =C2=A0 =C2=A0 =C2=A0 : draft-=
ietf-tcpm-rfc8312bis-06.txt<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Pages=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0: 35<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Date=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 : 2022-01-26<br>
&gt; <br>
&gt; Abstract:<br>
&gt;=C2=A0 =C2=A0CUBIC is a standard TCP congestion control algorithm that =
uses a<br>
&gt;=C2=A0 =C2=A0cubic function instead of a linear congestion window incre=
ase<br>
&gt;=C2=A0 =C2=A0function to improve scalability and stability over fast an=
d long-<br>
&gt;=C2=A0 =C2=A0distance networks.=C2=A0 CUBIC has been adopted as the def=
ault TCP<br>
&gt;=C2=A0 =C2=A0congestion control algorithm by the Linux, Windows, and Ap=
ple stacks.<br>
&gt; <br>
&gt;=C2=A0 =C2=A0This document updates the specification of CUBIC to includ=
e<br>
&gt;=C2=A0 =C2=A0algorithmic improvements based on these implementations an=
d recent<br>
&gt;=C2=A0 =C2=A0academic work.=C2=A0 Based on the extensive deployment exp=
erience with<br>
&gt;=C2=A0 =C2=A0CUBIC, it also moves the specification to the Standards Tr=
ack,<br>
&gt;=C2=A0 =C2=A0obsoleting RFC 8312.=C2=A0 This also requires updating RFC=
 5681, to allow<br>
&gt;=C2=A0 =C2=A0for CUBIC&#39;s occasionally more aggressive sending behav=
ior.<br>
&gt; <br>
&gt; <br>
&gt; The IETF datatracker status page for this draft is:<br>
&gt; <a href=3D"https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis=
/" rel=3D"noreferrer" target=3D"_blank">https://datatracker.ietf.org/doc/dr=
aft-ietf-tcpm-rfc8312bis/</a><br>
&gt; <br>
&gt; There is also an HTML version available at:<br>
&gt; <a href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-=
06.html" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/archive/=
id/draft-ietf-tcpm-rfc8312bis-06.html</a><br>
&gt; <br>
&gt; A diff from the previous version is available at:<br>
&gt; <a href=3D"https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312=
bis-06" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/rfcdiff?u=
rl2=3Ddraft-ietf-tcpm-rfc8312bis-06</a><br>
&gt; <br>
&gt; <br>
&gt; Internet-Drafts are also available by rsync at rsync.ietf.org::interne=
t-drafts<br>
&gt; <br>
&gt; <br>
&gt; _______________________________________________<br>
&gt; tcpm mailing list<br>
&gt; <a href=3D"mailto:tcpm@ietf.org" target=3D"_blank">tcpm@ietf.org</a><b=
r>
&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/tcpm" rel=3D"noreferr=
er" target=3D"_blank">https://www.ietf.org/mailman/listinfo/tcpm</a><br>
<br>
_______________________________________________<br>
tcpm mailing list<br>
<a href=3D"mailto:tcpm@ietf.org" target=3D"_blank">tcpm@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/tcpm" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/listinfo/tcpm</a><br>
</blockquote></div></div>
</blockquote></div></div>

--000000000000379b1405d75768bb--


From nobody Sun Feb  6 03:03:23 2022
Return-Path: <nsd.ietf@gmail.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 708B53A2321 for <tcpm@ietfa.amsl.com>; Sun,  6 Feb 2022 03:03:20 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.838
X-Spam-Level: 
X-Spam-Status: No, score=-1.838 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, HTML_OBFUSCATE_05_10=0.26, SPF_HELO_NONE=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1LcZ0yUYFNjG for <tcpm@ietfa.amsl.com>; Sun,  6 Feb 2022 03:03:17 -0800 (PST)
Received: from mail-qk1-x72e.google.com (mail-qk1-x72e.google.com [IPv6:2607:f8b0:4864:20::72e]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 141143A2436 for <tcpm@ietf.org>; Sun,  6 Feb 2022 03:02:44 -0800 (PST)
Received: by mail-qk1-x72e.google.com with SMTP id c189so8663429qkg.11 for <tcpm@ietf.org>; Sun, 06 Feb 2022 03:02:44 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112;  h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=SeVA0JYtYOWNW0lqsHNxuDP9xawlCBscRGdNVWt+Igw=; b=YJ39vu+fCbPaZMHSJ5/QUeLoQu6Oo0XJdh4iGbBZaM+yYxKmvYF6F3XRSJtS9grc8C /WcVXpiW6qYhbUDhv/MP8J5pocWs7Qe+yMnm8IvtT4P12VWuN/+9QfgdbuveJtnyMjpo abGd8HW9RznwC26fnD5fOuRRmhziIilByBM0NG9HV5mPlBthXlhW+V8mGX2Qfcp/u/Rn QopuBfOc3GCxHWIeqtcP7AQMPhJw/FhvPRBOFphr3Vi0S9GPxJVLSDGtu4srUEEVWP46 wntLx3JB9WdaOjW61M1wlgd1ogAzk0vNhuocYCcUPfVadqRdU+/cjUHjMRMvDJLuwdS1 xUuw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=SeVA0JYtYOWNW0lqsHNxuDP9xawlCBscRGdNVWt+Igw=; b=kHRT6dqTn6TwT0CriYnt4yTtPktBLNEuzxoazPt5A8vw2y1lr5BoB6lZ+GAAmGIQDX BjT0KmZOTB+1KzxwuDjgIaMpoKDwHjDizRRffur5e9Vp2/FNGaLKqNSMKRPpqLpR3+Qe Iu4dsOn5LX6bXxzNuFJde8KZSauNPh6DH23/28lhBbN02SAk+KHECFhTwN68VFQ44UVZ 7mM1+oHK4WBbiAL5Syh36iExN+zKTUzMwF/uzBooAsmoT96Bm2ZBLiBp6RrPaKPIjnVy F4HQZZjpO0gUqvzpmP7h3Dq4xG5O/KdtG7T34wN/pvkEmCLIOZtKDZqZbRAYERqt+NBt ASZg==
X-Gm-Message-State: AOAM531r55+vYY9xb+9R+r95MwQ0U+wlwEF2a0P81xBrhB+vdSKN/QNs hn0ZFkjGwTzKnhIb0uROi1mgtv4RKHAALJnhJxMWklf7
X-Google-Smtp-Source: ABdhPJzJS7zelp3rZC/Ij1wtwzyhU0YXUxrt40nJxLdtIuVFpdKt2X1VVtZDzkrTlVRJmBLEEqwRDdTfULVUvjyr8lI=
X-Received: by 2002:a05:620a:13ee:: with SMTP id h14mr3728253qkl.741.1644145363208;  Sun, 06 Feb 2022 03:02:43 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044SCF-L9Xmoc1vH=uQAe4k5XyYboamRcyTjSAaveu1uMGQ@mail.gmail.com> <BE6C47B6-FBBB-46A5-9048-FB0FCA9BC326@apple.com>
In-Reply-To: <BE6C47B6-FBBB-46A5-9048-FB0FCA9BC326@apple.com>
From: Yoshifumi Nishida <nsd.ietf@gmail.com>
Date: Sun, 6 Feb 2022 03:02:32 -0800
Message-ID: <CAAK044SdoURfZAFapRFeAjzLZ_3Bbu_H=N56AWLcT6B+QWU=TQ@mail.gmail.com>
To: Vidhi Goel <vidhi_goel@apple.com>
Cc: Lars Eggert <lars@eggert.org>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>
Content-Type: multipart/alternative; boundary="000000000000554cde05d7576c7d"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/_QmXXcr4bcnGdSxC-wqpvq8BqM8>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-rfc8312bis-06.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 06 Feb 2022 11:03:21 -0000

--000000000000554cde05d7576c7d
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi Vidhi,

On Thu, Jan 27, 2022 at 4:35 PM Vidhi Goel <vidhi_goel@apple.com> wrote:

> Also, I am wondering if https://github.com/NTAP/rfc8312bis/issues/135 has
> been solved completely.
> Have we decided to leave this part as it is? We don't need any texts here
> or are there any future plans?
>
> We have discussed internally and decided to leave it as is. The
> justification is - there is already large scale deployment of CUBIC in
> the field with Beta_cubic =3D 0.7. We had added the below line to the dra=
ft
> as a caution. To change B_cubic for so many deployments, it would require=
 a
> lot of experimental data and perhaps even some deployment experience. Suc=
h
> a change would be categorized as a research initiative before it makes it
> to an IETF draft.
>
> *A side effect of setting =CE=B2cubic to a value bigger than 0.5 is slowe=
r
> convergence. We believe that while a more adaptive setting of =CE=B2cubic=
 could
> result in faster convergence, it will make the analysis of CUBIC much
> harder.*
>

OK. I've thought about this and I am fine with this as long as we recommend
using hystart++ or other burst mitigations during slow start (which is
already done)
--
Yoshi

--000000000000554cde05d7576c7d
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi=C2=A0Vidhi,</div><div><br></div><div class=3D"gmai=
l_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Thu, Jan 27, 2022 at 4:35=
 PM Vidhi Goel &lt;<a href=3D"mailto:vidhi_goel@apple.com" target=3D"_blank=
">vidhi_goel@apple.com</a>&gt; wrote:<br></div><blockquote class=3D"gmail_q=
uote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,2=
04);padding-left:1ex"><div><blockquote type=3D"cite"><div dir=3D"ltr"><div>=
Also, I am wondering if=C2=A0<a href=3D"https://github.com/NTAP/rfc8312bis/=
issues/135" target=3D"_blank">https://github.com/NTAP/rfc8312bis/issues/135=
</a>=C2=A0has been solved completely.=C2=A0</div><div>Have we decided to le=
ave this part as it=C2=A0is? We don&#39;t need any texts here or are there =
any future plans?</div></div></blockquote>We have discussed internally and =
decided to leave it as is. The justification is - there is already=C2=A0<fo=
nt color=3D"#000000">large scale deployment of CUBIC in the field with Beta=
_cubic =3D 0.7. We had added the below line to the draft as a caution. To c=
hange B_cubic for so many deployments, it would require a lot of experiment=
al data and perhaps even=C2=A0some deployment=C2=A0experience.=C2=A0Such a =
change would be=C2=A0categorized as a research initiative before it makes i=
t to an IETF draft.</font></div><div><font color=3D"#000000"><br></font></d=
iv><div><i style=3D"color:rgb(0,0,0)"><span style=3D"color:rgb(34,34,34);fo=
nt-family:&quot;Noto Sans&quot;,Arial,Helvetica,sans-serif;background-color=
:rgb(255,255,255)">A side effect of setting=C2=A0</span><span style=3D"colo=
r:rgb(34,34,34);font-family:&quot;Noto Sans&quot;,Arial,Helvetica,sans-seri=
f">=CE=B2</span><span style=3D"color:rgb(34,34,34);font-family:&quot;Noto S=
ans&quot;,Arial,Helvetica,sans-serif"><sub>cubic</sub></span><span style=3D=
"color:rgb(34,34,34);font-family:&quot;Noto Sans&quot;,Arial,Helvetica,sans=
-serif;background-color:rgb(255,255,255)">=C2=A0to a value bigger than 0.5 =
is slower convergence. We believe that while a more adaptive setting of=C2=
=A0</span><span style=3D"color:rgb(34,34,34);font-family:&quot;Noto Sans&qu=
ot;,Arial,Helvetica,sans-serif">=CE=B2</span><span style=3D"color:rgb(34,34=
,34);font-family:&quot;Noto Sans&quot;,Arial,Helvetica,sans-serif"><sub>cub=
ic</sub></span><span style=3D"color:rgb(34,34,34);font-family:&quot;Noto Sa=
ns&quot;,Arial,Helvetica,sans-serif;background-color:rgb(255,255,255)">=C2=
=A0could result in faster convergence, it will make the analysis of CUBIC m=
uch harder.</span></i></div></blockquote><div><br></div><div>OK. I&#39;ve t=
hought about this and I am fine with this as long as we recommend using hys=
tart++ or other burst mitigations during slow start (which is already done)=
</div><div>--</div><div>Yoshi=C2=A0=C2=A0</div></div></div>

--000000000000554cde05d7576c7d--


From nobody Sun Feb  6 14:28:06 2022
Return-Path: <ietf@bobbriscoe.net>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A930C3A0E4F for <tcpm@ietfa.amsl.com>; Sun,  6 Feb 2022 14:28:01 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.813
X-Spam-Level: 
X-Spam-Status: No, score=-2.813 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, NICE_REPLY_A=-0.714, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=unavailable autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=bobbriscoe.net
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qXmIR4r3EM1Y for <tcpm@ietfa.amsl.com>; Sun,  6 Feb 2022 14:27:55 -0800 (PST)
Received: from mail-ssdrsserver2.hostinginterface.eu (mail-ssdrsserver2.hostinginterface.eu [185.185.85.90]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A4F7E3A0E52 for <tcpm@ietf.org>; Sun,  6 Feb 2022 14:27:54 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=bobbriscoe.net; s=default; h=In-Reply-To:From:References:Cc:To:Subject: MIME-Version:Date:Message-ID:Content-Type:Sender:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id: List-Help:List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=z3xqeXX7ZcdlvJxDqr8bOwpdBNsRq9ghaluUBL43xYc=; b=tDOaWi9Y7mHzULy3nGy+EsUaN3 ++zGZzf7hKC/leEoAMaIq0h/jImkCPiPfxYar04Y/UKmpzlP8zsCjmaTF1Sa1Sv1uIQqX8LkrkqNg zSf9rXN3dwGrNs+pRygii+rtjL0/dpVtBZMDbePZLO6OLt/V0NEpGEt7WOHM/Z9rDQZ00Nug7Q0cI o54FpUenGBXpTJZd7ApJZNex+whyHrNIQ9OPMSjAVsQgiaCtP5Dqk5KTq/omHq4mQipLr8amNML73 eM4xaN3TpQkFcStPVW7akf9Lga4GCpadPc2S1Ne/QfdwSdhp+Zwa3q/pZ73k78RGFh660EBlJaF8b ev+cPyIA==;
Received: from 67.153.238.178.in-addr.arpa ([178.238.153.67]:55194 helo=[192.168.1.11]) by ssdrsserver2.hostinginterface.eu with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 (Exim 4.94.2) (envelope-from <ietf@bobbriscoe.net>) id 1nGq0e-0004ar-OW; Sun, 06 Feb 2022 22:27:52 +0000
Content-Type: multipart/alternative; boundary="------------UO067S0NPkEoC6gOEK0ScEW1"
Message-ID: <615549a4-6d09-8a7b-f892-42a1ea52c2e2@bobbriscoe.net>
Date: Sun, 6 Feb 2022 22:27:45 +0000
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.5.0
Content-Language: en-GB
To: "Scheffenegger, Richard" <rs.ietf@gmx.at>, Vidhi Goel <vidhi_goel=40apple.com@dmarc.ietf.org>
Cc: Richard Scheffenegger <rscheff@gmx.at>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, Mirja Kuehlewind <ietf@kuehlewind.net>
References: <164389942954.13556.7754569279919863072@ietfa.amsl.com> <c2c5bdf9-6b34-fe19-3296-5f1cd831ed8e@bobbriscoe.net> <2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com> <e57355ae-5a9d-0a14-9478-29453d25622c@bobbriscoe.net> <BB23E172-C995-41FE-A47E-E02D5586B67D@apple.com> <1cbe3dbe-241b-3f24-096f-f1a2ffe17d98@gmx.at>
From: Bob Briscoe <ietf@bobbriscoe.net>
In-Reply-To: <1cbe3dbe-241b-3f24-096f-f1a2ffe17d98@gmx.at>
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - ssdrsserver2.hostinginterface.eu
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - bobbriscoe.net
X-Get-Message-Sender-Via: ssdrsserver2.hostinginterface.eu: authenticated_id: in@bobbriscoe.net
X-Authenticated-Sender: ssdrsserver2.hostinginterface.eu: in@bobbriscoe.net
X-Source: 
X-Source-Args: 
X-Source-Dir: 
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/kye9B4aqZsbNWS0tP-XnRasip8o>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-accurate-ecn-16.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 06 Feb 2022 22:28:02 -0000

This is a multi-part message in MIME format.
--------------UO067S0NPkEoC6gOEK0ScEW1
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Richard,

Indeed, I recently discovered I had missed an item that I added to my 
ToDo list in Mar 2021, based on a conversation between Gorry and Mirja 
where they both made the same point - that fall-back behaviour after 
discovering ECN mangling ought to be non-normative in this draft, 
because it's beyond the scope of the new feedback mechanism:
https://mailarchive.ietf.org/arch/msg/tcpm/00URm1uwptgrAzKq96T6KSh8dQ4/

We do need to say the following normatively in "Testing for Mangling of 
the IP/ECN Field" because they are relevant to feedback, the Data Sender:
* remains in AccECN mode
* MUST continue to feed back any ECN markings on arriving packets.

BTW, Vidhi, the text you quoted was from §2 on "Protocol Overview and 
Rationale". That was meant to be an example of why it is useful for the 
handshake to give feedback on exactly which IP-ECN codepoint arrived. It 
wasn't meant to be a full spec. So I'll make it clearer that it's only 
an example, rather than try to make it comprehensive, which isn't 
appropriate at that point in the doc.

Nonetheless, your points are correct, but if they are dealt with anyway, 
they should be dealt with in §3.2.2.3. "Testing for Mangling of the 
IP/ECN Field" where these scenarios are already covered (except for the 
congestion response, which was silently left out of scope):
https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3 


We could just say that the questions of whether to set ECT for 
subsequent packets, and whether to respond to feedback of CE markings 
are beyond the scope of the draft. I think we should, but I think we 
should also give some non-normative advice on what to do. It can only be 
interim advice, not normative requirements. Both because it's out of 
scope, and because the best behaviour and the alternatives will depend 
on deployment experience. So they shouldn't be written normatively in a 
Proposed Standard yet.

We will do what the WG wants, but assuming it agrees to including 
non-normative advice, following from your (Richard's) list of cases, 
here's a full matrix. I don't propose this matrix should go in the 
AccECN draft - it's just for this email, to check every case is covered.


	Feedback in SYN/ACK of SYN's IP-ECN
IP-ECN on SYN 	Not-ECT 	ECT(0) 	ECT(1) 	CE
Not ECT 	√ 	NR 	NR 	NR
ECT(0) 	ND 	√ 	√ 	√
ECT(1) 	ND 	CU
	√ 	√
CE 	n/a 	n/a 	n/a 	n/a


Key:
√       Valid response
NR    Non-normative advice: Set *N*ot-ECT for subsequent packets, but 
continue to *R*espond to CE feedback
ND    Non-normative advice: Set *N*ot-ECT for subsequent packets, and 
*D*isable response to CE feedback
CU    Currently unused transition (not listed as invalid in the draft)
n/a    not applicable in the AccECN spec, in which there is never a case 
where the sender sets CE when sending a packet.

In the draft we don't need to mention any cases where CE is sent (but a 
draft on fall-back probably would, because the sender might set CE to 
test the path). We can cover the other cases with two groups of invalid 
transitions: those where the SYN (or SYN-ACK) arriving at the receiver 
appears to have been
i) ECN-capable
ii) Not-ECN-capable.

I propose to also say that details (like what 'subsequent packets' 
implies about how long this behaviour continues) are for future study.

I'll try to write some text and propose it to this list. Perhaps it 
should be in a new appendix?


Bob

On 05/02/2022 12:01, Scheffenegger, Richard wrote:
> Vidhi,
>
> Thanks for reading the draft very closely!
>
> Well, in the non-ECT SYN case, where the SYNACK indicates an ECT (ect0,
> ect1, ce) is the obvious - and fully semantically correct - use case of
> that possible use.
>
> For the set of three codepoints (ECT0, ECT1, CE) the only other possible
> valid detection scenario would be if a SYN+CE is sent (deliberately,
> althought one can argue that a sender should not originate CE-marked
> packets), but a SYNACK without CE is returned (full or partial
> bleaching). At least under the current RFC3168 definition of IP ECN
> codepoints.
>
> allowed transitions:
> nonECT -> nonECT
>
> ECT0 -> ECT0
> ECT0 -> ECT1
> ECT0 -> CE
>
> ECT1 -> ECT1
> ECT1 -> ECT0 (possibly)
> ECT1 -> CE
>
> CE -> CE
>
>
> Thinking about this aspect more, I am now unsure, if the higher level
> functions, such as detection of IP ECN codepoint mangling, or validating
> across multiple RTTs, should be in this document - or not rather be in a
> separate draft, in order to facilitate independent implementation and
> improvement of mechanisms in that space.
>
> However, the above table of allowed IP ECN transistions may be updated
> independently of AccECN - and performing extensive checks as part of a
> signal protocol ossifies this IMHO.
>
> As a middle ground, perhaps having these descriptions in a non-normative
> section, clearly stating these to be examples of what an impemented
> could choose to do?
>
> Richard
>
> Am 05.02.2022 um 02:14 schrieb Vidhi Goel:
>> I like your response for all the points. I can review the diff for the
>> proposed changes, if you’d like before committing them to the draft.
>>
>> Sorry, I want to add a new comment for this text,
>>
>> If a TCP client has set the SYN to Not-ECT, but
>>     receives feedback that the IP-ECN field on the SYN arrived with a
>>     different codepoint, it can detect such middlebox interference and
>>     send Not-ECT for the rest of the connection.
>>
>>
>> This statement holds good for Not-ECT but doesn’t if lets say the SYN
>> was ECT0 or ECT1 as those could be marked CE and still be valid. Should
>> we add a statement for ECT marked SYN as well?
>>
>> Thanks,
>> Vidhi
>>
>>> On Feb 4, 2022, at 5:44 AM, Bob Briscoe <ietf@bobbriscoe.net
>>> <mailto:ietf@bobbriscoe.net>> wrote:
>>>
>>> Vidhi,
>>>
>>> I've increased the table of contents to tocdepth = "4" which might
>>> help find the relevant sections better, because a lot of the meat in
>>> this draft is under sections 3.2.2 (ACE) and 3.2.3 (Option).
>>>
>>> I've responded below, but I'd prefer it if you would say these things
>>> on the list, so we have a better justification for changing the draft.
>>> Reviewing a draft that has just be re-posted shouldn't imply anything
>>> about whether Apple is implementing the draft or not.
>>>
>>> (BTW, regarding the proposed change to the initial value of r.e1b, I
>>> was going to send that  to the list once co-authors agreed to it. But
>>> it would be nice if some of the edits were not self-generated.)
>>>
>>> See [BB] inline for responses (which I'll repeat if you send to the 
>>> list).
>>>
>>> On 04/02/2022 07:37, Vidhi Goel wrote:
>>>> Hello Bob and authors,
>>>>
>>>> I have read the draft a few times and will send my extensive review
>>>> at a later point. But right now, I want to ask some critical things
>>>> that an implementation could benefit from.
>>>>
>>>> The draft says,
>>>> /If a TCP client has set the SYN to Not-ECT, but/
>>>> /   receives feedback that the IP-ECN field on the SYN arrived with a/
>>>> /   different codepoint, it can detect such middlebox interference 
>>>> and/
>>>> /   send Not-ECT for the rest of the connection/
>>>>
>>>> 1. On the forward path from client to server, the client will revert
>>>> to Not-ECT when it sees for example, that it sent a Non-ECT SYN but
>>>> received ACE encoding other than 0 1 0. What does the client do in
>>>> the last ACK of the 3WHS -
>>>>  a. does it stay in AccECN mode and still send AccECN encoding based
>>>> on the IP code point of SYN-ACK (Table 4)? This would mean that
>>>> client won’t participate in ECN on the sender half of the connection
>>>> and only provide AccECN feedback to the server as a receiver.
>>>>  b.  does it disable AccECN mode and set ACE= 0 0 0 so that a server
>>>> in AccECN mode can disable ECN based on Table4?
>>>>
>>>> While writing this, I realized that the intention is probably a.
>>>> Could you confirm? Also, when the sender sets Not-ECT in its data
>>>> packets, it should also disable acting upon any ACE feedback as we
>>>> could still receive false ACE feedback from the server if the
>>>> network, lets say, changed 00 (not ECT) to 01(ECT1). If you agree, we
>>>> should add some text around this. Based on the current text, the
>>>> sender will always respond to the ACE feedback even if it sends 
>>>> Not-ECT.
>>>>
>>>> TBH, this is a complicated scenario, where sender said to the network
>>>> - I don’t trust you so I can’t use ECN. Feel free to drop my packets.
>>>> And the network mangles the IP to ECT1 and then set CE (when
>>>> congested) which would be feedback’ed from the receiver. Now, the
>>>> packet wasn’t dropped which it should have been. So, is it better to
>>>> just ignore this feedback because sender doesn’t trust the network or
>>>> just act on it and reduce cwnd in order to reduce congestion in the
>>>> network somewhere.
>>>
>>> [BB] Your points are all good ones.
>>> I'll address this last complex scenario first, because it has
>>> implications for the earlier questions. First I'll define some
>>> terminology:
>>>
>>> Simple mangling scenario:
>>>
>>>   * some network function, e.g. broken Diffserv ToS-byte remarking,
>>>     illegally remarks a Not-ECT  SYN to CE.
>>>
>>> Your complex mangling scenario, repeated here:
>>>
>>>   * some network function, e.g. broken Diffserv ToS-byte remarking,
>>>     illegally remarks a Not-ECT SYN to ECT0 or ECT1,
>>>   * then congestion at a subsequent bottleneck is marking some 
>>> packets CE
>>>
>>> Then potentially the client could tailor its behaviour after sending a
>>> Not-ECT SYN, If the SYN/ACK feedback is
>>>
>>>  1. ECT: disable sending ECT, but continue responding to ECN feedback
>>>  2. CE, disable sending ECT, and disable response to ECN feedback
>>>
>>> However, this is uncertain, because CE feedback on the SYN/ACK could
>>> indicate either the simple or the complex mangling scenario.
>>>
>>> A simpler alternative would be to always continue responding to ECN
>>> feedback. Rationale:
>>>
>>>   * Whether case #1 or #2, assume that mangling of the SYN might have
>>>     been Not-ECT to ECT, even if the feedback off the SYN is CE.
>>>   * Then as the connection progresses, if /all/ feedback is CE,
>>>     there's already a recommendation to fall-back to disabling
>>>     congestion response.
>>>
>>> If we do this, I think we ought to say "SHOULD continue to respond to
>>> ECN feedback", not "MUST".
>>>
>>> And we'll need to put this all to the WG.
>>>
>>> 1.a) Now back to the beginning of your point. The text you quote is
>>> from §2.5 which is in the non-normative "Overview and Rationale"
>>> section (§2).
>>> You really need the normative text from:
>>>
>>>
>>>           3.2.2.3.
>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3>Testing
>>>           for Mangling of the IP/ECN Field
>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the>
>>>
>>>
>>> where it says (end of 1st para):
>>>
>>>     " ...for the remainder of the connection the client MUST NOT send
>>>     ECN-capable packets, but it MUST continue to feed back any ECN
>>>     markings on arriving packets."
>>>
>>> So, you're right, the answer is (a).
>>> It can't be (b), because, once in AccECN mode, 000 on the ACE field
>>> just becomes a counter value, and no longer negotiates the feedback 
>>> mode.
>>>
>>> 1.b) As per the above, let's conservatively assume complex mangling.
>>> So, at the end of the first para quoted above I suggest we add:
>>>
>>>     " ...for the remainder of the connection the client MUST NOT send
>>>     ECN-capable packets *but**it**SHOULD continue to respond to ECN
>>>     feedback even though **it **is no longer sending ECN-capable
>>>     packets (see reasoning below)**. T**he client**MUST remain in the
>>>     same feedback mode and***it MUST continue to feed back any ECN
>>>     markings on arriving packets."
>>>
>>> I'll do the same for next para about the server.
>>> I'll work out some text for the reasoning, but I won't give it here.
>>>
>>> There are two other cases where it disables sending ECT, which don't
>>> say whether it continues to respond to congestion:
>>>
>>>   * "Testing for Mangling" section, penultimate para,  where it's
>>>     receiving solid CE:
>>>     "Once a Data Sender has entered AccECN mode it SHOULD check
>>>     whether all feedback received for the first three or four rounds
>>>     indicated that every packet it sent was CE-marked. If so, for the
>>>     remainder of the connection, the Data Sender SHOULD NOT send
>>>     ECN-capable packets*and it SHOULD NOT respond to ECN feedback*,
>>>     but *it MUST remain in the same feedback mode and *it MUST
>>>     continue to feed back any ECN markings on arriving packets*(in its
>>>     role as Data Receiver)*."
>>>   * Next section "Zeroing of the ACE Field", 2nd para:
>>>     "If the value of this ACE field is zero (0b000), the Data Sender
>>>     disables sending ECN-capable packets for the remainder of the
>>>     half-connection by setting the IP/ECN field in all subsequent
>>>     packets to Not-ECT, *but**it**SHOULD continue to respond to ECN
>>>     feedback even though **it **is no longer sending ECN-capable
>>>     packets.* *It MUST also remain in the same feedback mode and it
>>>     MUST continue to feed back any ECN markings on arriving packets
>>>     (in its role as Data Receiver).*"
>>>
>>> I'll also add reasoning (in the zeroing section, it already says that
>>> ACE=0b000 is not an unambiguous indication of mangling).
>>>
>>> And this prompts me to edit the bullet in an earlier section about the
>>> obligation of a sender to respond to congestion feedback:
>>>
>>>
>>>         3.1.5.
>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.1.5>Implications
>>>         of AccECN Mode
>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-implications-of-accecn-mode>
>>>
>>>
>>>     "It is still obliged to respond appropriately to AccECN feedback
>>>     that indicates there were ECN marks on *ECN-capable *packets it
>>>     had previously sent."
>>>
>>> And I'll add a bullet:
>>>
>>>     "*If the sender chooses not to send ECN-capable packets (e.g.
>>>     because path traversal of the ECN field is suspect), it can ignore
>>>     any **ECN **feedback about those packets if it is certain that it
>>>     cannot be valid (see Section 3.2.2, which gives normative
>>>     requirements for certain specific cases).*"
>>>
>>> I've said 'can' rather than 'MUST' because it's hard to cover all
>>> cases, e.g. single packets without ECT when it's not clear whether the
>>> feedback covered another packet that was ECT.
>>>
>>> How does all this sound?
>>>
>>>> 2. On the reverse path from server to client, if a server sends a
>>>> Not-ECT SYN-ACK and receives ACE handshake encoding on last ACK other
>>>> than 0 1 0, there is no text like above that says server should send
>>>> Not-ECT for the rest of the connection (or at least I didn’t find
>>>> it). I think the server should also do same as client as the two
>>>> paths could be different. One could make it more complicated by
>>>> saying, if both client and server decide to not use ECN on their
>>>> corresponding sender half, then ECN should be disabled but let’s talk
>>>> about that later.
>>>
>>> [BB] It's in the second para of:
>>>
>>>
>>>           3.2.2.3.
>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3>Testing
>>>           for Mangling of the IP/ECN Field
>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the>
>>>
>>>
>>>> 3. In general, not setting the ECT (ECT0 or ECT1) code point on an
>>>> outgoing packet is different from supporting AccECN right as in the
>>>> host can still provide AccECN feedback on the receive path.
>>>
>>> [BB] Yup.
>>>
>>> Do you think any further explanation is needed in the above sections?
>>>
>>>
>>> Bob
>>>
>>>> ….
>>>>
>>>> To be continued if more questions come to my mind.
>>>>
>>>>
>>>> Thanks,
>>>> Vidhi
>>>>
>>>>> On Feb 3, 2022, at 7:24 AM, Bob Briscoe <ietf@bobbriscoe.net> wrote:
>>>>>
>>>>> tcpm folks,
>>>>>
>>>>> This rev to accurate-ecn is the first of two. The second will
>>>>> hopefully follow on its heels in the next couple of days.
>>>>>
>>>>> Diffs between this first rev (-16) and -15:
>>>>>
>>>>> 1.  switches round two fairly large sections, so I've deferred other
>>>>> changes to a second rev so the diffs won't be masked by the switch
>>>>> round of sections.
>>>>> Suggested by Ilpo to match the order in which the tests in these
>>>>> sections will be executed:
>>>>> * Test for mangling the IP-ECN field (now 3.2.2.3),
>>>>> * Then test for zeroing the ACE field (now 3.2.2.4).
>>>>>
>>>>> 2. Ilpo suggested some clarifications in "3.2.3.2.5. Consistency
>>>>> between AccECN Feedback Fields", which is about the receiver of
>>>>> feedback ensuring consistency between the mandatory 3-bit ACE field
>>>>> and the optional 24-bit counters. In brief (paraphrasing) it
>>>>> previously only said "MUST consider both fields", when it is now
>>>>> clearer that it actually meant "MUST reconcile both fields", so that
>>>>> there is always a consistent baseline for subsequent ACKs.
>>>>>
>>>>> 3. A minor point is added in an appendix about the details that the
>>>>> pseudocode doesn't cover.
>>>>>
>>>>>
>>>>>
>>>>> Bob
>>>>>
>>>>> PS. #2 & #3 were added to the XML ages ago (Jul '21), so you will
>>>>> have seen them in the HTML. However, prob due to my clumsiness, the
>>>>> posted TXT didn't include them whereas the posted XML did (ironic
>>>>> for a section about consistency). In turn, inconsistency was only
>>>>> possible because I am having to manually post the TXT for this
>>>>> draft, due to an unresolved issue with v3 RFC XML tables.
>>>>>
>>>>>
>>>>> On 03/02/2022 14:43, 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 TCP Maintenance and Minor
>>>>>> Extensions WG of the IETF.
>>>>>>
>>>>>>         Title           : More Accurate ECN Feedback in TCP
>>>>>>         Authors         : Bob Briscoe
>>>>>>                           Mirja Kühlewind
>>>>>>                           Richard Scheffenegger
>>>>>> Filename        : draft-ietf-tcpm-accurate-ecn-16.txt
>>>>>> Pages           : 60
>>>>>> Date            : 2022-02-03
>>>>>>
>>>>>> Abstract:
>>>>>>    Explicit Congestion Notification (ECN) is a mechanism where 
>>>>>> network
>>>>>>    nodes can mark IP packets instead of dropping them to indicate
>>>>>>    incipient congestion to the end-points.  Receivers with an ECN-
>>>>>>    capable transport protocol feed back this information to the 
>>>>>> sender.
>>>>>>    ECN was originally specified for TCP in such a way that only one
>>>>>>    feedback signal can be transmitted per Round-Trip Time (RTT).
>>>>>>  Recent
>>>>>>    new TCP mechanisms like Congestion Exposure (ConEx), Data 
>>>>>> Center TCP
>>>>>>    (DCTCP) or Low Latency Low Loss Scalable Throughput (L4S) need 
>>>>>> more
>>>>>>    accurate ECN feedback information whenever more than one 
>>>>>> marking is
>>>>>>    received in one RTT.  This document specifies a scheme to provide
>>>>>>    more than one feedback signal per RTT in the TCP header. 
>>>>>>  Given TCP
>>>>>>    header space is scarce, it allocates a reserved header bit
>>>>>> previously
>>>>>>    assigned to the ECN-Nonce.  It also overloads the two existing 
>>>>>> ECN
>>>>>>    flags in the TCP header.  The resulting extra space is 
>>>>>> exploited to
>>>>>>    feed back the IP-ECN field received during the 3-way handshake as
>>>>>>    well.  Supplementary feedback information can optionally be 
>>>>>> provided
>>>>>>    in a new TCP option, which is never used on the TCP SYN.  The
>>>>>>    document also specifies the treatment of this updated TCP wire
>>>>>>    protocol by middleboxes.
>>>>>>
>>>>>>
>>>>>> The IETF datatracker status page for this draft is:
>>>>>> https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/
>>>>>>
>>>>>> There is also an htmlized version available at:
>>>>>> https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16 
>>>>>>
>>>>>>
>>>>>> A diff from the previous version is available at:
>>>>>> https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16
>>>>>>
>>>>>>
>>>>>> Internet-Drafts are also available by rsync at rsync.ietf.org
>>>>>> <http://rsync.ietf.org>::internet-drafts
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> tcpm mailing list
>>>>>> tcpm@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>>
>>>>> -- 
>>>>> ________________________________________________________________
>>>>> Bob Briscoe http://bobbriscoe.net/
>>>>>
>>>>> _______________________________________________
>>>>> tcpm mailing list
>>>>> tcpm@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>
>>>
>>> -- 
>>> ________________________________________________________________
>>> Bob Briscoehttp://bobbriscoe.net/
>>
>>
>> _______________________________________________
>> tcpm mailing list
>> tcpm@ietf.org
>> https://www.ietf.org/mailman/listinfo/tcpm

-- 
________________________________________________________________
Bob Briscoehttp://bobbriscoe.net/

--------------UO067S0NPkEoC6gOEK0ScEW1
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    Richard,<br>
    <br>
    Indeed, I recently discovered I had missed an item that I added to
    my ToDo list in Mar 2021, based on a conversation between Gorry and
    Mirja where they both made the same point - that fall-back behaviour
    after discovering ECN mangling ought to be non-normative in this
    draft, because it's beyond the scope of the new feedback mechanism:<br>
<a class="moz-txt-link-freetext" href="https://mailarchive.ietf.org/arch/msg/tcpm/00URm1uwptgrAzKq96T6KSh8dQ4/">https://mailarchive.ietf.org/arch/msg/tcpm/00URm1uwptgrAzKq96T6KSh8dQ4/</a><br>
    <br>
    We do need to say the following normatively in "Testing for Mangling
    of the IP/ECN Field" because they are relevant to feedback, the Data
    Sender:<br>
    * remains in AccECN mode <br>
    * MUST continue to feed back any ECN markings on arriving packets.<br>
    <br>
    BTW, Vidhi, the text you quoted was from §2 on "Protocol Overview
    and Rationale". That was meant to be an example of why it is useful
    for the handshake to give feedback on exactly which IP-ECN codepoint
    arrived. It wasn't meant to be a full spec. So I'll make it clearer
    that it's only an example, rather than try to make it comprehensive,
    which isn't appropriate at that point in the doc.<br>
    <br>
    Nonetheless, your points are correct, but if they are dealt with
    anyway, they should be dealt with in §3.2.2.3. "Testing for Mangling
    of the IP/ECN Field" where these scenarios are already covered
    (except for the congestion response, which was silently left out of
    scope):<br>
<a class="moz-txt-link-freetext" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3">https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3</a>
    <br>
    <br>
    We could just say that the questions of whether to set ECT for
    subsequent packets, and whether to respond to feedback of CE
    markings are beyond the scope of the draft. I think we should, but I
    think we should also give some non-normative advice on what to do.
    It can only be interim advice, not normative requirements. Both
    because it's out of scope, and because the best behaviour and the
    alternatives will depend on deployment experience. So they shouldn't
    be written normatively in a Proposed Standard yet. <br>
    <br>
    We will do what the WG wants, but assuming it agrees to including
    non-normative advice, following from your (Richard's) list of cases,
    here's a full matrix. I don't propose this matrix should go in the
    AccECN draft - it's just for this email, to check every case is
    covered. <br>
    <br>
    <table width="737" height="293" cellspacing="0" border="0">
      <colgroup width="85" span="5"></colgroup> <tbody>
        <tr>
          <td height="17" align="left"><br>
          </td>
          <th colspan="4" valign="middle" align="center">Feedback in
            SYN/ACK of SYN's IP-ECN </th>
        </tr>
        <tr>
          <th height="17" align="left">IP-ECN on SYN</th>
          <th align="left">Not-ECT</th>
          <th align="left">ECT(0)</th>
          <th align="left">ECT(1)</th>
          <th align="left">CE</th>
        </tr>
        <tr>
          <th height="17" align="left">Not ECT</th>
          <td align="left">√</td>
          <td align="left">NR</td>
          <td align="left">NR</td>
          <td align="left">NR</td>
        </tr>
        <tr>
          <th height="17" align="left">ECT(0)</th>
          <td align="left">ND</td>
          <td align="left">√</td>
          <td align="left">√</td>
          <td align="left">√</td>
        </tr>
        <tr>
          <th height="17" align="left">ECT(1)</th>
          <td align="left">ND</td>
          <td align="left">CU<br>
          </td>
          <td align="left">√</td>
          <td align="left">√</td>
        </tr>
        <tr>
          <th height="17" align="left">CE</th>
          <td align="left">n/a</td>
          <td align="left">n/a</td>
          <td align="left">n/a</td>
          <td align="left">n/a</td>
        </tr>
      </tbody>
    </table>
    <style type="text/css">body,div,table,thead,tbody,tfoot,tr,th,td,p { font-family:"Liberation Sans"; font-size:x-small }a.comment-indicator:hover + comment { background:#ffd; position:absolute; display:block; border:1px solid black; padding:0.5em;  }a.comment-indicator { background:red; display:inline-block; border:1px solid black; width:0.5em; height:0.5em;  }comment { display:none;  }</style><br>
    Key:<br>
    √       Valid response<br>
    NR    Non-normative advice: Set <b>N</b>ot-ECT for subsequent
    packets, but continue to <b>R</b>espond to CE feedback<br>
    ND    Non-normative advice: Set <b>N</b>ot-ECT for subsequent
    packets, and <b>D</b>isable response to CE feedback<br>
    CU    Currently unused transition (not listed as invalid in the
    draft)<br>
    n/a    not applicable in the AccECN spec, in which there is never a
    case where the sender sets CE when sending a packet.<br>
    <br>
    In the draft we don't need to mention any cases where CE is sent
    (but a draft on fall-back probably would, because the sender might
    set CE to test the path). We can cover the other cases with two
    groups of invalid transitions: those where the SYN (or SYN-ACK)
    arriving at the receiver appears to have been <br>
    i) ECN-capable<br>
    ii) Not-ECN-capable.<br>
    <br>
    I propose to also say that details (like what 'subsequent packets'
    implies about how long this behaviour continues) are for future
    study.<br>
    <br>
    I'll try to write some text and propose it to this list. Perhaps it
    should be in a new appendix?<br>
    <br>
    <br>
    Bob<br>
    <br>
    <div class="moz-cite-prefix">On 05/02/2022 12:01, Scheffenegger,
      Richard wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:1cbe3dbe-241b-3f24-096f-f1a2ffe17d98@gmx.at">Vidhi,
      <br>
      <br>
      Thanks for reading the draft very closely!
      <br>
      <br>
      Well, in the non-ECT SYN case, where the SYNACK indicates an ECT
      (ect0,
      <br>
      ect1, ce) is the obvious - and fully semantically correct - use
      case of
      <br>
      that possible use.
      <br>
      <br>
      For the set of three codepoints (ECT0, ECT1, CE) the only other
      possible
      <br>
      valid detection scenario would be if a SYN+CE is sent
      (deliberately,
      <br>
      althought one can argue that a sender should not originate
      CE-marked
      <br>
      packets), but a SYNACK without CE is returned (full or partial
      <br>
      bleaching). At least under the current RFC3168 definition of IP
      ECN
      <br>
      codepoints.
      <br>
      <br>
      allowed transitions:
      <br>
      nonECT -&gt; nonECT
      <br>
      <br>
      ECT0 -&gt; ECT0
      <br>
      ECT0 -&gt; ECT1
      <br>
      ECT0 -&gt; CE
      <br>
      <br>
      ECT1 -&gt; ECT1
      <br>
      ECT1 -&gt; ECT0 (possibly)
      <br>
      ECT1 -&gt; CE
      <br>
      <br>
      CE -&gt; CE
      <br>
      <br>
      <br>
      Thinking about this aspect more, I am now unsure, if the higher
      level
      <br>
      functions, such as detection of IP ECN codepoint mangling, or
      validating
      <br>
      across multiple RTTs, should be in this document - or not rather
      be in a
      <br>
      separate draft, in order to facilitate independent implementation
      and
      <br>
      improvement of mechanisms in that space.
      <br>
      <br>
      However, the above table of allowed IP ECN transistions may be
      updated
      <br>
      independently of AccECN - and performing extensive checks as part
      of a
      <br>
      signal protocol ossifies this IMHO.
      <br>
      <br>
      As a middle ground, perhaps having these descriptions in a
      non-normative
      <br>
      section, clearly stating these to be examples of what an
      impemented
      <br>
      could choose to do?
      <br>
      <br>
      Richard
      <br>
      <br>
      Am 05.02.2022 um 02:14 schrieb Vidhi Goel:
      <br>
      <blockquote type="cite">I like your response for all the points. I
        can review the diff for the
        <br>
        proposed changes, if you’d like before committing them to the
        draft.
        <br>
        <br>
        Sorry, I want to add a new comment for this text,
        <br>
        <br>
        If a TCP client has set the SYN to Not-ECT, but
        <br>
            receives feedback that the IP-ECN field on the SYN arrived
        with a
        <br>
            different codepoint, it can detect such middlebox
        interference and
        <br>
            send Not-ECT for the rest of the connection.
        <br>
        <br>
        <br>
        This statement holds good for Not-ECT but doesn’t if lets say
        the SYN
        <br>
        was ECT0 or ECT1 as those could be marked CE and still be
        valid. Should
        <br>
        we add a statement for ECT marked SYN as well?
        <br>
        <br>
        Thanks,
        <br>
        Vidhi
        <br>
        <br>
        <blockquote type="cite">On Feb 4, 2022, at 5:44 AM, Bob Briscoe
          &lt;<a class="moz-txt-link-abbreviated" href="mailto:ietf@bobbriscoe.net">ietf@bobbriscoe.net</a>
          <br>
          <a class="moz-txt-link-rfc2396E" href="mailto:ietf@bobbriscoe.net">&lt;mailto:ietf@bobbriscoe.net&gt;</a>&gt; wrote:
          <br>
          <br>
          Vidhi,
          <br>
          <br>
          I've increased the table of contents to tocdepth = "4" which
          might
          <br>
          help find the relevant sections better, because a lot of the
          meat in
          <br>
          this draft is under sections 3.2.2 (ACE) and 3.2.3 (Option).
          <br>
          <br>
          I've responded below, but I'd prefer it if you would say these
          things
          <br>
          on the list, so we have a better justification for changing
          the draft.
          <br>
          Reviewing a draft that has just be re-posted shouldn't imply
          anything
          <br>
          about whether Apple is implementing the draft or not.
          <br>
          <br>
          (BTW, regarding the proposed change to the initial value of
          r.e1b, I
          <br>
          was going to send that  to the list once co-authors agreed to
          it. But
          <br>
          it would be nice if some of the edits were not
          self-generated.)
          <br>
          <br>
          See [BB] inline for responses (which I'll repeat if you send
          to the list).
          <br>
          <br>
          On 04/02/2022 07:37, Vidhi Goel wrote:
          <br>
          <blockquote type="cite">Hello Bob and authors,
            <br>
            <br>
            I have read the draft a few times and will send my extensive
            review
            <br>
            at a later point. But right now, I want to ask some critical
            things
            <br>
            that an implementation could benefit from.
            <br>
            <br>
            The draft says,
            <br>
            /If a TCP client has set the SYN to Not-ECT, but/
            <br>
            /   receives feedback that the IP-ECN field on the SYN
            arrived with a/
            <br>
            /   different codepoint, it can detect such middlebox
            interference and/
            <br>
            /   send Not-ECT for the rest of the connection/
            <br>
            <br>
            1. On the forward path from client to server, the client
            will revert
            <br>
            to Not-ECT when it sees for example, that it sent a Non-ECT
            SYN but
            <br>
            received ACE encoding other than 0 1 0. What does the client
            do in
            <br>
            the last ACK of the 3WHS -
            <br>
             a. does it stay in AccECN mode and still send AccECN
            encoding based
            <br>
            on the IP code point of SYN-ACK (Table 4)? This would mean
            that
            <br>
            client won’t participate in ECN on the sender half of the
            connection
            <br>
            and only provide AccECN feedback to the server as a
            receiver.
            <br>
             b.  does it disable AccECN mode and set ACE= 0 0 0 so that
            a server
            <br>
            in AccECN mode can disable ECN based on Table4?
            <br>
            <br>
            While writing this, I realized that the intention
            is probably a.
            <br>
            Could you confirm? Also, when the sender sets Not-ECT in its
            data
            <br>
            packets, it should also disable acting upon any ACE feedback
            as we
            <br>
            could still receive false ACE feedback from the server
            if the
            <br>
            network, lets say, changed 00 (not ECT) to 01(ECT1). If you
            agree, we
            <br>
            should add some text around this. Based on the current text,
            the
            <br>
            sender will always respond to the ACE feedback even if it
            sends Not-ECT.
            <br>
            <br>
            TBH, this is a complicated scenario, where sender said to
            the network
            <br>
            - I don’t trust you so I can’t use ECN. Feel free to drop my
            packets.
            <br>
            And the network mangles the IP to ECT1 and then set CE (when
            <br>
            congested) which would be feedback’ed from the receiver.
            Now, the
            <br>
            packet wasn’t dropped which it should have been. So, is it
            better to
            <br>
            just ignore this feedback because sender doesn’t trust the
            network or
            <br>
            just act on it and reduce cwnd in order to reduce congestion
            in the
            <br>
            network somewhere.
            <br>
          </blockquote>
          <br>
          [BB] Your points are all good ones.
          <br>
          I'll address this last complex scenario first, because it has
          <br>
          implications for the earlier questions. First I'll define some
          <br>
          terminology:
          <br>
          <br>
          Simple mangling scenario:
          <br>
          <br>
            * some network function, e.g. broken Diffserv ToS-byte
          remarking,
          <br>
              illegally remarks a Not-ECT  SYN to CE.
          <br>
          <br>
          Your complex mangling scenario, repeated here:
          <br>
          <br>
            * some network function, e.g. broken Diffserv ToS-byte
          remarking,
          <br>
              illegally remarks a Not-ECT SYN to ECT0 or ECT1,
          <br>
            * then congestion at a subsequent bottleneck is marking some
          packets CE
          <br>
          <br>
          Then potentially the client could tailor its behaviour after
          sending a
          <br>
          Not-ECT SYN, If the SYN/ACK feedback is
          <br>
          <br>
           1. ECT: disable sending ECT, but continue responding to ECN
          feedback
          <br>
           2. CE, disable sending ECT, and disable response to ECN
          feedback
          <br>
          <br>
          However, this is uncertain, because CE feedback on the SYN/ACK
          could
          <br>
          indicate either the simple or the complex mangling scenario.
          <br>
          <br>
          A simpler alternative would be to always continue responding
          to ECN
          <br>
          feedback. Rationale:
          <br>
          <br>
            * Whether case #1 or #2, assume that mangling of the SYN
          might have
          <br>
              been Not-ECT to ECT, even if the feedback off the SYN is
          CE.
          <br>
            * Then as the connection progresses, if /all/ feedback is
          CE,
          <br>
              there's already a recommendation to fall-back to disabling
          <br>
              congestion response.
          <br>
          <br>
          If we do this, I think we ought to say "SHOULD continue to
          respond to
          <br>
          ECN feedback", not "MUST".
          <br>
          <br>
          And we'll need to put this all to the WG.
          <br>
          <br>
          1.a) Now back to the beginning of your point. The text you
          quote is
          <br>
          from §2.5 which is in the non-normative "Overview and
          Rationale"
          <br>
          section (§2).
          <br>
          You really need the normative text from:
          <br>
          <br>
          <br>
                    3.2.2.3.
          <br>
                   
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3&gt;</a>Testing<br>
                    for Mangling of the IP/ECN Field
          <br>
                   
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the&gt;</a><br>
          <br>
          <br>
          where it says (end of 1st para):
          <br>
          <br>
              " ...for the remainder of the connection the client MUST
          NOT send
          <br>
              ECN-capable packets, but it MUST continue to feed back any
          ECN
          <br>
              markings on arriving packets."
          <br>
          <br>
          So, you're right, the answer is (a).
          <br>
          It can't be (b), because, once in AccECN mode, 000 on the ACE
          field
          <br>
          just becomes a counter value, and no longer negotiates the
          feedback mode.
          <br>
          <br>
          1.b) As per the above, let's conservatively assume complex
          mangling.
          <br>
          So, at the end of the first para quoted above I suggest we
          add:
          <br>
          <br>
              " ...for the remainder of the connection the client MUST
          NOT send
          <br>
              ECN-capable packets *but**it**SHOULD continue to respond
          to ECN
          <br>
              feedback even though **it **is no longer sending
          ECN-capable
          <br>
              packets (see reasoning below)**. T**he client**MUST remain
          in the
          <br>
              same feedback mode and***it MUST continue to feed back any
          ECN
          <br>
              markings on arriving packets."
          <br>
          <br>
          I'll do the same for next para about the server.
          <br>
          I'll work out some text for the reasoning, but I won't give it
          here.
          <br>
          <br>
          There are two other cases where it disables sending ECT, which
          don't
          <br>
          say whether it continues to respond to congestion:
          <br>
          <br>
            * "Testing for Mangling" section, penultimate para,  where
          it's
          <br>
              receiving solid CE:
          <br>
              "Once a Data Sender has entered AccECN mode it SHOULD
          check
          <br>
              whether all feedback received for the first three or four
          rounds
          <br>
              indicated that every packet it sent was CE-marked. If so,
          for the
          <br>
              remainder of the connection, the Data Sender SHOULD NOT
          send
          <br>
              ECN-capable packets*and it SHOULD NOT respond to ECN
          feedback*,
          <br>
              but *it MUST remain in the same feedback mode and *it MUST
          <br>
              continue to feed back any ECN markings on arriving
          packets*(in its
          <br>
              role as Data Receiver)*."
          <br>
            * Next section "Zeroing of the ACE Field", 2nd para:
          <br>
              "If the value of this ACE field is zero (0b000), the Data
          Sender
          <br>
              disables sending ECN-capable packets for the remainder of
          the
          <br>
              half-connection by setting the IP/ECN field in all
          subsequent
          <br>
              packets to Not-ECT, *but**it**SHOULD continue to respond
          to ECN
          <br>
              feedback even though **it **is no longer sending
          ECN-capable
          <br>
              packets.* *It MUST also remain in the same feedback mode
          and it
          <br>
              MUST continue to feed back any ECN markings on arriving
          packets
          <br>
              (in its role as Data Receiver).*"
          <br>
          <br>
          I'll also add reasoning (in the zeroing section, it already
          says that
          <br>
          ACE=0b000 is not an unambiguous indication of mangling).
          <br>
          <br>
          And this prompts me to edit the bullet in an earlier section
          about the
          <br>
          obligation of a sender to respond to congestion feedback:
          <br>
          <br>
          <br>
                  3.1.5.
          <br>
                 
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.1.5">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.1.5&gt;</a>Implications<br>
                  of AccECN Mode
          <br>
                 
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-implications-of-accecn-mode">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-implications-of-accecn-mode&gt;</a><br>
          <br>
          <br>
              "It is still obliged to respond appropriately to AccECN
          feedback
          <br>
              that indicates there were ECN marks on *ECN-capable
          *packets it
          <br>
              had previously sent."
          <br>
          <br>
          And I'll add a bullet:
          <br>
          <br>
              "*If the sender chooses not to send ECN-capable packets
          (e.g.
          <br>
              because path traversal of the ECN field is suspect), it
          can ignore
          <br>
              any **ECN **feedback about those packets if it is certain
          that it
          <br>
              cannot be valid (see Section 3.2.2, which gives normative
          <br>
              requirements for certain specific cases).*"
          <br>
          <br>
          I've said 'can' rather than 'MUST' because it's hard to cover
          all
          <br>
          cases, e.g. single packets without ECT when it's not clear
          whether the
          <br>
          feedback covered another packet that was ECT.
          <br>
          <br>
          How does all this sound?
          <br>
          <br>
          <blockquote type="cite">2. On the reverse path from server to
            client, if a server sends a
            <br>
            Not-ECT SYN-ACK and receives ACE handshake encoding on last
            ACK other
            <br>
            than 0 1 0, there is no text like above that says server
            should send
            <br>
            Not-ECT for the rest of the connection (or at least I didn’t
            find
            <br>
            it). I think the server should also do same as client as the
            two
            <br>
            paths could be different. One could make it more complicated
            by
            <br>
            saying, if both client and server decide to not use ECN on
            their
            <br>
            corresponding sender half, then ECN should be disabled but
            let’s talk
            <br>
            about that later.
            <br>
          </blockquote>
          <br>
          [BB] It's in the second para of:
          <br>
          <br>
          <br>
                    3.2.2.3.
          <br>
                   
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3&gt;</a>Testing<br>
                    for Mangling of the IP/ECN Field
          <br>
                   
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the&gt;</a><br>
          <br>
          <br>
          <blockquote type="cite">3. In general, not setting the ECT
            (ECT0 or ECT1) code point on an
            <br>
            outgoing packet is different from supporting AccECN right as
            in the
            <br>
            host can still provide AccECN feedback on the receive path.
            <br>
          </blockquote>
          <br>
          [BB] Yup.
          <br>
          <br>
          Do you think any further explanation is needed in the above
          sections?
          <br>
          <br>
          <br>
          Bob
          <br>
          <br>
          <blockquote type="cite">….
            <br>
            <br>
            To be continued if more questions come to my mind.
            <br>
            <br>
            <br>
            Thanks,
            <br>
            Vidhi
            <br>
            <br>
            <blockquote type="cite">On Feb 3, 2022, at 7:24 AM, Bob
              Briscoe <a class="moz-txt-link-rfc2396E" href="mailto:ietf@bobbriscoe.net">&lt;ietf@bobbriscoe.net&gt;</a> wrote:
              <br>
              <br>
              tcpm folks,
              <br>
              <br>
              This rev to accurate-ecn is the first of two. The second
              will
              <br>
              hopefully follow on its heels in the next couple of days.
              <br>
              <br>
              Diffs between this first rev (-16) and -15:
              <br>
              <br>
              1.  switches round two fairly large sections, so I've
              deferred other
              <br>
              changes to a second rev so the diffs won't be masked by
              the switch
              <br>
              round of sections.
              <br>
              Suggested by Ilpo to match the order in which the tests in
              these
              <br>
              sections will be executed:
              <br>
              * Test for mangling the IP-ECN field (now 3.2.2.3),
              <br>
              * Then test for zeroing the ACE field (now 3.2.2.4).
              <br>
              <br>
              2. Ilpo suggested some clarifications in "3.2.3.2.5.
              Consistency
              <br>
              between AccECN Feedback Fields", which is about the
              receiver of
              <br>
              feedback ensuring consistency between the mandatory 3-bit
              ACE field
              <br>
              and the optional 24-bit counters. In brief (paraphrasing)
              it
              <br>
              previously only said "MUST consider both fields", when it
              is now
              <br>
              clearer that it actually meant "MUST reconcile both
              fields", so that
              <br>
              there is always a consistent baseline for subsequent ACKs.
              <br>
              <br>
              3. A minor point is added in an appendix about the details
              that the
              <br>
              pseudocode doesn't cover.
              <br>
              <br>
              <br>
              <br>
              Bob
              <br>
              <br>
              PS. #2 &amp; #3 were added to the XML ages ago (Jul '21),
              so you will
              <br>
              have seen them in the HTML. However, prob due to my
              clumsiness, the
              <br>
              posted TXT didn't include them whereas the posted XML did
              (ironic
              <br>
              for a section about consistency). In turn, inconsistency
              was only
              <br>
              possible because I am having to manually post the TXT for
              this
              <br>
              draft, due to an unresolved issue with v3 RFC XML tables.
              <br>
              <br>
              <br>
              On 03/02/2022 14:43, <a class="moz-txt-link-abbreviated" href="mailto:internet-drafts@ietf.org">internet-drafts@ietf.org</a> wrote:
              <br>
              <blockquote type="cite">A New Internet-Draft is available
                from the on-line Internet-Drafts
                <br>
                directories.
                <br>
                This draft is a work item of the TCP Maintenance and
                Minor
                <br>
                Extensions WG of the IETF.
                <br>
                <br>
                        Title           : More Accurate ECN Feedback in
                TCP
                <br>
                        Authors         : Bob Briscoe
                <br>
                                          Mirja Kühlewind
                <br>
                                          Richard Scheffenegger
                <br>
                Filename        : draft-ietf-tcpm-accurate-ecn-16.txt
                <br>
                Pages           : 60
                <br>
                Date            : 2022-02-03
                <br>
                <br>
                Abstract:
                <br>
                   Explicit Congestion Notification (ECN) is a mechanism
                where network
                <br>
                   nodes can mark IP packets instead of dropping them to
                indicate
                <br>
                   incipient congestion to the end-points.  Receivers
                with an ECN-
                <br>
                   capable transport protocol feed back this information
                to the sender.
                <br>
                   ECN was originally specified for TCP in such a way
                that only one
                <br>
                   feedback signal can be transmitted per Round-Trip
                Time (RTT).
                <br>
                 Recent
                <br>
                   new TCP mechanisms like Congestion Exposure (ConEx),
                Data Center TCP
                <br>
                   (DCTCP) or Low Latency Low Loss Scalable Throughput
                (L4S) need more
                <br>
                   accurate ECN feedback information whenever more than
                one marking is
                <br>
                   received in one RTT.  This document specifies a
                scheme to provide
                <br>
                   more than one feedback signal per RTT in the TCP
                header.  Given TCP
                <br>
                   header space is scarce, it allocates a reserved
                header bit
                <br>
                previously
                <br>
                   assigned to the ECN-Nonce.  It also overloads the two
                existing ECN
                <br>
                   flags in the TCP header.  The resulting extra space
                is exploited to
                <br>
                   feed back the IP-ECN field received during the 3-way
                handshake as
                <br>
                   well.  Supplementary feedback information can
                optionally be provided
                <br>
                   in a new TCP option, which is never used on the TCP
                SYN.  The
                <br>
                   document also specifies the treatment of this updated
                TCP wire
                <br>
                   protocol by middleboxes.
                <br>
                <br>
                <br>
                The IETF datatracker status page for this draft is:
                <br>
<a class="moz-txt-link-freetext" href="https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/">https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/</a>
                <br>
                <br>
                There is also an htmlized version available at:
                <br>
<a class="moz-txt-link-freetext" href="https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16">https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16</a>
                <br>
                <br>
                A diff from the previous version is available at:
                <br>
<a class="moz-txt-link-freetext" href="https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16">https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16</a>
                <br>
                <br>
                <br>
                Internet-Drafts are also available by rsync at
                rsync.ietf.org
                <br>
                <a class="moz-txt-link-rfc2396E" href="http://rsync.ietf.org">&lt;http://rsync.ietf.org&gt;</a>::internet-drafts
                <br>
                <br>
                <br>
                _______________________________________________
                <br>
                tcpm mailing list
                <br>
                <a class="moz-txt-link-abbreviated" href="mailto:tcpm@ietf.org">tcpm@ietf.org</a>
                <br>
                <a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/tcpm">https://www.ietf.org/mailman/listinfo/tcpm</a>
                <br>
              </blockquote>
              <br>
              --
              <br>
________________________________________________________________
              <br>
              Bob Briscoe <a class="moz-txt-link-freetext" href="http://bobbriscoe.net/">http://bobbriscoe.net/</a>
              <br>
              <br>
              _______________________________________________
              <br>
              tcpm mailing list
              <br>
              <a class="moz-txt-link-abbreviated" href="mailto:tcpm@ietf.org">tcpm@ietf.org</a>
              <br>
              <a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/tcpm">https://www.ietf.org/mailman/listinfo/tcpm</a>
              <br>
            </blockquote>
            <br>
          </blockquote>
          <br>
          --
          <br>
________________________________________________________________
          <br>
          Bob Briscoehttp://bobbriscoe.net/
          <br>
        </blockquote>
        <br>
        <br>
        _______________________________________________
        <br>
        tcpm mailing list
        <br>
        <a class="moz-txt-link-abbreviated" href="mailto:tcpm@ietf.org">tcpm@ietf.org</a>
        <br>
        <a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/tcpm">https://www.ietf.org/mailman/listinfo/tcpm</a>
        <br>
      </blockquote>
    </blockquote>
    <br>
    <pre class="moz-signature" cols="72">-- 
________________________________________________________________
Bob Briscoe                               <a class="moz-txt-link-freetext" href="http://bobbriscoe.net/">http://bobbriscoe.net/</a></pre>
  </body>
</html>

--------------UO067S0NPkEoC6gOEK0ScEW1--


From nobody Sun Feb  6 15:01:56 2022
Return-Path: <rs.ietf@gmx.at>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B67083A0EDC; Sun,  6 Feb 2022 15:01:52 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.612
X-Spam-Level: 
X-Spam-Status: No, score=-2.612 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, FREEMAIL_FROM=0.001, NICE_REPLY_A=-0.714, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=gmx.net
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id F7BxRrT3GGh5; Sun,  6 Feb 2022 15:01:47 -0800 (PST)
Received: from mout.gmx.net (mout.gmx.net [212.227.17.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id DE8C83A0EDB; Sun,  6 Feb 2022 15:01:46 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=gmx.net; s=badeba3b8450; t=1644188456; bh=vXJ7BSXkcjG4+ygreHeBMFuH4mDV1+/YRHwdlsIz1Ns=; h=X-UI-Sender-Class:Date:Subject:To:Cc:References:From:In-Reply-To; b=kQ341pB8BzO+xbKF0mF4ydNTOK090/mjgr1SCrBAXqOYWLTa8QQ9VKA+RCM6HurxR jyk8NGHCPEYQNrH65+mGnOIiGbUouk3FFSmaKVX1o5I1ODbzKt48tPZVFgBd9gbUWw s4PiG0AWg+tCYbUqjj83ZBjNsA4GN6WLl2apDjB8=
X-UI-Sender-Class: 01bb95c1-4bf8-414a-932a-4f6e2808ef9c
Received: from [192.168.233.112] ([185.236.167.136]) by mail.gmx.net (mrgmx105 [212.227.17.168]) with ESMTPSA (Nemesis) id 1MY68T-1npBBd226H-00YToa; Mon, 07 Feb 2022 00:00:56 +0100
Message-ID: <24dd6cbd-6668-23e2-6902-9772d0a1ef9a@gmx.at>
Date: Mon, 7 Feb 2022 00:00:52 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Thunderbird/91.5.1
To: Bob Briscoe <ietf@bobbriscoe.net>, Vidhi Goel <vidhi_goel=40apple.com@dmarc.ietf.org>
Cc: Richard Scheffenegger <rscheff@gmx.at>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, Mirja Kuehlewind <ietf@kuehlewind.net>
References: <164389942954.13556.7754569279919863072@ietfa.amsl.com> <c2c5bdf9-6b34-fe19-3296-5f1cd831ed8e@bobbriscoe.net> <2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com> <e57355ae-5a9d-0a14-9478-29453d25622c@bobbriscoe.net> <BB23E172-C995-41FE-A47E-E02D5586B67D@apple.com> <1cbe3dbe-241b-3f24-096f-f1a2ffe17d98@gmx.at> <615549a4-6d09-8a7b-f892-42a1ea52c2e2@bobbriscoe.net>
From: "Scheffenegger, Richard" <rs.ietf@gmx.at>
In-Reply-To: <615549a4-6d09-8a7b-f892-42a1ea52c2e2@bobbriscoe.net>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: quoted-printable
X-Provags-ID: V03:K1:7Rlj3Xb94u63W8Al9yWh1Vr0XNXzrBo0RuPyW2bpaS936nMig7j Kg6U0frhhArkGxTJ9IzNSTH5swRbC9BmV1qgIoxdMH00QI6PhIitA5wZwG80hONbVC7hJpn iMR5WkVwVs5km1EEn0ms4GP9KuNgJ/qR6Y83hmiOEhNq6r0T+KFgts/HUnGfrlVCpSwxS+w eopz86x0mselO12LpzUxQ==
X-UI-Out-Filterresults: notjunk:1;V03:K0:gKeCFyY7Jtk=:aQ9g+Jxe5yW5xlayQdnsPD 4mhmHwviOrJcNzk0PWQ/EzxQK+qNo0mVAxedfVKj2IejuocrM9QatFtdQAHGcWGomLw3q1HoM Tt1bHY2V8wB7Wls87g7ojzCxVcZJDot4b46NjL9ioIvNSGrerk0Hqj+UVi3PNI5nKfVNcVqyy heK8nSsvzV+saOJL4pN0C2G1lBDesQwyYcFLJN0ZclIIVNPxCUxVsn98h0R3dYVzFb2iq1MVP cs3lZQPm+nJAJmMaYzdFAQGsLXt1AFl71HEvaY7uJGzvLSXBFcVBMlJnraM1vZEyV+6DnQcbg vKcOiy5FaANw9IvpRiEq0YhVZUFl6qE97kxsjxyQlRXQPAmUHTBKpq5YaNoUA+Z28mSy7zw+9 j/VVSJkiV2730jb2HCj17LEzvGKOU1S6onAfVAivy6HhkHQZPogyJ51nlEcUNIvZbg8CG8Gf6 OJv0gi8OxcdsGJpSkdCkhCboPdRWvrk6Uou9NEPIs28FIroqNoA5OSnyMGXs0MxhIIjjadDWE lvlIdzwRy1aR094w3q5DBi9OvqOLRl4IKj252p8oSFtWkKSVPBiwhpxstD7hxHtsOdWKw3KKM Edsvsc8pJ4lK4ZvxovfZOsmFHkTIO9F7WZuAS1kobMvXMoNI1BuGCvbnd6zAMfncB8ffbzxW/ ckBCQaOcG6YPTzebXjhs156bF1qdzmnZoUXASi0DRzZCT+sjuiCwZPi/AfvqQC/+2rN3Ai/ds hCBMGFb/ICLUxHx5fEZdQHL8KvQBbyrQiIuDib4tOrBA2GqMNan56JGxxmc7KQj2H0JZnjbmM BZhwoOKlQFW6ecwj/kfMAsfEaItW3XxcTklCtXKt6Xr45F6oeujCH2G+Fa6Zg+AqHp3dhQlm9 7PURnwuMX1okGxWrb3cKCfhJ6EBn3TFTL2kI+GMKS1R2SVZxgjthPQQsJrRj1UWgiJSo6uoFY rOvXLSZBgCPySYHHOeT3iYCZGJ6iDGIVfzwoyGfnmvPfDlrh9kyXWcxxUrE8aU9PMtMTNVBTd 8H5jZEcg15RRmqIwtxLicPt3yHmrtFyhK79q/Ey7zjuTmUwWOxaGkhy7tG8Je3TyqnFb4fsWE /EFdYbiaXGHyJI=
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/-Jk78ZM9xBUNR3vgqKoM4nq49eE>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-accurate-ecn-16.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 06 Feb 2022 23:01:53 -0000

Hi Bob,

could it be, that the first "data" row and column in this table are
parially transposed?

In the case of sending a Non-ECT SYN, but getting an indication of a
received CE on the SYN-ACK, why would you respond to that indication of
an incorrect CE marking? (With the response being, that the cwnd would
probably never grow beyond 1 or 2 packets per RTT, or possibly diminish
to n RTTs per packet, one a stack supports such low rates).

I am thinking of the most common probable cause - a middlebox
unconditionally overriding the full TOS byte, and simply setting both
rightmost bits continously.

Such a path would be rendered utterly unusable for an AccECN session, not?

So at least that case would be a ND in your nomenclature.

In the column - if the IP ECN is bleached, why disable response to CE
(if that is unlike to come back, as long as the session traverses the
bleaching path)? A "NR" seems more logical here in case of path changes?

But again, all these ins-and-outs should be discussed in a different draft=
.

I fully agree, that only the normative response of the AccECN feedback
mechanism has to be fully defined - the two bullet point is exactly what
I expect to be stated.

Regards,

    Richard


Am 06.02.2022 um 23:27 schrieb Bob Briscoe:
> Richard,
>
> Indeed, I recently discovered I had missed an item that I added to my
> ToDo list in Mar 2021, based on a conversation between Gorry and Mirja
> where they both made the same point - that fall-back behaviour after
> discovering ECN mangling ought to be non-normative in this draft,
> because it's beyond the scope of the new feedback mechanism:
> https://mailarchive.ietf.org/arch/msg/tcpm/00URm1uwptgrAzKq96T6KSh8dQ4/
>
> We do need to say the following normatively in "Testing for Mangling of
> the IP/ECN Field" because they are relevant to feedback, the Data Sender=
:
> * remains in AccECN mode
> * MUST continue to feed back any ECN markings on arriving packets.
>
> BTW, Vidhi, the text you quoted was from =C2=A72 on "Protocol Overview a=
nd
> Rationale". That was meant to be an example of why it is useful for the
> handshake to give feedback on exactly which IP-ECN codepoint arrived. It
> wasn't meant to be a full spec. So I'll make it clearer that it's only
> an example, rather than try to make it comprehensive, which isn't
> appropriate at that point in the doc.
>
> Nonetheless, your points are correct, but if they are dealt with anyway,
> they should be dealt with in =C2=A73.2.2.3. "Testing for Mangling of the
> IP/ECN Field" where these scenarios are already covered (except for the
> congestion response, which was silently left out of scope):
> https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#sec=
tion-3.2.2.3
>
>
> We could just say that the questions of whether to set ECT for
> subsequent packets, and whether to respond to feedback of CE markings
> are beyond the scope of the draft. I think we should, but I think we
> should also give some non-normative advice on what to do. It can only be
> interim advice, not normative requirements. Both because it's out of
> scope, and because the best behaviour and the alternatives will depend
> on deployment experience. So they shouldn't be written normatively in a
> Proposed Standard yet.
>
> We will do what the WG wants, but assuming it agrees to including
> non-normative advice, following from your (Richard's) list of cases,
> here's a full matrix. I don't propose this matrix should go in the
> AccECN draft - it's just for this email, to check every case is covered.
>
>
> 	Feedback in SYN/ACK of SYN's IP-ECN
> IP-ECN on SYN 	Not-ECT 	ECT(0) 	ECT(1) 	CE
> Not ECT 	=E2=88=9A 	NR 	NR 	NR
> ECT(0) 	ND 	=E2=88=9A 	=E2=88=9A 	=E2=88=9A
> ECT(1) 	ND 	CU
> 	=E2=88=9A 	=E2=88=9A
> CE 	n/a 	n/a 	n/a 	n/a
>
>
> Key:
> =E2=88=9A=C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0 Valid response
> NR=C2=A0=C2=A0=C2=A0 Non-normative advice: Set *N*ot-ECT for subsequent =
packets, but
> continue to *R*espond to CE feedback
> ND=C2=A0=C2=A0=C2=A0 Non-normative advice: Set *N*ot-ECT for subsequent =
packets, and
> *D*isable response to CE feedback
> CU =C2=A0=C2=A0 Currently unused transition (not listed as invalid in th=
e draft)
> n/a=C2=A0=C2=A0=C2=A0 not applicable in the AccECN spec, in which there =
is never a case
> where the sender sets CE when sending a packet.
>
> In the draft we don't need to mention any cases where CE is sent (but a
> draft on fall-back probably would, because the sender might set CE to
> test the path). We can cover the other cases with two groups of invalid
> transitions: those where the SYN (or SYN-ACK) arriving at the receiver
> appears to have been
> i) ECN-capable
> ii) Not-ECN-capable.
>
> I propose to also say that details (like what 'subsequent packets'
> implies about how long this behaviour continues) are for future study.
>
> I'll try to write some text and propose it to this list. Perhaps it
> should be in a new appendix?
>
>
> Bob
>
> On 05/02/2022 12:01, Scheffenegger, Richard wrote:
>> Vidhi,
>>
>> Thanks for reading the draft very closely!
>>
>> Well, in the non-ECT SYN case, where the SYNACK indicates an ECT (ect0,
>> ect1, ce) is the obvious - and fully semantically correct - use case of
>> that possible use.
>>
>> For the set of three codepoints (ECT0, ECT1, CE) the only other possibl=
e
>> valid detection scenario would be if a SYN+CE is sent (deliberately,
>> althought one can argue that a sender should not originate CE-marked
>> packets), but a SYNACK without CE is returned (full or partial
>> bleaching). At least under the current RFC3168 definition of IP ECN
>> codepoints.
>>
>> allowed transitions:
>> nonECT -> nonECT
>>
>> ECT0 -> ECT0
>> ECT0 -> ECT1
>> ECT0 -> CE
>>
>> ECT1 -> ECT1
>> ECT1 -> ECT0 (possibly)
>> ECT1 -> CE
>>
>> CE -> CE
>>
>>
>> Thinking about this aspect more, I am now unsure, if the higher level
>> functions, such as detection of IP ECN codepoint mangling, or validatin=
g
>> across multiple RTTs, should be in this document - or not rather be in =
a
>> separate draft, in order to facilitate independent implementation and
>> improvement of mechanisms in that space.
>>
>> However, the above table of allowed IP ECN transistions may be updated
>> independently of AccECN - and performing extensive checks as part of a
>> signal protocol ossifies this IMHO.
>>
>> As a middle ground, perhaps having these descriptions in a non-normativ=
e
>> section, clearly stating these to be examples of what an impemented
>> could choose to do?
>>
>> Richard
>>
>> Am 05.02.2022 um 02:14 schrieb Vidhi Goel:
>>> I like your response for all the points. I can review the diff for the
>>> proposed changes, if you=E2=80=99d like before committing them to the =
draft.
>>>
>>> Sorry, I want to add a new comment for this text,
>>>
>>> If a TCP client has set the SYN to Not-ECT, but
>>> =C2=A0=C2=A0=C2=A0 receives feedback that the IP-ECN field on the SYN =
arrived with a
>>> =C2=A0=C2=A0=C2=A0 different codepoint, it can detect such middlebox i=
nterference and
>>> =C2=A0=C2=A0=C2=A0 send Not-ECT for the rest of the connection.
>>>
>>>
>>> This statement holds good for Not-ECT but doesn=E2=80=99t if lets say =
the SYN
>>> was ECT0 or ECT1=C2=A0as those could be marked CE and still be valid.=
=C2=A0Should
>>> we add a statement for ECT marked SYN as well?
>>>
>>> Thanks,
>>> Vidhi
>>>
>>>> On Feb 4, 2022, at 5:44 AM, Bob Briscoe <ietf@bobbriscoe.net
>>>> <mailto:ietf@bobbriscoe.net>> wrote:
>>>>
>>>> Vidhi,
>>>>
>>>> I've increased the table of contents to tocdepth =3D "4" which might
>>>> help find the relevant sections better, because a lot of the meat in
>>>> this draft is under sections 3.2.2 (ACE) and 3.2.3 (Option).
>>>>
>>>> I've responded below, but I'd prefer it if you would say these things
>>>> on the list, so we have a better justification for changing the draft=
.
>>>> Reviewing a draft that has just be re-posted shouldn't imply anything
>>>> about whether Apple is implementing the draft or not.
>>>>
>>>> (BTW, regarding the proposed change to the initial value of r.e1b, I
>>>> was going to send that=C2=A0 to the list once co-authors agreed to it=
. But
>>>> it would be nice if some of the edits were not self-generated.)
>>>>
>>>> See [BB] inline for responses (which I'll repeat if you send to the
>>>> list).
>>>>
>>>> On 04/02/2022 07:37, Vidhi Goel wrote:
>>>>> Hello Bob and authors,
>>>>>
>>>>> I have read the draft a few times and will send my extensive review
>>>>> at a later point. But right now, I want to ask some critical things
>>>>> that an implementation could benefit from.
>>>>>
>>>>> The draft says,
>>>>> /If a TCP client has set the SYN to Not-ECT, but/
>>>>> /=C2=A0=C2=A0 receives feedback that the IP-ECN field on the SYN arr=
ived with a/
>>>>> /=C2=A0=C2=A0 different codepoint, it can detect such middlebox inte=
rference
>>>>> and/
>>>>> /=C2=A0=C2=A0 send Not-ECT for the rest of the connection/
>>>>>
>>>>> 1. On the forward path from client to server, the client will revert
>>>>> to Not-ECT when it sees for example, that it sent a Non-ECT SYN but
>>>>> received ACE encoding other than 0 1 0.=C2=A0What does the client do=
 in
>>>>> the last ACK of the 3WHS -
>>>>> =C2=A0a. does it stay in AccECN mode and still send AccECN encoding =
based
>>>>> on the IP=C2=A0code point of SYN-ACK (Table 4)? This would mean that
>>>>> client won=E2=80=99t participate in ECN on the sender half of the co=
nnection
>>>>> and only=C2=A0provide AccECN feedback to the server as a receiver.
>>>>> =C2=A0b. =C2=A0does it disable AccECN mode and set ACE=3D 0 0 0 so t=
hat a server
>>>>> in AccECN mode can disable ECN based on Table4?
>>>>>
>>>>> While writing this, I realized that the intention is=C2=A0probably a=
.
>>>>> Could you confirm? Also, when the sender sets Not-ECT in its data
>>>>> packets, it should also disable acting upon any ACE feedback as we
>>>>> could still receive false ACE feedback from the server if=C2=A0the
>>>>> network, lets say, changed 00 (not ECT) to 01(ECT1). If you agree, w=
e
>>>>> should add some text around this. Based on the current text, the
>>>>> sender will always respond to the ACE feedback even if it sends
>>>>> Not-ECT.
>>>>>
>>>>> TBH, this is a complicated scenario, where sender said to the networ=
k
>>>>> - I don=E2=80=99t trust you so I can=E2=80=99t use ECN. Feel=C2=A0fr=
ee to drop my packets.
>>>>> And the network mangles the IP to ECT1=C2=A0and then set CE (when
>>>>> congested)=C2=A0which would be feedback=E2=80=99ed from the receiver=
. Now, the
>>>>> packet wasn=E2=80=99t dropped which it should have been. So, is it b=
etter to
>>>>> just ignore this feedback because sender doesn=E2=80=99t trust the n=
etwork or
>>>>> just act on it and reduce cwnd in order to reduce congestion in the
>>>>> network somewhere.
>>>>
>>>> [BB] Your points are all good ones.
>>>> I'll address this last complex scenario first, because it has
>>>> implications for the earlier questions. First I'll define some
>>>> terminology:
>>>>
>>>> Simple mangling scenario:
>>>>
>>>> =C2=A0 * some network function, e.g. broken Diffserv ToS-byte remarki=
ng,
>>>> =C2=A0=C2=A0=C2=A0 illegally remarks a Not-ECT=C2=A0 SYN to CE.
>>>>
>>>> Your complex mangling scenario, repeated here:
>>>>
>>>> =C2=A0 * some network function, e.g. broken Diffserv ToS-byte remarki=
ng,
>>>> =C2=A0=C2=A0=C2=A0 illegally remarks a Not-ECT SYN to ECT0 or ECT1,
>>>> =C2=A0 * then congestion at a subsequent bottleneck is marking some
>>>> packets CE
>>>>
>>>> Then potentially the client could tailor its behaviour after sending =
a
>>>> Not-ECT SYN, If the SYN/ACK feedback is
>>>>
>>>> =C2=A01. ECT: disable sending ECT, but continue responding to ECN fee=
dback
>>>> =C2=A02. CE, disable sending ECT, and disable response to ECN feedbac=
k
>>>>
>>>> However, this is uncertain, because CE feedback on the SYN/ACK could
>>>> indicate either the simple or the complex mangling scenario.
>>>>
>>>> A simpler alternative would be to always continue responding to ECN
>>>> feedback. Rationale:
>>>>
>>>> =C2=A0 * Whether case #1 or #2, assume that mangling of the SYN might=
 have
>>>> =C2=A0=C2=A0=C2=A0 been Not-ECT to ECT, even if the feedback off the =
SYN is CE.
>>>> =C2=A0 * Then as the connection progresses, if /all/ feedback is CE,
>>>> =C2=A0=C2=A0=C2=A0 there's already a recommendation to fall-back to d=
isabling
>>>> =C2=A0=C2=A0=C2=A0 congestion response.
>>>>
>>>> If we do this, I think we ought to say "SHOULD continue to respond to
>>>> ECN feedback", not "MUST".
>>>>
>>>> And we'll need to put this all to the WG.
>>>>
>>>> 1.a) Now back to the beginning of your point. The text you quote is
>>>> from =C2=A72.5 which is in the non-normative "Overview and Rationale"
>>>> section (=C2=A72).
>>>> You really need the normative text from:
>>>>
>>>>
>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 3.2.2.3.
>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html=
#section-3.2.2.3>Testing
>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 for Mangling o=
f the IP/ECN Field
>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html=
#name-testing-for-mangling-of-the>
>>>>
>>>>
>>>> where it says (end of 1st para):
>>>>
>>>> =C2=A0=C2=A0=C2=A0 " ...for the remainder of the connection the clien=
t MUST NOT send
>>>> =C2=A0=C2=A0=C2=A0 ECN-capable packets, but it MUST continue to feed =
back any ECN
>>>> =C2=A0=C2=A0=C2=A0 markings on arriving packets."
>>>>
>>>> So, you're right, the answer is (a).
>>>> It can't be (b), because, once in AccECN mode, 000 on the ACE field
>>>> just becomes a counter value, and no longer negotiates the feedback
>>>> mode.
>>>>
>>>> 1.b) As per the above, let's conservatively assume complex mangling.
>>>> So, at the end of the first para quoted above I suggest we add:
>>>>
>>>> =C2=A0=C2=A0=C2=A0 " ...for the remainder of the connection the clien=
t MUST NOT send
>>>> =C2=A0=C2=A0=C2=A0 ECN-capable packets *but**it**SHOULD continue to r=
espond to ECN
>>>> =C2=A0=C2=A0=C2=A0 feedback even though **it **is no longer sending E=
CN-capable
>>>> =C2=A0=C2=A0=C2=A0 packets (see reasoning below)**. T**he client**MUS=
T remain in the
>>>> =C2=A0=C2=A0=C2=A0 same feedback mode and***it MUST continue to feed =
back any ECN
>>>> =C2=A0=C2=A0=C2=A0 markings on arriving packets."
>>>>
>>>> I'll do the same for next para about the server.
>>>> I'll work out some text for the reasoning, but I won't give it here.
>>>>
>>>> There are two other cases where it disables sending ECT, which don't
>>>> say whether it continues to respond to congestion:
>>>>
>>>> =C2=A0 * "Testing for Mangling" section, penultimate para,=C2=A0 wher=
e it's
>>>> =C2=A0=C2=A0=C2=A0 receiving solid CE:
>>>> =C2=A0=C2=A0=C2=A0 "Once a Data Sender has entered AccECN mode it SHO=
ULD check
>>>> =C2=A0=C2=A0=C2=A0 whether all feedback received for the first three =
or four rounds
>>>> =C2=A0=C2=A0=C2=A0 indicated that every packet it sent was CE-marked.=
 If so, for the
>>>> =C2=A0=C2=A0=C2=A0 remainder of the connection, the Data Sender SHOUL=
D NOT send
>>>> =C2=A0=C2=A0=C2=A0 ECN-capable packets*and it SHOULD NOT respond to E=
CN feedback*,
>>>> =C2=A0=C2=A0=C2=A0 but *it MUST remain in the same feedback mode and =
*it MUST
>>>> =C2=A0=C2=A0=C2=A0 continue to feed back any ECN markings on arriving=
 packets*(in its
>>>> =C2=A0=C2=A0=C2=A0 role as Data Receiver)*."
>>>> =C2=A0 * Next section "Zeroing of the ACE Field", 2nd para:
>>>> =C2=A0=C2=A0=C2=A0 "If the value of this ACE field is zero (0b000), t=
he Data Sender
>>>> =C2=A0=C2=A0=C2=A0 disables sending ECN-capable packets for the remai=
nder of the
>>>> =C2=A0=C2=A0=C2=A0 half-connection by setting the IP/ECN field in all=
 subsequent
>>>> =C2=A0=C2=A0=C2=A0 packets to Not-ECT, *but**it**SHOULD continue to r=
espond to ECN
>>>> =C2=A0=C2=A0=C2=A0 feedback even though **it **is no longer sending E=
CN-capable
>>>> =C2=A0=C2=A0=C2=A0 packets.* *It MUST also remain in the same feedbac=
k mode and it
>>>> =C2=A0=C2=A0=C2=A0 MUST continue to feed back any ECN markings on arr=
iving packets
>>>> =C2=A0=C2=A0=C2=A0 (in its role as Data Receiver).*"
>>>>
>>>> I'll also add reasoning (in the zeroing section, it already says that
>>>> ACE=3D0b000 is not an unambiguous indication of mangling).
>>>>
>>>> And this prompts me to edit the bullet in an earlier section about th=
e
>>>> obligation of a sender to respond to congestion feedback:
>>>>
>>>>
>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 3.1.5.
>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html=
#section-3.1.5>Implications
>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 of AccECN Mode
>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html=
#name-implications-of-accecn-mode>
>>>>
>>>>
>>>> =C2=A0=C2=A0=C2=A0 "It is still obliged to respond appropriately to A=
ccECN feedback
>>>> =C2=A0=C2=A0=C2=A0 that indicates there were ECN marks on *ECN-capabl=
e *packets it
>>>> =C2=A0=C2=A0=C2=A0 had previously sent."
>>>>
>>>> And I'll add a bullet:
>>>>
>>>> =C2=A0=C2=A0=C2=A0 "*If the sender chooses not to send ECN-capable pa=
ckets (e.g.
>>>> =C2=A0=C2=A0=C2=A0 because path traversal of the ECN field is suspect=
), it can ignore
>>>> =C2=A0=C2=A0=C2=A0 any **ECN **feedback about those packets if it is =
certain that it
>>>> =C2=A0=C2=A0=C2=A0 cannot be valid (see Section 3.2.2, which gives no=
rmative
>>>> =C2=A0=C2=A0=C2=A0 requirements for certain specific cases).*"
>>>>
>>>> I've said 'can' rather than 'MUST' because it's hard to cover all
>>>> cases, e.g. single packets without ECT when it's not clear whether th=
e
>>>> feedback covered another packet that was ECT.
>>>>
>>>> How does all this sound?
>>>>
>>>>> 2. On the reverse path=C2=A0from server to client, if a server sends=
 a
>>>>> Not-ECT SYN-ACK and receives ACE handshake encoding on last ACK=C2=
=A0other
>>>>> than 0 1 0, there is no text like above that says server should send
>>>>> Not-ECT for the rest of the connection (or=C2=A0at least I=C2=A0didn=
=E2=80=99t find
>>>>> it). I think the server should also do same as client as=C2=A0the tw=
o
>>>>> paths could be different. One could make it more complicated by
>>>>> saying, if both client and server decide to not use ECN on their
>>>>> corresponding sender half, then ECN should be disabled but let=E2=80=
=99s talk
>>>>> about that later.
>>>>
>>>> [BB] It's in the second para of:
>>>>
>>>>
>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 3.2.2.3.
>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html=
#section-3.2.2.3>Testing
>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 for Mangling o=
f the IP/ECN Field
>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html=
#name-testing-for-mangling-of-the>
>>>>
>>>>
>>>>> 3. In general, not setting the ECT (ECT0 or ECT1)=C2=A0code point on=
 an
>>>>> outgoing packet is different from supporting AccECN right as in the
>>>>> host can still provide AccECN feedback on the=C2=A0receive path.
>>>>
>>>> [BB] Yup.
>>>>
>>>> Do you think any further explanation is needed in the above sections?
>>>>
>>>>
>>>> Bob
>>>>
>>>>> =E2=80=A6.
>>>>>
>>>>> To be continued if more questions come to my mind.
>>>>>
>>>>>
>>>>> Thanks,
>>>>> Vidhi
>>>>>
>>>>>> On Feb 3, 2022, at 7:24 AM, Bob Briscoe <ietf@bobbriscoe.net> wrote=
:
>>>>>>
>>>>>> tcpm folks,
>>>>>>
>>>>>> This rev to accurate-ecn is the first of two. The second will
>>>>>> hopefully follow on its heels in the next couple of days.
>>>>>>
>>>>>> Diffs between this first rev (-16) and -15:
>>>>>>
>>>>>> 1.=C2=A0 switches round two fairly large sections, so I've deferred=
 other
>>>>>> changes to a second rev so the diffs won't be masked by the switch
>>>>>> round of sections.
>>>>>> Suggested by Ilpo to match the order in which the tests in these
>>>>>> sections will be executed:
>>>>>> * Test for mangling the IP-ECN field (now 3.2.2.3),
>>>>>> * Then test for zeroing the ACE field (now 3.2.2.4).
>>>>>>
>>>>>> 2. Ilpo suggested some clarifications in "3.2.3.2.5. Consistency
>>>>>> between AccECN Feedback Fields", which is about the receiver of
>>>>>> feedback ensuring consistency between the mandatory 3-bit ACE field
>>>>>> and the optional 24-bit counters. In brief (paraphrasing) it
>>>>>> previously only said "MUST consider both fields", when it is now
>>>>>> clearer that it actually meant "MUST reconcile both fields", so tha=
t
>>>>>> there is always a consistent baseline for subsequent ACKs.
>>>>>>
>>>>>> 3. A minor point is added in an appendix about the details that the
>>>>>> pseudocode doesn't cover.
>>>>>>
>>>>>>
>>>>>>
>>>>>> Bob
>>>>>>
>>>>>> PS. #2 & #3 were added to the XML ages ago (Jul '21), so you will
>>>>>> have seen them in the HTML. However, prob due to my clumsiness, the
>>>>>> posted TXT didn't include them whereas the posted XML did (ironic
>>>>>> for a section about consistency). In turn, inconsistency was only
>>>>>> possible because I am having to manually post the TXT for this
>>>>>> draft, due to an unresolved issue with v3 RFC XML tables.
>>>>>>
>>>>>>
>>>>>> On 03/02/2022 14:43, 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 TCP Maintenance and Minor
>>>>>>> Extensions WG of the IETF.
>>>>>>>
>>>>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Title =C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0: More Accurate ECN Feedba=
ck in TCP
>>>>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Authors =C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0: Bob Briscoe
>>>>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0Mirja K=C3=BChlewind
>>>>>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0Richard Scheffenegger
>>>>>>> Filename =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0: draft-ietf-tc=
pm-accurate-ecn-16.txt
>>>>>>> Pages =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
: 60
>>>>>>> Date =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0: 2022-02-03
>>>>>>>
>>>>>>> Abstract:
>>>>>>> =C2=A0=C2=A0=C2=A0Explicit Congestion Notification (ECN) is a mech=
anism where
>>>>>>> network
>>>>>>> =C2=A0=C2=A0=C2=A0nodes can mark IP packets instead of dropping th=
em to indicate
>>>>>>> =C2=A0=C2=A0=C2=A0incipient congestion to the end-points. =C2=A0Re=
ceivers with an ECN-
>>>>>>> =C2=A0=C2=A0=C2=A0capable transport protocol feed back this inform=
ation to the
>>>>>>> sender.
>>>>>>> =C2=A0=C2=A0=C2=A0ECN was originally specified for TCP in such a w=
ay that only one
>>>>>>> =C2=A0=C2=A0=C2=A0feedback signal can be transmitted per Round-Tri=
p Time (RTT).
>>>>>>> =C2=A0Recent
>>>>>>> =C2=A0=C2=A0=C2=A0new TCP mechanisms like Congestion Exposure (Con=
Ex), Data
>>>>>>> Center TCP
>>>>>>> =C2=A0=C2=A0=C2=A0(DCTCP) or Low Latency Low Loss Scalable Through=
put (L4S) need
>>>>>>> more
>>>>>>> =C2=A0=C2=A0=C2=A0accurate ECN feedback information whenever more =
than one
>>>>>>> marking is
>>>>>>> =C2=A0=C2=A0=C2=A0received in one RTT. =C2=A0This document specifi=
es a scheme to provide
>>>>>>> =C2=A0=C2=A0=C2=A0more than one feedback signal per RTT in the TCP=
 header.
>>>>>>> =C2=A0Given TCP
>>>>>>> =C2=A0=C2=A0=C2=A0header space is scarce, it allocates a reserved =
header bit
>>>>>>> previously
>>>>>>> =C2=A0=C2=A0=C2=A0assigned to the ECN-Nonce. =C2=A0It also overloa=
ds the two existing
>>>>>>> ECN
>>>>>>> =C2=A0=C2=A0=C2=A0flags in the TCP header. =C2=A0The resulting ext=
ra space is
>>>>>>> exploited to
>>>>>>> =C2=A0=C2=A0=C2=A0feed back the IP-ECN field received during the 3=
-way handshake as
>>>>>>> =C2=A0=C2=A0=C2=A0well. =C2=A0Supplementary feedback information c=
an optionally be
>>>>>>> provided
>>>>>>> =C2=A0=C2=A0=C2=A0in a new TCP option, which is never used on the =
TCP SYN. =C2=A0The
>>>>>>> =C2=A0=C2=A0=C2=A0document also specifies the treatment of this up=
dated TCP wire
>>>>>>> =C2=A0=C2=A0=C2=A0protocol by middleboxes.
>>>>>>>
>>>>>>>
>>>>>>> The IETF datatracker status page for this draft is:
>>>>>>> https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/
>>>>>>>
>>>>>>> There is also an htmlized version available at:
>>>>>>> https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn=
-16
>>>>>>>
>>>>>>>
>>>>>>> A diff from the previous version is available at:
>>>>>>> https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-accurate-ecn-1=
6
>>>>>>>
>>>>>>>
>>>>>>> Internet-Drafts are also available by rsync at rsync.ietf.org
>>>>>>> <http://rsync.ietf.org>::internet-drafts
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> tcpm mailing list
>>>>>>> tcpm@ietf.org
>>>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>>>
>>>>>> --
>>>>>> ________________________________________________________________
>>>>>> Bob Briscoe http://bobbriscoe.net/
>>>>>>
>>>>>> _______________________________________________
>>>>>> tcpm mailing list
>>>>>> tcpm@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>>
>>>>
>>>> --
>>>> ________________________________________________________________
>>>> Bob Briscoehttp://bobbriscoe.net/
>>>
>>>
>>> _______________________________________________
>>> tcpm mailing list
>>> tcpm@ietf.org
>>> https://www.ietf.org/mailman/listinfo/tcpm
>
> --
> ________________________________________________________________
> Bob Briscoehttp://bobbriscoe.net/
>


From nobody Mon Feb  7 15:32:50 2022
Return-Path: <vidhi_goel@apple.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5F5223A0F09 for <tcpm@ietfa.amsl.com>; Mon,  7 Feb 2022 15:32:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.675
X-Spam-Level: 
X-Spam-Status: No, score=-2.675 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.576, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=apple.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id dunv2lK0Mesz for <tcpm@ietfa.amsl.com>; Mon,  7 Feb 2022 15:32:43 -0800 (PST)
Received: from rn-mailsvcp-ppex-lapp34.apple.com (rn-mailsvcp-ppex-lapp34.rno.apple.com [17.179.253.43]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 723983A0F2E for <tcpm@ietf.org>; Mon,  7 Feb 2022 15:32:43 -0800 (PST)
Received: from pps.filterd (rn-mailsvcp-ppex-lapp34.rno.apple.com [127.0.0.1]) by rn-mailsvcp-ppex-lapp34.rno.apple.com (8.16.1.2/8.16.1.2) with SMTP id 217NFYBr010547; Mon, 7 Feb 2022 15:32:41 -0800
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=apple.com; h=from : message-id : content-type : mime-version : subject : date : in-reply-to : cc : to : references; s=20180706; bh=i73DVC4qbwqACy5VBSk9tvvVtJvlSr6pfq7XHWbZcPA=; b=UyuLOa/rl13pXrOPxvev6+MJOTlWfpKn5+kv4Olm2S2B2ATpipK6JYrqpOloyFzEzzqV zgmBpNLt0Hcue1RDoQuYh4bnQsVp1dvL6f+gcXo6/dT5+aItS5tPr1xSxFiT7Xnn2rR0 dg0U4X6al7Uj9fXux8Yb+b3ftnZaxp48Y5IjfGJmUqIoUghVcW8b07gKNNX7donl4dhc 8UxRQPsjtuDZ8AaFho2Fw3Nl8s62Vk0WA4a6e9v7bezuAOQAv6ZDA2/cQgW6mXUIE5yS OMuorAzP8FTjLd4CGwaT5SGQMtS0qVeVS6T+hzBgR+QQuzTkXId1o9TlX6IO1DSYlvuV hQ== 
Received: from rn-mailsvcp-mta-lapp01.rno.apple.com (rn-mailsvcp-mta-lapp01.rno.apple.com [10.225.203.149]) by rn-mailsvcp-ppex-lapp34.rno.apple.com with ESMTP id 3e26b3we9t-2 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NO); Mon, 07 Feb 2022 15:32:41 -0800
Received: from rn-mailsvcp-mmp-lapp03.rno.apple.com (rn-mailsvcp-mmp-lapp03.rno.apple.com [17.179.253.16]) by rn-mailsvcp-mta-lapp01.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPS id <0R6Y004M2K2HMP30@rn-mailsvcp-mta-lapp01.rno.apple.com>;  Mon, 07 Feb 2022 15:32:41 -0800 (PST)
Received: from process_milters-daemon.rn-mailsvcp-mmp-lapp03.rno.apple.com by rn-mailsvcp-mmp-lapp03.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) id <0R6Y00M00JY8GL00@rn-mailsvcp-mmp-lapp03.rno.apple.com>; Mon, 07 Feb 2022 15:32:41 -0800 (PST)
X-Va-A: 
X-Va-T-CD: 201b35371ed429fb4e4fc3ed10b3bd6f
X-Va-E-CD: 749c6e7503a1c30456497a36735b331a
X-Va-R-CD: 59a455eaeb25ffef4add4d7c725dcf79
X-Va-CD: 0
X-Va-ID: 0f6a47c3-8b75-41e0-bf63-ebc77df376f3
X-V-A: 
X-V-T-CD: 201b35371ed429fb4e4fc3ed10b3bd6f
X-V-E-CD: 749c6e7503a1c30456497a36735b331a
X-V-R-CD: 59a455eaeb25ffef4add4d7c725dcf79
X-V-CD: 0
X-V-ID: 12f0c5f6-74e8-416a-8fda-188b19408506
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-07_07:2022-02-07, 2022-02-07 signatures=0
Received: from smtpclient.apple (unknown [17.234.14.73]) by rn-mailsvcp-mmp-lapp03.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPSA id <0R6Y00WKEK2GGM00@rn-mailsvcp-mmp-lapp03.rno.apple.com>; Mon, 07 Feb 2022 15:32:41 -0800 (PST)
From: Vidhi Goel <vidhi_goel@apple.com>
Message-id: <492FC264-2D3C-4D3D-B7C2-D547593E4A7D@apple.com>
Content-type: multipart/signed; boundary="Apple-Mail=_45C2D04E-78BF-4827-A84D-52F7F454FFD1"; protocol="application/pkcs7-signature"; micalg=sha-256
MIME-version: 1.0 (Mac OS X Mail 14.0 \(3654.100.0.2.11\))
Date: Mon, 07 Feb 2022 15:32:39 -0800
In-reply-to: <CAAK044R6B7HT0mdSDwHU=FhvN=A=M-8pxYb8wGfwxczwtCT-yw@mail.gmail.com>
Cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>
To: Yoshifumi Nishida <nsd.ietf@gmail.com>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <CAAK044R6B7HT0mdSDwHU=FhvN=A=M-8pxYb8wGfwxczwtCT-yw@mail.gmail.com>
X-Mailer: Apple Mail (2.3654.100.0.2.11)
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-07_07:2022-02-07, 2022-02-07 signatures=0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/vGlc5AML1jX01Ugg08QOh8dXaAI>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 07 Feb 2022 23:32:48 -0000

--Apple-Mail=_45C2D04E-78BF-4827-A84D-52F7F454FFD1
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_C451B4F0-F4F0-4693-BF86-9465A6AD561C"


--Apple-Mail=_C451B4F0-F4F0-4693-BF86-9465A6AD561C
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

Thanks Yoshi for reviewing the draft.

> So, it seems one PS doc mentions not to use cwnd, while another PS doc =
mentions using cwnd is fine (with RFC7661).
RFC5681 is conservative in its approach to use FlightSize and will =
result in very low throughput in scenarios when the congestion event =
occurs at the tail of send window (when flight size could be very small)

> I am not sure which can cause this situation yet, but in any case, =
does this mean there is no concern on=20
> "some implementations may incidentally increase well beyond rwnd" in =
cubic?
> Is this because this is an old implementation bug or are there other =
reasons?
> I am wondering if we might want some clarification on this point.
If an implementation uses RFC7661, there is no concern about increasing =
cwnd beyond rwnd as RFC7661 tracks bytes acknowledged in a RTT which =
means that `rwnd` check has already been performed.
Usually cwnd doesn=E2=80=99t grow beyond rwnd  (ACK-clocking increases =
cwnd which means the sender already checked 	`rwnd` before sending =
out the data).

But there could be implementations that may want to keep `rwnd` even =
smaller than initial `cwnd`. In those cases, RFC7661 would make the cwnd =
invalidated when the number of acknowledged bytes is much lower than the =
current cwnd.

Do you have any suggestions on how to make it more clear?

> Also, I am wondering if we might want to use "MAY be used" here =
instead of "can be used" =20
Yeah, we can make this change.

I have opened an issue in GitHub =
https://github.com/NTAP/rfc8312bis/issues/143 =
<https://github.com/NTAP/rfc8312bis/issues/143>. We can continue the =
discussion there.


Thanks,
Vidhi

> On Feb 6, 2022, at 3:01 AM, Yoshifumi Nishida <nsd.ietf@gmail.com> =
wrote:
>=20
> Hi,
> I just have one more comment on the draft.
> In RFC5681, it mentions in Section 3:
> ''
>    Implementation Note: An easy mistake to make is to simply use cwnd,
>    rather than FlightSize, which in some implementations may
>    incidentally increase well beyond rwnd.
> "
>=20
> while this draft mentions in Section 4.6:
> "
>    To avoid suboptimal performance with such applications,
>    the mechanisms described in [RFC7661 =
<https://datatracker.ietf.org/doc/html/rfc7661>] can be used to mitigate =
this
>    issue as it would allow using a value between _cwnd_ and
>    _flight_size_ to calculate the new _ssthresh_ in Figure 5.=20
> "
> So, it seems one PS doc mentions not to use cwnd, while another PS doc =
mentions using cwnd is fine (with RFC7661).
>=20
> I am not sure which can cause this situation yet, but in any case, =
does this mean there is no concern on=20
> "some implementations may incidentally increase well beyond rwnd" in =
cubic?
> Is this because this is an old implementation bug or are there other =
reasons?
> I am wondering if we might want some clarification on this point.
> =20
> Also, I am wondering if we might want to use "MAY be used" here =
instead of "can be used" =20
>=20
> Thanks,
> --
> Yoshi
>=20
> On Mon, Jan 31, 2022 at 11:43 PM Yoshifumi Nishida <nsd.ietf@gmail.com =
<mailto:nsd.ietf@gmail.com>> wrote:
> Hello,
>=20
> After some discussions among chairs, we decided to run the 2nd WGLC on =
draft-ietf-tcpm-rfc8312bis in consideration of the importance of the =
draft.=20
> We'll be grateful if you could send your feedback to the ML. The WGLC =
runs until *Feb 11*.
>=20
> If interested, you can check in-depth past discussions in the =
following URL.
> https://github.com/NTAP/rfc8312bis/ =
<https://github.com/NTAP/rfc8312bis/>
>=20
> Thank you so much!
> --
> tcpm co-chairs
>=20
>=20
> On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <lars@eggert.org =
<mailto:lars@eggert.org>> wrote:
> Hi,
>=20
> this -06 version rolls in all the changes requested during (and after) =
WGLC ended.
>=20
> I'll leave it up to the chairs to decide if another WGLC is warranted =
or the document can progress as-is.
>=20
> Thanks,
> Lars
>=20
>=20
> > On 2022-1-26, at 11:12, internet-drafts@ietf.org =
<mailto:internet-drafts@ietf.org> wrote:
> >=20
> >=20
> > A New Internet-Draft is available from the on-line Internet-Drafts =
directories.
> > This draft is a work item of the TCP Maintenance and Minor =
Extensions WG of the IETF.
> >=20
> >        Title           : CUBIC for Fast and Long-Distance Networks
> >        Authors         : Lisong Xu
> >                          Sangtae Ha
> >                          Injong Rhee
> >                          Vidhi Goel
> >                          Lars Eggert
> >       Filename        : draft-ietf-tcpm-rfc8312bis-06.txt
> >       Pages           : 35
> >       Date            : 2022-01-26
> >=20
> > Abstract:
> >   CUBIC is a standard TCP congestion control algorithm that uses a
> >   cubic function instead of a linear congestion window increase
> >   function to improve scalability and stability over fast and long-
> >   distance networks.  CUBIC has been adopted as the default TCP
> >   congestion control algorithm by the Linux, Windows, and Apple =
stacks.
> >=20
> >   This document updates the specification of CUBIC to include
> >   algorithmic improvements based on these implementations and recent
> >   academic work.  Based on the extensive deployment experience with
> >   CUBIC, it also moves the specification to the Standards Track,
> >   obsoleting RFC 8312.  This also requires updating RFC 5681, to =
allow
> >   for CUBIC's occasionally more aggressive sending behavior.
> >=20
> >=20
> > The IETF datatracker status page for this draft is:
> > https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/ =
<https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/>
> >=20
> > There is also an HTML version available at:
> > https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html =
<https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html>
> >=20
> > A diff from the previous version is available at:
> > https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06 =
<https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06>
> >=20
> >=20
> > Internet-Drafts are also available by rsync at =
rsync.ietf.org::internet-drafts
> >=20
> >=20
> > _______________________________________________
> > tcpm mailing list
> > tcpm@ietf.org <mailto:tcpm@ietf.org>
> > https://www.ietf.org/mailman/listinfo/tcpm =
<https://www.ietf.org/mailman/listinfo/tcpm>
>=20
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org <mailto:tcpm@ietf.org>
> https://www.ietf.org/mailman/listinfo/tcpm =
<https://www.ietf.org/mailman/listinfo/tcpm>
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm


--Apple-Mail=_C451B4F0-F4F0-4693-BF86-9465A6AD561C
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=utf-8

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3Dutf-8"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: after-white-space;" =
class=3D"">Thanks Yoshi for reviewing the draft.<div class=3D""><br =
class=3D""></div><div class=3D""><blockquote type=3D"cite" class=3D""><div=
 dir=3D"ltr" class=3D""><div class=3D"">So, it seems one PS doc mentions =
not to use cwnd, while another PS doc mentions using cwnd is fine (with =
RFC7661).</div></div></blockquote>RFC5681 is conservative in its =
approach to use FlightSize and will result in very low throughput in =
scenarios when the congestion event occurs at the tail of send window =
(when flight size could be very small)</div><div class=3D""><br =
class=3D""></div><div class=3D""><blockquote type=3D"cite" class=3D""><div=
 dir=3D"ltr" class=3D""><div class=3D""><div class=3D"">I am not sure =
which can cause this situation yet, but in any case, does this mean =
there is no concern on&nbsp;</div><div class=3D"">"some implementations =
may incidentally increase well beyond rwnd" in cubic?</div></div><div =
class=3D"">Is this because this is an old implementation bug or are =
there other reasons?<br class=3D""></div><div class=3D"">I am wondering =
if we might want some clarification on this =
point.</div></div></blockquote><div class=3D"">If an implementation uses =
RFC7661, there is no concern about increasing cwnd beyond rwnd as =
RFC7661 tracks bytes acknowledged in a RTT which means that `rwnd` check =
has already been performed.</div><div class=3D"">Usually cwnd doesn=E2=80=99=
t grow beyond rwnd &nbsp;(ACK-clocking increases cwnd which means the =
sender already checked <span class=3D"Apple-tab-span" =
style=3D"white-space:pre">	</span>`rwnd` before sending out the =
data).</div><div class=3D""><br class=3D""></div><div class=3D"">But =
there could be implementations that may want to keep `rwnd` even smaller =
than initial `cwnd`. In those cases, RFC7661 would make the cwnd =
invalidated when the number of acknowledged bytes is much lower than the =
current cwnd.</div><div class=3D""><br class=3D""></div><div class=3D"">Do=
 you have any suggestions on how to make it more clear?</div><div =
class=3D""><br class=3D""></div><blockquote type=3D"cite" class=3D""><div =
dir=3D"ltr" class=3D""><div class=3D"">Also, I am wondering if we might =
want to use "MAY be used" here instead of "can be =
used"&nbsp;&nbsp;</div></div></blockquote><div class=3D""><div dir=3D"ltr"=
 class=3D""><div class=3D"">Yeah, we can make this change.</div><div =
class=3D""><br class=3D""></div><div class=3D"">I have opened an issue =
in GitHub&nbsp;<a href=3D"https://github.com/NTAP/rfc8312bis/issues/143" =
class=3D"">https://github.com/NTAP/rfc8312bis/issues/143</a>. We can =
continue the discussion there.</div><div class=3D""><br =
class=3D""></div><div class=3D""><br class=3D""></div><div =
class=3D"">Thanks,</div><div class=3D"">Vidhi</div></div></div><div><br =
class=3D""><blockquote type=3D"cite" class=3D""><div class=3D"">On Feb =
6, 2022, at 3:01 AM, Yoshifumi Nishida &lt;<a =
href=3D"mailto:nsd.ietf@gmail.com" class=3D"">nsd.ietf@gmail.com</a>&gt; =
wrote:</div><br class=3D"Apple-interchange-newline"><div class=3D""><div =
dir=3D"ltr" class=3D""><div class=3D"">Hi,</div><div class=3D"">I just =
have one more comment on the draft.</div><div class=3D"">In RFC5681, it =
mentions in Section 3:</div><div class=3D"">''</div><div class=3D""><pre =
style=3D"font-size: 13.3333px; margin-top: 0px; margin-bottom: 0px; =
break-before: page;" class=3D"">   Implementation Note: An easy mistake =
to make is to simply use cwnd,
   rather than FlightSize, which in some implementations may
   incidentally increase well beyond rwnd.</pre></div><div =
class=3D"">"</div><div class=3D""><br class=3D""></div><div =
class=3D"">while this draft mentions in Section 4.6:</div><div =
class=3D"">"</div><div class=3D""><pre style=3D"font-size: 13.3333px; =
margin-top: 0px; margin-bottom: 0px; break-before: page;" class=3D"">   =
To avoid suboptimal performance with such applications,</pre><pre =
style=3D"font-size: 13.3333px; margin-top: 0px; margin-bottom: 0px; =
break-before: page;" class=3D"">   the mechanisms described in [<a =
href=3D"https://datatracker.ietf.org/doc/html/rfc7661" =
title=3D"&quot;Updating TCP to Support Rate-Limited Traffic&quot;" =
target=3D"_blank" class=3D"">RFC7661</a>] can be used to mitigate this
   issue as it would allow using a value between _cwnd_ and
   _flight_size_ to calculate the new _ssthresh_ in Figure 5. </pre><pre =
style=3D"font-size: 13.3333px; margin-top: 0px; margin-bottom: 0px; =
break-before: page;" class=3D"">"</pre></div><div class=3D"">So, it =
seems one PS doc mentions not to use cwnd, while another PS doc mentions =
using cwnd is fine (with RFC7661).</div><div class=3D""><br =
class=3D""></div><div class=3D""><div class=3D"">I am not sure which can =
cause this situation yet, but in any case, does this mean there is no =
concern on&nbsp;</div><div class=3D"">"some implementations may =
incidentally increase well beyond rwnd" in cubic?</div></div><div =
class=3D"">Is this because this is an old implementation bug or are =
there other reasons?<br class=3D""></div><div class=3D"">I am wondering =
if we might want some clarification on this point.<br =
class=3D""></div><div class=3D"">&nbsp;</div><div class=3D"">Also, I am =
wondering if we might want to use "MAY be used" here instead of "can be =
used"&nbsp;&nbsp;</div><div class=3D""><br class=3D""></div><div =
class=3D"">Thanks,</div><div class=3D"">--</div><div =
class=3D"">Yoshi</div><div class=3D""><br class=3D""></div><div =
class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Mon, Jan =
31, 2022 at 11:43 PM Yoshifumi Nishida &lt;<a =
href=3D"mailto:nsd.ietf@gmail.com" target=3D"_blank" =
class=3D"">nsd.ietf@gmail.com</a>&gt; wrote:<br =
class=3D""></div><blockquote class=3D"gmail_quote" style=3D"margin:0px =
0px 0px 0.8ex;border-left:1px solid =
rgb(204,204,204);padding-left:1ex"><div dir=3D"ltr" class=3D""><div =
class=3D"">Hello,</div><div class=3D""><br class=3D""></div><div =
class=3D"">After some discussions among chairs, we decided to run the =
2nd WGLC on draft-ietf-tcpm-rfc8312bis in consideration of the =
importance of the draft.&nbsp;</div><div class=3D"">We'll be grateful if =
you could send your feedback to the ML. The WGLC runs until *Feb =
11*.</div><div class=3D""><br class=3D""></div><div class=3D"">If =
interested, you can check in-depth past discussions in the following =
URL.</div><div class=3D""><a href=3D"https://github.com/NTAP/rfc8312bis/" =
target=3D"_blank" class=3D"">https://github.com/NTAP/rfc8312bis/</a><br =
class=3D""></div><div class=3D""><br class=3D""></div><div =
class=3D"">Thank you so much!</div><div class=3D"">--</div><div =
class=3D"">tcpm co-chairs</div><div class=3D""><br class=3D""></div><br =
class=3D""><div class=3D"gmail_quote"><div dir=3D"ltr" =
class=3D"gmail_attr">On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert &lt;<a =
href=3D"mailto:lars@eggert.org" target=3D"_blank" =
class=3D"">lars@eggert.org</a>&gt; wrote:<br class=3D""></div><blockquote =
class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px =
solid rgb(204,204,204);padding-left:1ex">Hi,<br class=3D"">
<br class=3D"">
this -06 version rolls in all the changes requested during (and after) =
WGLC ended.<br class=3D"">
<br class=3D"">
I'll leave it up to the chairs to decide if another WGLC is warranted or =
the document can progress as-is.<br class=3D"">
<br class=3D"">
Thanks,<br class=3D"">
Lars<br class=3D"">
<br class=3D"">
<br class=3D"">
&gt; On 2022-1-26, at 11:12, <a href=3D"mailto:internet-drafts@ietf.org" =
target=3D"_blank" class=3D"">internet-drafts@ietf.org</a> wrote:<br =
class=3D"">
&gt; <br class=3D"">
&gt; <br class=3D"">
&gt; A New Internet-Draft is available from the on-line Internet-Drafts =
directories.<br class=3D"">
&gt; This draft is a work item of the TCP Maintenance and Minor =
Extensions WG of the IETF.<br class=3D"">
&gt; <br class=3D"">
&gt;&nbsp; &nbsp; &nbsp; &nbsp; Title&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp;: CUBIC for Fast and Long-Distance Networks<br class=3D"">
&gt;&nbsp; &nbsp; &nbsp; &nbsp; Authors&nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp;: Lisong Xu<br class=3D"">
&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; Sangtae Ha<br class=3D"">
&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; Injong Rhee<br class=3D"">
&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; Vidhi Goel<br class=3D"">
&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; Lars Eggert<br class=3D"">
&gt;&nbsp; &nbsp; &nbsp; &nbsp;Filename&nbsp; &nbsp; &nbsp; &nbsp; : =
draft-ietf-tcpm-rfc8312bis-06.txt<br class=3D"">
&gt;&nbsp; &nbsp; &nbsp; &nbsp;Pages&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp;: 35<br class=3D"">
&gt;&nbsp; &nbsp; &nbsp; &nbsp;Date&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; : 2022-01-26<br class=3D"">
&gt; <br class=3D"">
&gt; Abstract:<br class=3D"">
&gt;&nbsp; &nbsp;CUBIC is a standard TCP congestion control algorithm =
that uses a<br class=3D"">
&gt;&nbsp; &nbsp;cubic function instead of a linear congestion window =
increase<br class=3D"">
&gt;&nbsp; &nbsp;function to improve scalability and stability over fast =
and long-<br class=3D"">
&gt;&nbsp; &nbsp;distance networks.&nbsp; CUBIC has been adopted as the =
default TCP<br class=3D"">
&gt;&nbsp; &nbsp;congestion control algorithm by the Linux, Windows, and =
Apple stacks.<br class=3D"">
&gt; <br class=3D"">
&gt;&nbsp; &nbsp;This document updates the specification of CUBIC to =
include<br class=3D"">
&gt;&nbsp; &nbsp;algorithmic improvements based on these implementations =
and recent<br class=3D"">
&gt;&nbsp; &nbsp;academic work.&nbsp; Based on the extensive deployment =
experience with<br class=3D"">
&gt;&nbsp; &nbsp;CUBIC, it also moves the specification to the Standards =
Track,<br class=3D"">
&gt;&nbsp; &nbsp;obsoleting RFC 8312.&nbsp; This also requires updating =
RFC 5681, to allow<br class=3D"">
&gt;&nbsp; &nbsp;for CUBIC's occasionally more aggressive sending =
behavior.<br class=3D"">
&gt; <br class=3D"">
&gt; <br class=3D"">
&gt; The IETF datatracker status page for this draft is:<br class=3D"">
&gt; <a =
href=3D"https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/" =
rel=3D"noreferrer" target=3D"_blank" =
class=3D"">https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/</a=
><br class=3D"">
&gt; <br class=3D"">
&gt; There is also an HTML version available at:<br class=3D"">
&gt; <a =
href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html=
" rel=3D"noreferrer" target=3D"_blank" =
class=3D"">https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.h=
tml</a><br class=3D"">
&gt; <br class=3D"">
&gt; A diff from the previous version is available at:<br class=3D"">
&gt; <a =
href=3D"https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06"=
 rel=3D"noreferrer" target=3D"_blank" =
class=3D"">https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-=
06</a><br class=3D"">
&gt; <br class=3D"">
&gt; <br class=3D"">
&gt; Internet-Drafts are also available by rsync at <a =
href=3D"http://rsync.ietf.org" =
class=3D"">rsync.ietf.org</a>::internet-drafts<br class=3D"">
&gt; <br class=3D"">
&gt; <br class=3D"">
&gt; _______________________________________________<br class=3D"">
&gt; tcpm mailing list<br class=3D"">
&gt; <a href=3D"mailto:tcpm@ietf.org" target=3D"_blank" =
class=3D"">tcpm@ietf.org</a><br class=3D"">
&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/tcpm" =
rel=3D"noreferrer" target=3D"_blank" =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm</a><br class=3D"">
<br class=3D"">
_______________________________________________<br class=3D"">
tcpm mailing list<br class=3D"">
<a href=3D"mailto:tcpm@ietf.org" target=3D"_blank" =
class=3D"">tcpm@ietf.org</a><br class=3D"">
<a href=3D"https://www.ietf.org/mailman/listinfo/tcpm" rel=3D"noreferrer" =
target=3D"_blank" =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm</a><br class=3D"">
</blockquote></div></div>
</blockquote></div></div>
_______________________________________________<br class=3D"">tcpm =
mailing list<br class=3D""><a href=3D"mailto:tcpm@ietf.org" =
class=3D"">tcpm@ietf.org</a><br =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm<br =
class=3D""></div></blockquote></div><br class=3D""></div></body></html>=

--Apple-Mail=_C451B4F0-F4F0-4693-BF86-9465A6AD561C--

--Apple-Mail=_45C2D04E-78BF-4827-A84D-52F7F454FFD1
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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=
--Apple-Mail=_45C2D04E-78BF-4827-A84D-52F7F454FFD1--


From nobody Tue Feb  8 00:39:33 2022
Return-Path: <mohamed.boucadair@orange.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A0F7C3A0CAC for <tcpm@ietfa.amsl.com>; Tue,  8 Feb 2022 00:39:31 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.095
X-Spam-Level: 
X-Spam-Status: No, score=-2.095 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_BLOCKED=0.001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, UNPARSEABLE_RELAY=0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=orange.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Yd7ou0adOMnI for <tcpm@ietfa.amsl.com>; Tue,  8 Feb 2022 00:39:26 -0800 (PST)
Received: from relais-inet.orange.com (relais-inet.orange.com [80.12.66.41]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 92EFC3A0CB0 for <tcpm@ietf.org>; Tue,  8 Feb 2022 00:39:26 -0800 (PST)
Received: from opfedar02.francetelecom.fr (unknown [xx.xx.xx.4]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits)) (No client certificate requested) by opfedar20.francetelecom.fr (ESMTP service) with ESMTPS id 4JtGdn0Jpvz8svW;  Tue,  8 Feb 2022 09:39:25 +0100 (CET)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=orange.com; s=ORANGE001; t=1644309565; bh=HWW0WSL1Dua/F5rORfRblPFvqoBkPpUF2FH3AhnLFoU=; h=From:To:Subject:Date:Message-ID:Content-Type: Content-Transfer-Encoding:MIME-Version; b=OwbLb3OjVfMp9lDGc+ebCUgE1sddZBUHMpX15JbAxIuuR0U/+RSuOYgdNAqecZOOj XJjHlS6nECUboieklLPDt+IlVZ4v1S/IcRvvyFaXg7gZlaf1CSn44BGcCZKHdGTTsd wWW3Pq45APj2DswCQ8x+e/QoCH7EQ19jjbeoogeS04qMCbuSylfvbfxPVyuzQkUZEa 0NpeUBCO7HXEyZxqEs9Mv3x6QI3m2xpxmwHPnPx8UHgFh8KGiukWmHVsPz4Yr+JugE K4+0A9awffVv9+h8Znyc8lMT92rr0rGymG8inKo11QoPa0MIllWag+nCptrg/6snaQ RqL60X06SBYrQ==
Received: from Exchangemail-eme6.itn.ftgroup (unknown [xx.xx.13.48]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by opfedar02.francetelecom.fr (ESMTP service) with ESMTPS id 4JtGdm6bsHzCqkP;  Tue,  8 Feb 2022 09:39:24 +0100 (CET)
From: <mohamed.boucadair@orange.com>
To: "Scharf, Michael" <Michael.Scharf@hs-esslingen.de>
CC: "tcpm@ietf.org" <tcpm@ietf.org>
Thread-Topic: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
Thread-Index: AQHXyc25qfTSowdLSEa6dKhesfqRYKv/uF0AgEvBUuCABenxUIAYa72QgBqz19CABXaAYA==
Content-Class: 
Date: Tue, 8 Feb 2022 08:39:23 +0000
Message-ID: <18668_1644309564_62022C3C_18668_151_1_787AE7BB302AE849A7480A190F8B93303548DD57@OPEXCAUBMA2.corporate.adroot.infra.ftgroup>
References: <163518613944.7645.2494692847367626249@ietfa.amsl.com> <18622_1636712195_618E3F03_18622_35_1_787AE7BB302AE849A7480A190F8B933035450859@OPEXCAUBMA2.corporate.adroot.infra.ftgroup> <bde902389b8a4c76aae776ea2d68d9ce@hs-esslingen.de> <3337_1641201204_61D2BE34_3337_132_1_787AE7BB302AE849A7480A190F8B93303546BDC5@OPEXCAUBMA2.corporate.adroot.infra.ftgroup> <706848a62fd74d18a34566046369d049@hs-esslingen.de> <bdc3831619dd442dbc5e2914fff7a2b5@hs-esslingen.de>
In-Reply-To: <bdc3831619dd442dbc5e2914fff7a2b5@hs-esslingen.de>
Accept-Language: fr-FR, en-US
Content-Language: fr-FR
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
msip_labels: MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_Enabled=true; MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_SetDate=2022-02-08T08:34:21Z;  MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_Method=Privileged; MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_Name=unrestricted_parent.2; MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_SiteId=90c7a20a-f34b-40bf-bc48-b9253b6f5d20; MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_ActionId=918be82b-2241-4d63-9cc9-9ad158f11070; MSIP_Label_07222825-62ea-40f3-96b5-5375c07996e2_ContentBits=0
x-originating-ip: [10.114.13.245]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/Uk70daQeFiKWGKtxG4c2SxeaeeY>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Feb 2022 08:39:32 -0000

Hi Michael,=20

Thank you.=20

There is one minor comment about the ipv6 addresses in the examples: they s=
hould be derived from the prefix in RFC 3849. This should be fixed in draft=
-ietf-tcpm-ao-test-vectors (the same applies for 10.11.12.13, etcs that are=
 not from the blocks defined in 5737).=20

Other than that, this looks good to me. Thanks.

Cheers,
Med

> -----Message d'origine-----
> De=A0: Scharf, Michael <Michael.Scharf@hs-esslingen.de>
> Envoy=E9=A0: vendredi 4 f=E9vrier 2022 22:12
> =C0=A0: BOUCADAIR Mohamed INNOV/NET <mohamed.boucadair@orange.com>
> Cc=A0: tcpm@ietf.org
> Objet=A0: RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
>=20
> For what it is worth, -06 with the mentioned changes has been published.
>=20
> The diff can be found at: https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-
> tcpm-yang-tcp-06.txt
>=20
> I hope that this update fixes all issues.
>=20
> Michael
>=20
> > -----Original Message-----
> > From: tcpm <tcpm-bounces@ietf.org> On Behalf Of Scharf, Michael
> > Sent: Tuesday, January 18, 2022 10:32 PM
> > To: mohamed.boucadair@orange.com
> > Cc: tcpm@ietf.org
> > Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
> >
> > Hi Med,
> >
> > Thanks a lot for this further review with good catches.
> >
> > We have addressed all comments in the current working copy for -06,
> > except for the disclaimer text regarding line wrap in B.1. This
> > disclaimer is currently automatically introduced by the build tools no
> matter whether it is needed.
> > That detail can be sorted out later by the RFC editor.
> >
> > Details can be found at the repository with the working copy:
> > https://github.com/mjethanandani/ietf-tcp
> >
> > Our suggestion is to change the TCP-AO example so that it illustrates
> > the use of send-id and recv-id with parameters documented in
> > draft-ietf-tcpm-ao- test-vectors.
> >
> > As the WGLC is ongoing, we would wait for further comments before
> > finalizing -06.
> >
> > Thanks
> >
> > Michael (as author)
> >
> >
> > > -----Original Message-----
> > > From: mohamed.boucadair@orange.com
> > > <mohamed.boucadair@orange.com>
> > > Sent: Monday, January 3, 2022 10:13 AM
> > > To: Scharf, Michael <Michael.Scharf@hs-esslingen.de>
> > > Cc: tcpm@ietf.org
> > > Subject: RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
> > >
> > > Hi Michael,
> > >
> > > The changes look good. Thank you.
> > >
> > > Please find below some comments on the new version:
> > > - Section 4
> > >
> > > (1)
> > > OLD:
> > >        reference
> > >          "RFC XXXX, YANG Model for Transmission Control Protocol
> (TCP)
> > >                     Configuration.";
> > >
> > > NEW:
> > >        reference
> > >          "RFC XXXX: A YANG Model for Transmission Control Protocol
> (TCP)
> > >                     Configuration";
> > >
> > > (2)
> > >
> > > Add "when set to true" in the description of almost all the Boolean
> nodes.
> > >
> > > (3)
> > >
> > > I still don't think "forms the connection identifier" is accurate.
> > > The
> > connection
> > > is identified by the 4-uple (which is captured in the key
> > > statement). You
> > may
> > > consider simply: s/forms the connection identifier/of the
> connection.
> > > Another better approach would be, e.g.:
> > >
> > > OLD:
> > >  "Local address that forms the connection identifier.";
> > > NEW:
> > >  "Identifies the address that is used locally by an endpoint to bind
> > > the connection.";
> > >
> > > OLD:
> > >  "Remote address that forms the connection identifier.";
> > > NEW:
> > >  "Identifies the address that is used by the peer endpoint to bind
> > > the connection.";
> > >
> > > etc.
> > >
> > > - Make this change in 5.2: s/registrations are
> > > requested/registration is requested
> > > - Update 5.2 to make it explicit that the module is not maintained
> by IANA:
> > >
> > > OLD:
> > >       name:         ietf-tcp
> > >       namespace:    urn:ietf:params:xml:ns:yang:ietf-tcp
> > >       prefix:       tcp
> > >       reference:    RFC XXXX
> > >
> > > NEW:
> > >       name:         ietf-tcp
> > >       namespace:    urn:ietf:params:xml:ns:yang:ietf-tcp
> > >       prefix:       tcp
> > >       maintained by IANA: N
> > >       reference:    RFC XXXX
> > >
> > > - I think you can remove the note in B.1
> > > - remove "\" right after "</description>" in B.2
> > > - As you are referring to RFC 8792, you need to add the 8792 header
> > > to adhere with the following (B.2):
> > >
> > > =3D=3D=3D=3D=3D=3D
> > > ...
> > >    Text content that has been folded as specified by this strategy
> MUST
> > >    adhere to the following structure.
> > >
> > > 7.1.1.  Header
> > >
> > >    The header is two lines long.
> > >
> > >    The first line is the following 36-character string; this string
> MAY
> > >    be surrounded by any number of printable characters.  This first
> line
> > >    cannot itself be folded.
> > >
> > >    NOTE: '\' line wrapping per RFC 8792
> > >
> > >    The second line is an empty line, containing only the end-of-line
> > >    character sequence.  This line provides visual separation for
> > >    readability.
> > > =3D=3D=3D=3D=3D
> > >
> > > - B.2:
> > >
> > > <!--
> > > This example sets TCP-AO configuration parameters as demonstrated by
> > > examples in draft-touch-tcpm-ao-test-vectors.
> > > -->
> > >
> > > Not sure why this I-D is cited here given the example does not echo
> > > an example from that I-D (e.g., same local/remote port, local
> address, etc.).
> > > Also, I was expecting to see an example that illustrates the use of
> > > send-id and recv-id.
> > >
> > > If the citation is maintained, please make this change:
> > >
> > > s/demonstrated by examples in draft-touch-tcpm-ao-test-
> > > vectors/demonstrated by examples in I-D.ietf-tcpm-ao-test-vectors
> > >
> > > Cheers,
> > > Med
> > >
> > > > -----Message d'origine-----
> > > > De=A0: Scharf, Michael <Michael.Scharf@hs-esslingen.de> Envoy=E9=A0:
> > > > jeudi 30 d=E9cembre 2021 15:19 =C0=A0: BOUCADAIR Mohamed INNOV/NET
> > > <mohamed.boucadair@orange.com>
> > > > Cc=A0: tcpm@ietf.org
> > > > Objet=A0: RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
> > > >
> > > > Hi Med,
> > > >
> > > > Thanks a lot for the detailed review. I went through all your
> comments.
> > > > Most were straightforward to address.
> > > >
> > > > Our suggestion is to clearly state that MPTCP is outside the scope
> > > > of this model. MPTCP may need a different YANG model. Yet, the
> > > > most obvious
> > > use
> > > > case of the YANG model are TCP-based control plane protocols on
> > routers
> > > > (BGP, LDP, ...). At least I am not aware of widespread use of BGP
> > > > over MPTCP.
> > > >
> > > > Can you please have a look at -05 and let me know if -05 works for
> you?
> > > >
> > > > The full diff is available at:
> > > > https://www.ietf.org/rfcdiff?url2=3Ddraft-
> > > > ietf-tcpm-yang-tcp-05
> > > >
> > > > Thanks
> > > >
> > > > Michael
> > > >
> > > > > -----Original Message-----
> > > > > From: mohamed.boucadair@orange.com
> > > > > <mohamed.boucadair@orange.com>
> > > > > Sent: Friday, November 12, 2021 11:17 AM
> > > > > To: Scharf, Michael <Michael.Scharf@hs-esslingen.de>
> > > > > Cc: tcpm@ietf.org
> > > > > Subject: RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
> > > > >
> > > > > Hi Michael,
> > > > >
> > > > > FWIW, please find below some comments to the latest version of
> > > > > the
> > > > draft:
> > > > >
> > > > > * pdf: https://raw.githubusercontent.com/boucadair/IETF-Drafts-
> > > > > Reviews/master/draft-ietf-tcpm-yang-tcp-04-rev%20Med.pdf
> > > > > * doc: https://github.com/boucadair/IETF-Drafts-
> > > > > Reviews/raw/master/draft-ietf-tcpm-yang-tcp-04-rev%20Med.doc
> > > > >
> > > > > I think that the document is almost stable. Thank you.
> > > > >
> > > > > Cheers,
> > > > > Med
> > > > >
> > > > > > -----Message d'origine-----
> > > > > > De=A0: tcpm <tcpm-bounces@ietf.org> De la part de
> > > > > > internet-drafts@ietf.org Envoy=E9=A0: lundi 25 octobre 2021 20:=
22
> =C0=A0:
> > > > > > i-d-announce@ietf.org Cc=A0: tcpm@ietf.org Objet=A0: [tcpm] I-D
> Action:
> > > > > > draft-ietf-tcpm-yang-tcp-04.txt
> > > > > >
> > > > > >
> > > > > > A New Internet-Draft is available from the on-line
> > > > > > Internet-Drafts directories.
> > > > > > This draft is a work item of the TCP Maintenance and Minor
> > > > > > Extensions WG of the IETF.
> > > > > >
> > > > > >         Title           : YANG Model for Transmission Control
> Protocol
> > > > > > (TCP) Configuration
> > > > > >         Authors         : Michael Scharf
> > > > > >                           Mahesh Jethanandani
> > > > > >                           Vishal Murgai
> > > > > > 	Filename        : draft-ietf-tcpm-yang-tcp-04.txt
> > > > > > 	Pages           : 23
> > > > > > 	Date            : 2021-10-25
> > > > > >
> > > > > > Abstract:
> > > > > >    This document specifies a minimal YANG model for TCP on
> > > > > > devices
> > > > that
> > > > > >    are configured by network management protocols.  The YANG
> > model
> > > > > >    defines a container for all TCP connections and groupings
> of
> > > > > >    authentication parameters that can be imported and used in
> TCP
> > > > > >    implementations or by other models that need to configure
> TCP
> > > > > >    parameters.  The model also includes basic TCP statistics.
> > > > > > The
> > > > model
> > > > > >    is NMDA (RFC 8342) compliant.
> > > > > >
> > > > > >
> > > > > > The IETF datatracker status page for this draft is:
> > > > > > https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/
> > > > > >
> > > > > > There is also an htmlized version available at:
> > > > > > https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-yang-tcp
> > > > > > -04
> > > > > >
> > > > > > A diff from the previous version is available at:
> > > > > > https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-yang-tcp-04
> > > > > >
> > > > > >
> > > > > > Internet-Drafts are also available by anonymous FTP at:
> > > > > > ftp://ftp.ietf.org/internet-drafts/
> > > > > >
> > > > > >
> > > > > > _______________________________________________
> > > > > > tcpm mailing list
> > > > > > tcpm@ietf.org
> > > > > > https://www.ietf.org/mailman/listinfo/tcpm
> > > > >
> > > > >
> > >
> > __________________________________________________________
> > > > >
> > >
> > __________________________________________________________
> > > > > _____
> > > > >
> > > > > Ce message et ses pieces jointes peuvent contenir des
> > > > > informations confidentielles ou privilegiees et ne doivent donc
> > > > > pas etre diffuses, exploites ou copies sans autorisation. Si
> > > > > vous avez recu ce message par erreur, veuillez le signaler a
> > > > > l'expediteur et le detruire ainsi que les pieces jointes. Les
> > > > > messages electroniques etant susceptibles d'alteration, Orange
> > > > > decline toute responsabilite si ce message a ete altere, deforme
> ou falsifie. Merci.
> > > > >
> > > > > This message and its attachments may contain confidential or
> > > > > privileged information that may be protected by law; they should
> > > > > not be distributed, used or copied without authorisation.
> > > > > If you have received this email in error, please notify the
> > > > > sender and delete this message and its attachments.
> > > > > As emails may be altered, Orange is not liable for messages that
> > > > > have been modified, changed or falsified.
> > > > > Thank you.
> > >
> > >
> > >
> > __________________________________________________________
> > >
> > __________________________________________________________
> > > _____
> > >
> > > Ce message et ses pieces jointes peuvent contenir des informations
> > > confidentielles ou privilegiees et ne doivent donc pas etre
> > > diffuses, exploites ou copies sans autorisation. Si vous avez recu
> > > ce message par erreur, veuillez le signaler a l'expediteur et le
> > > detruire ainsi que les pieces jointes. Les messages electroniques
> > > etant susceptibles d'alteration, Orange decline toute responsabilite
> > > si ce message a ete altere, deforme ou falsifie. Merci.
> > >
> > > This message and its attachments may contain confidential or
> > > privileged information that may be protected by law; they should not
> > > be distributed, used or copied without authorisation.
> > > If you have received this email in error, please notify the sender
> > > and delete this message and its attachments.
> > > As emails may be altered, Orange is not liable for messages that
> > > have been modified, changed or falsified.
> > > Thank you.
> >
> > _______________________________________________
> > tcpm mailing list
> > tcpm@ietf.org
> > https://www.ietf.org/mailman/listinfo/tcpm

___________________________________________________________________________=
______________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confiden=
tielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu=
 ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages el=
ectroniques etant susceptibles d'alteration,
Orange decline toute responsabilite si ce message a ete altere, deforme ou =
falsifie. Merci.

This message and its attachments may contain confidential or privileged inf=
ormation that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and dele=
te this message and its attachments.
As emails may be altered, Orange is not liable for messages that have been =
modified, changed or falsified.
Thank you.


From nobody Tue Feb  8 07:16:49 2022
Return-Path: <touch@strayalpha.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 938223A0E66 for <tcpm@ietfa.amsl.com>; Tue,  8 Feb 2022 07:16:46 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.317
X-Spam-Level: 
X-Spam-Status: No, score=-1.317 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_NEUTRAL=0.779, URIBL_BLOCKED=0.001] autolearn=no autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=strayalpha.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7JdJ2goi2Uja for <tcpm@ietfa.amsl.com>; Tue,  8 Feb 2022 07:16:42 -0800 (PST)
Received: from server217-1.web-hosting.com (server217-1.web-hosting.com [198.54.114.226]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E3BC83A0891 for <tcpm@ietf.org>; Tue,  8 Feb 2022 07:16:41 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=strayalpha.com; s=default; h=To:References:Message-Id:Cc:Date:In-Reply-To: From:Subject:Mime-Version:Content-Type:Sender:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id: List-Help:List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=4nwlpGWpfgRo79XAPnzd+WdvsyDpM8hB1VB4NRjX2tU=; b=dG3fECLMOjpvQS3jnb/Yzp20O8 qQo0LFIqjMF1Ow2FvBoKp4svWTmmp6FvpCJDES8OHvRSB5upXaypweDEG0dEIz2Qrap+gbRdJrx7X J8E31t4wuvJwnxFzaq6FXajz9CFC2F8lOTwB4MaMP+2mft0OhrHOzL4IPfLo+zQ1GNyRCth+yakWB m2wnlWyYM1mc9Pbkvsbmtv1HdXCtTSI+YTIfyX+Wm2Bi1hEoj8w1t+GV5Lm/ouG+MsfTh7sbZYg+l GAINg7jyYHrceVJ3A78yZspnbFgwBBiZkk0mi1Scnyh7z8ecKmwFfnQyP0Zk3+XkIjU29G5jiw/IN +fskbYhg==;
Received: from cpe-172-114-237-88.socal.res.rr.com ([172.114.237.88]:58966 helo=smtpclient.apple) by server217.web-hosting.com with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from <touch@strayalpha.com>) id 1nHSEN-003yPq-H6; Tue, 08 Feb 2022 10:16:41 -0500
Content-Type: multipart/alternative; boundary="Apple-Mail=_B8EEAA2F-6A8D-4449-9DC3-22E4D82DC951"
Mime-Version: 1.0 (Mac OS X Mail 15.0 \(3693.40.0.1.81\))
From: "touch@strayalpha.com" <touch@strayalpha.com>
In-Reply-To: <18668_1644309564_62022C3C_18668_151_1_787AE7BB302AE849A7480A190F8B93303548DD57@OPEXCAUBMA2.corporate.adroot.infra.ftgroup>
Date: Tue, 8 Feb 2022 07:16:31 -0800
Cc: "Scharf, Michael" <Michael.Scharf@hs-esslingen.de>, "tcpm@ietf.org" <tcpm@ietf.org>
Message-Id: <745AF8A3-5F02-4752-BA28-43C6F10FFCAB@strayalpha.com>
References: <163518613944.7645.2494692847367626249@ietfa.amsl.com> <18622_1636712195_618E3F03_18622_35_1_787AE7BB302AE849A7480A190F8B933035450859@OPEXCAUBMA2.corporate.adroot.infra.ftgroup> <bde902389b8a4c76aae776ea2d68d9ce@hs-esslingen.de> <3337_1641201204_61D2BE34_3337_132_1_787AE7BB302AE849A7480A190F8B93303546BDC5@OPEXCAUBMA2.corporate.adroot.infra.ftgroup> <706848a62fd74d18a34566046369d049@hs-esslingen.de> <bdc3831619dd442dbc5e2914fff7a2b5@hs-esslingen.de> <18668_1644309564_62022C3C_18668_151_1_787AE7BB302AE849A7480A190F8B93303548DD57@OPEXCAUBMA2.corporate.adroot.infra.ftgroup>
To: mohamed.boucadair@orange.com
X-Mailer: Apple Mail (2.3693.40.0.1.81)
X-OutGoing-Spam-Status: No, score=-1.0
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - server217.web-hosting.com
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - strayalpha.com
X-Get-Message-Sender-Via: server217.web-hosting.com: authenticated_id: touch@strayalpha.com
X-Authenticated-Sender: server217.web-hosting.com: touch@strayalpha.com
X-Source: 
X-Source-Args: 
X-Source-Dir: 
X-From-Rewrite: unmodified, already matched
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/2cm8973YRKhzqDvxyuOCc8MJrb8>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Feb 2022 15:16:47 -0000

--Apple-Mail=_B8EEAA2F-6A8D-4449-9DC3-22E4D82DC951
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

Hi, Med,

I disagree. The addresses in test vectors are not examples for =
documentation, but rather an actual configuration intended for testing =
purposes.

I found only one other RFC that provides test vectors based on packet =
fields that include IP addresses: RFC 5769, but it (IMO) incorrectly =
uses the documentation addresses you suggest (I expect to submit an =
errata on that shortly - let me know if you know of others).

Joe

=E2=80=94
Joe Touch, temporal epistemologist
www.strayalpha.com

> On Feb 8, 2022, at 12:39 AM, mohamed.boucadair@orange.com wrote:
>=20
> Hi Michael,=20
>=20
> Thank you.=20
>=20
> There is one minor comment about the ipv6 addresses in the examples: =
they should be derived from the prefix in RFC 3849. This should be fixed =
in draft-ietf-tcpm-ao-test-vectors (the same applies for 10.11.12.13, =
etcs that are not from the blocks defined in 5737).=20
>=20
> Other than that, this looks good to me. Thanks.
>=20
> Cheers,
> Med
>=20
>> -----Message d'origine-----
>> De : Scharf, Michael <Michael.Scharf@hs-esslingen.de>
>> Envoy=C3=A9 : vendredi 4 f=C3=A9vrier 2022 22:12
>> =C3=80 : BOUCADAIR Mohamed INNOV/NET <mohamed.boucadair@orange.com>
>> Cc : tcpm@ietf.org
>> Objet : RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
>>=20
>> For what it is worth, -06 with the mentioned changes has been =
published.
>>=20
>> The diff can be found at: =
https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-
>> tcpm-yang-tcp-06.txt
>>=20
>> I hope that this update fixes all issues.
>>=20
>> Michael
>>=20
>>> -----Original Message-----
>>> From: tcpm <tcpm-bounces@ietf.org> On Behalf Of Scharf, Michael
>>> Sent: Tuesday, January 18, 2022 10:32 PM
>>> To: mohamed.boucadair@orange.com
>>> Cc: tcpm@ietf.org
>>> Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
>>>=20
>>> Hi Med,
>>>=20
>>> Thanks a lot for this further review with good catches.
>>>=20
>>> We have addressed all comments in the current working copy for -06,
>>> except for the disclaimer text regarding line wrap in B.1. This
>>> disclaimer is currently automatically introduced by the build tools =
no
>> matter whether it is needed.
>>> That detail can be sorted out later by the RFC editor.
>>>=20
>>> Details can be found at the repository with the working copy:
>>> https://github.com/mjethanandani/ietf-tcp
>>>=20
>>> Our suggestion is to change the TCP-AO example so that it =
illustrates
>>> the use of send-id and recv-id with parameters documented in
>>> draft-ietf-tcpm-ao- test-vectors.
>>>=20
>>> As the WGLC is ongoing, we would wait for further comments before
>>> finalizing -06.
>>>=20
>>> Thanks
>>>=20
>>> Michael (as author)
>>>=20
>>>=20
>>>> -----Original Message-----
>>>> From: mohamed.boucadair@orange.com
>>>> <mohamed.boucadair@orange.com>
>>>> Sent: Monday, January 3, 2022 10:13 AM
>>>> To: Scharf, Michael <Michael.Scharf@hs-esslingen.de>
>>>> Cc: tcpm@ietf.org
>>>> Subject: RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
>>>>=20
>>>> Hi Michael,
>>>>=20
>>>> The changes look good. Thank you.
>>>>=20
>>>> Please find below some comments on the new version:
>>>> - Section 4
>>>>=20
>>>> (1)
>>>> OLD:
>>>>       reference
>>>>         "RFC XXXX, YANG Model for Transmission Control Protocol
>> (TCP)
>>>>                    Configuration.";
>>>>=20
>>>> NEW:
>>>>       reference
>>>>         "RFC XXXX: A YANG Model for Transmission Control Protocol
>> (TCP)
>>>>                    Configuration";
>>>>=20
>>>> (2)
>>>>=20
>>>> Add "when set to true" in the description of almost all the Boolean
>> nodes.
>>>>=20
>>>> (3)
>>>>=20
>>>> I still don't think "forms the connection identifier" is accurate.
>>>> The
>>> connection
>>>> is identified by the 4-uple (which is captured in the key
>>>> statement). You
>>> may
>>>> consider simply: s/forms the connection identifier/of the
>> connection.
>>>> Another better approach would be, e.g.:
>>>>=20
>>>> OLD:
>>>> "Local address that forms the connection identifier.";
>>>> NEW:
>>>> "Identifies the address that is used locally by an endpoint to bind
>>>> the connection.";
>>>>=20
>>>> OLD:
>>>> "Remote address that forms the connection identifier.";
>>>> NEW:
>>>> "Identifies the address that is used by the peer endpoint to bind
>>>> the connection.";
>>>>=20
>>>> etc.
>>>>=20
>>>> - Make this change in 5.2: s/registrations are
>>>> requested/registration is requested
>>>> - Update 5.2 to make it explicit that the module is not maintained
>> by IANA:
>>>>=20
>>>> OLD:
>>>>      name:         ietf-tcp
>>>>      namespace:    urn:ietf:params:xml:ns:yang:ietf-tcp
>>>>      prefix:       tcp
>>>>      reference:    RFC XXXX
>>>>=20
>>>> NEW:
>>>>      name:         ietf-tcp
>>>>      namespace:    urn:ietf:params:xml:ns:yang:ietf-tcp
>>>>      prefix:       tcp
>>>>      maintained by IANA: N
>>>>      reference:    RFC XXXX
>>>>=20
>>>> - I think you can remove the note in B.1
>>>> - remove "\" right after "</description>" in B.2
>>>> - As you are referring to RFC 8792, you need to add the 8792 header
>>>> to adhere with the following (B.2):
>>>>=20
>>>> =3D=3D=3D=3D=3D=3D
>>>> ...
>>>>   Text content that has been folded as specified by this strategy
>> MUST
>>>>   adhere to the following structure.
>>>>=20
>>>> 7.1.1.  Header
>>>>=20
>>>>   The header is two lines long.
>>>>=20
>>>>   The first line is the following 36-character string; this string
>> MAY
>>>>   be surrounded by any number of printable characters.  This first
>> line
>>>>   cannot itself be folded.
>>>>=20
>>>>   NOTE: '\' line wrapping per RFC 8792
>>>>=20
>>>>   The second line is an empty line, containing only the end-of-line
>>>>   character sequence.  This line provides visual separation for
>>>>   readability.
>>>> =3D=3D=3D=3D=3D
>>>>=20
>>>> - B.2:
>>>>=20
>>>> <!--
>>>> This example sets TCP-AO configuration parameters as demonstrated =
by
>>>> examples in draft-touch-tcpm-ao-test-vectors.
>>>> -->
>>>>=20
>>>> Not sure why this I-D is cited here given the example does not echo
>>>> an example from that I-D (e.g., same local/remote port, local
>> address, etc.).
>>>> Also, I was expecting to see an example that illustrates the use of
>>>> send-id and recv-id.
>>>>=20
>>>> If the citation is maintained, please make this change:
>>>>=20
>>>> s/demonstrated by examples in draft-touch-tcpm-ao-test-
>>>> vectors/demonstrated by examples in I-D.ietf-tcpm-ao-test-vectors
>>>>=20
>>>> Cheers,
>>>> Med
>>>>=20
>>>>> -----Message d'origine-----
>>>>> De : Scharf, Michael <Michael.Scharf@hs-esslingen.de> Envoy=C3=A9 =
:
>>>>> jeudi 30 d=C3=A9cembre 2021 15:19 =C3=80 : BOUCADAIR Mohamed =
INNOV/NET
>>>> <mohamed.boucadair@orange.com>
>>>>> Cc : tcpm@ietf.org
>>>>> Objet : RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
>>>>>=20
>>>>> Hi Med,
>>>>>=20
>>>>> Thanks a lot for the detailed review. I went through all your
>> comments.
>>>>> Most were straightforward to address.
>>>>>=20
>>>>> Our suggestion is to clearly state that MPTCP is outside the scope
>>>>> of this model. MPTCP may need a different YANG model. Yet, the
>>>>> most obvious
>>>> use
>>>>> case of the YANG model are TCP-based control plane protocols on
>>> routers
>>>>> (BGP, LDP, ...). At least I am not aware of widespread use of BGP
>>>>> over MPTCP.
>>>>>=20
>>>>> Can you please have a look at -05 and let me know if -05 works for
>> you?
>>>>>=20
>>>>> The full diff is available at:
>>>>> https://www.ietf.org/rfcdiff?url2=3Ddraft-
>>>>> ietf-tcpm-yang-tcp-05
>>>>>=20
>>>>> Thanks
>>>>>=20
>>>>> Michael
>>>>>=20
>>>>>> -----Original Message-----
>>>>>> From: mohamed.boucadair@orange.com
>>>>>> <mohamed.boucadair@orange.com>
>>>>>> Sent: Friday, November 12, 2021 11:17 AM
>>>>>> To: Scharf, Michael <Michael.Scharf@hs-esslingen.de>
>>>>>> Cc: tcpm@ietf.org
>>>>>> Subject: RE: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
>>>>>>=20
>>>>>> Hi Michael,
>>>>>>=20
>>>>>> FWIW, please find below some comments to the latest version of
>>>>>> the
>>>>> draft:
>>>>>>=20
>>>>>> * pdf: https://raw.githubusercontent.com/boucadair/IETF-Drafts-
>>>>>> Reviews/master/draft-ietf-tcpm-yang-tcp-04-rev%20Med.pdf
>>>>>> * doc: https://github.com/boucadair/IETF-Drafts-
>>>>>> Reviews/raw/master/draft-ietf-tcpm-yang-tcp-04-rev%20Med.doc
>>>>>>=20
>>>>>> I think that the document is almost stable. Thank you.
>>>>>>=20
>>>>>> Cheers,
>>>>>> Med
>>>>>>=20
>>>>>>> -----Message d'origine-----
>>>>>>> De : tcpm <tcpm-bounces@ietf.org> De la part de
>>>>>>> internet-drafts@ietf.org Envoy=C3=A9 : lundi 25 octobre 2021 =
20:22
>> =C3=80 :
>>>>>>> i-d-announce@ietf.org Cc : tcpm@ietf.org Objet : [tcpm] I-D
>> Action:
>>>>>>> draft-ietf-tcpm-yang-tcp-04.txt
>>>>>>>=20
>>>>>>>=20
>>>>>>> A New Internet-Draft is available from the on-line
>>>>>>> Internet-Drafts directories.
>>>>>>> This draft is a work item of the TCP Maintenance and Minor
>>>>>>> Extensions WG of the IETF.
>>>>>>>=20
>>>>>>>        Title           : YANG Model for Transmission Control
>> Protocol
>>>>>>> (TCP) Configuration
>>>>>>>        Authors         : Michael Scharf
>>>>>>>                          Mahesh Jethanandani
>>>>>>>                          Vishal Murgai
>>>>>>> 	Filename        : draft-ietf-tcpm-yang-tcp-04.txt
>>>>>>> 	Pages           : 23
>>>>>>> 	Date            : 2021-10-25
>>>>>>>=20
>>>>>>> Abstract:
>>>>>>>   This document specifies a minimal YANG model for TCP on
>>>>>>> devices
>>>>> that
>>>>>>>   are configured by network management protocols.  The YANG
>>> model
>>>>>>>   defines a container for all TCP connections and groupings
>> of
>>>>>>>   authentication parameters that can be imported and used in
>> TCP
>>>>>>>   implementations or by other models that need to configure
>> TCP
>>>>>>>   parameters.  The model also includes basic TCP statistics.
>>>>>>> The
>>>>> model
>>>>>>>   is NMDA (RFC 8342) compliant.
>>>>>>>=20
>>>>>>>=20
>>>>>>> The IETF datatracker status page for this draft is:
>>>>>>> https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/
>>>>>>>=20
>>>>>>> There is also an htmlized version available at:
>>>>>>> https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-yang-tcp
>>>>>>> -04
>>>>>>>=20
>>>>>>> A diff from the previous version is available at:
>>>>>>> https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-yang-tcp-04
>>>>>>>=20
>>>>>>>=20
>>>>>>> Internet-Drafts are also available by anonymous FTP at:
>>>>>>> ftp://ftp.ietf.org/internet-drafts/
>>>>>>>=20
>>>>>>>=20
>>>>>>> _______________________________________________
>>>>>>> tcpm mailing list
>>>>>>> tcpm@ietf.org
>>>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>>>=20
>>>>>>=20
>>>>=20
>>> __________________________________________________________
>>>>>>=20
>>>>=20
>>> __________________________________________________________
>>>>>> _____
>>>>>>=20
>>>>>> Ce message et ses pieces jointes peuvent contenir des
>>>>>> informations confidentielles ou privilegiees et ne doivent donc
>>>>>> pas etre diffuses, exploites ou copies sans autorisation. Si
>>>>>> vous avez recu ce message par erreur, veuillez le signaler a
>>>>>> l'expediteur et le detruire ainsi que les pieces jointes. Les
>>>>>> messages electroniques etant susceptibles d'alteration, Orange
>>>>>> decline toute responsabilite si ce message a ete altere, deforme
>> ou falsifie. Merci.
>>>>>>=20
>>>>>> This message and its attachments may contain confidential or
>>>>>> privileged information that may be protected by law; they should
>>>>>> not be distributed, used or copied without authorisation.
>>>>>> If you have received this email in error, please notify the
>>>>>> sender and delete this message and its attachments.
>>>>>> As emails may be altered, Orange is not liable for messages that
>>>>>> have been modified, changed or falsified.
>>>>>> Thank you.
>>>>=20
>>>>=20
>>>>=20
>>> __________________________________________________________
>>>>=20
>>> __________________________________________________________
>>>> _____
>>>>=20
>>>> Ce message et ses pieces jointes peuvent contenir des informations
>>>> confidentielles ou privilegiees et ne doivent donc pas etre
>>>> diffuses, exploites ou copies sans autorisation. Si vous avez recu
>>>> ce message par erreur, veuillez le signaler a l'expediteur et le
>>>> detruire ainsi que les pieces jointes. Les messages electroniques
>>>> etant susceptibles d'alteration, Orange decline toute =
responsabilite
>>>> si ce message a ete altere, deforme ou falsifie. Merci.
>>>>=20
>>>> This message and its attachments may contain confidential or
>>>> privileged information that may be protected by law; they should =
not
>>>> be distributed, used or copied without authorisation.
>>>> If you have received this email in error, please notify the sender
>>>> and delete this message and its attachments.
>>>> As emails may be altered, Orange is not liable for messages that
>>>> have been modified, changed or falsified.
>>>> Thank you.
>>>=20
>>> _______________________________________________
>>> tcpm mailing list
>>> tcpm@ietf.org
>>> https://www.ietf.org/mailman/listinfo/tcpm
>=20
> =
__________________________________________________________________________=
_______________________________________________
>=20
> Ce message et ses pieces jointes peuvent contenir des informations =
confidentielles ou privilegiees et ne doivent donc
> pas etre diffuses, exploites ou copies sans autorisation. Si vous avez =
recu ce message par erreur, veuillez le signaler
> a l'expediteur et le detruire ainsi que les pieces jointes. Les =
messages electroniques etant susceptibles d'alteration,
> Orange decline toute responsabilite si ce message a ete altere, =
deforme ou falsifie. Merci.
>=20
> This message and its attachments may contain confidential or =
privileged information that may be protected by law;
> they should not be distributed, used or copied without authorisation.
> If you have received this email in error, please notify the sender and =
delete this message and its attachments.
> As emails may be altered, Orange is not liable for messages that have =
been modified, changed or falsified.
> Thank you.
>=20
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm


--Apple-Mail=_B8EEAA2F-6A8D-4449-9DC3-22E4D82DC951
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=utf-8

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3Dutf-8"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: after-white-space;" class=3D"">Hi, =
Med,<div class=3D""><br class=3D""></div><div class=3D"">I disagree. The =
addresses in test vectors are not examples for documentation, but rather =
an actual configuration intended for testing purposes.</div><div =
class=3D""><br class=3D""></div><div class=3D"">I found only one other =
RFC that provides test vectors based on packet fields that include IP =
addresses: RFC&nbsp;5769, but it (IMO) incorrectly uses the =
documentation addresses you suggest (I expect to submit an errata on =
that shortly - let me know if you know of others).</div><div =
class=3D""><br class=3D""></div><div class=3D"">Joe</div><div =
class=3D""><br class=3D""><div class=3D"">
<meta charset=3D"UTF-8" class=3D""><div dir=3D"auto" style=3D"caret-color:=
 rgb(0, 0, 0); color: rgb(0, 0, 0); letter-spacing: normal; text-align: =
start; text-indent: 0px; text-transform: none; white-space: normal; =
word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: =
none; word-wrap: break-word; -webkit-nbsp-mode: space; line-break: =
after-white-space;" class=3D"">=E2=80=94<div class=3D"">Joe Touch, =
temporal epistemologist<div class=3D""><a =
href=3D"http://www.strayalpha.com" =
class=3D"">www.strayalpha.com</a></div></div></div>
</div>
<div><br class=3D""><blockquote type=3D"cite" class=3D""><div =
class=3D"">On Feb 8, 2022, at 12:39 AM, <a =
href=3D"mailto:mohamed.boucadair@orange.com" =
class=3D"">mohamed.boucadair@orange.com</a> wrote:</div><br =
class=3D"Apple-interchange-newline"><div class=3D""><div class=3D"">Hi =
Michael, <br class=3D""><br class=3D"">Thank you. <br class=3D""><br =
class=3D"">There is one minor comment about the ipv6 addresses in the =
examples: they should be derived from the prefix in RFC 3849. This =
should be fixed in draft-ietf-tcpm-ao-test-vectors (the same applies for =
10.11.12.13, etcs that are not from the blocks defined in 5737). <br =
class=3D""><br class=3D"">Other than that, this looks good to me. =
Thanks.<br class=3D""><br class=3D"">Cheers,<br class=3D"">Med<br =
class=3D""><br class=3D""><blockquote type=3D"cite" =
class=3D"">-----Message d'origine-----<br class=3D"">De&nbsp;: Scharf, =
Michael &lt;<a href=3D"mailto:Michael.Scharf@hs-esslingen.de" =
class=3D"">Michael.Scharf@hs-esslingen.de</a>&gt;<br =
class=3D"">Envoy=C3=A9&nbsp;: vendredi 4 f=C3=A9vrier 2022 22:12<br =
class=3D"">=C3=80&nbsp;: BOUCADAIR Mohamed INNOV/NET &lt;<a =
href=3D"mailto:mohamed.boucadair@orange.com" =
class=3D"">mohamed.boucadair@orange.com</a>&gt;<br class=3D"">Cc&nbsp;: =
<a href=3D"mailto:tcpm@ietf.org" class=3D"">tcpm@ietf.org</a><br =
class=3D"">Objet&nbsp;: RE: [tcpm] I-D Action: =
draft-ietf-tcpm-yang-tcp-04.txt<br class=3D""><br class=3D"">For what it =
is worth, -06 with the mentioned changes has been published.<br =
class=3D""><br class=3D"">The diff can be found at: <a =
href=3D"https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-" =
class=3D"">https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-</a><br =
class=3D"">tcpm-yang-tcp-06.txt<br class=3D""><br class=3D"">I hope that =
this update fixes all issues.<br class=3D""><br class=3D"">Michael<br =
class=3D""><br class=3D""><blockquote type=3D"cite" =
class=3D"">-----Original Message-----<br class=3D"">From: tcpm &lt;<a =
href=3D"mailto:tcpm-bounces@ietf.org" =
class=3D"">tcpm-bounces@ietf.org</a>&gt; On Behalf Of Scharf, Michael<br =
class=3D"">Sent: Tuesday, January 18, 2022 10:32 PM<br class=3D"">To: <a =
href=3D"mailto:mohamed.boucadair@orange.com" =
class=3D"">mohamed.boucadair@orange.com</a><br class=3D"">Cc: <a =
href=3D"mailto:tcpm@ietf.org" class=3D"">tcpm@ietf.org</a><br =
class=3D"">Subject: Re: [tcpm] I-D Action: =
draft-ietf-tcpm-yang-tcp-04.txt<br class=3D""><br class=3D"">Hi Med,<br =
class=3D""><br class=3D"">Thanks a lot for this further review with good =
catches.<br class=3D""><br class=3D"">We have addressed all comments in =
the current working copy for -06,<br class=3D"">except for the =
disclaimer text regarding line wrap in B.1. This<br class=3D"">disclaimer =
is currently automatically introduced by the build tools no<br =
class=3D""></blockquote>matter whether it is needed.<br =
class=3D""><blockquote type=3D"cite" class=3D"">That detail can be =
sorted out later by the RFC editor.<br class=3D""><br class=3D"">Details =
can be found at the repository with the working copy:<br class=3D""><a =
href=3D"https://github.com/mjethanandani/ietf-tcp" =
class=3D"">https://github.com/mjethanandani/ietf-tcp</a><br class=3D""><br=
 class=3D"">Our suggestion is to change the TCP-AO example so that it =
illustrates<br class=3D"">the use of send-id and recv-id with parameters =
documented in<br class=3D"">draft-ietf-tcpm-ao- test-vectors.<br =
class=3D""><br class=3D"">As the WGLC is ongoing, we would wait for =
further comments before<br class=3D"">finalizing -06.<br class=3D""><br =
class=3D"">Thanks<br class=3D""><br class=3D"">Michael (as author)<br =
class=3D""><br class=3D""><br class=3D""><blockquote type=3D"cite" =
class=3D"">-----Original Message-----<br class=3D"">From: =
mohamed.boucadair@orange.com<br =
class=3D"">&lt;mohamed.boucadair@orange.com&gt;<br class=3D"">Sent: =
Monday, January 3, 2022 10:13 AM<br class=3D"">To: Scharf, Michael =
&lt;Michael.Scharf@hs-esslingen.de&gt;<br class=3D"">Cc: =
tcpm@ietf.org<br class=3D"">Subject: RE: [tcpm] I-D Action: =
draft-ietf-tcpm-yang-tcp-04.txt<br class=3D""><br class=3D"">Hi =
Michael,<br class=3D""><br class=3D"">The changes look good. Thank =
you.<br class=3D""><br class=3D"">Please find below some comments on the =
new version:<br class=3D"">- Section 4<br class=3D""><br class=3D"">(1)<br=
 class=3D"">OLD:<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;reference<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;"RFC XXXX, YANG Model =
for Transmission Control Protocol<br =
class=3D""></blockquote></blockquote>(TCP)<br class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Configuration.";<br class=3D""><br=
 class=3D"">NEW:<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;reference<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;"RFC XXXX: A YANG Model =
for Transmission Control Protocol<br =
class=3D""></blockquote></blockquote>(TCP)<br class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Configuration";<br class=3D""><br =
class=3D"">(2)<br class=3D""><br class=3D"">Add "when set to true" in =
the description of almost all the Boolean<br =
class=3D""></blockquote></blockquote>nodes.<br class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""><br =
class=3D"">(3)<br class=3D""><br class=3D"">I still don't think "forms =
the connection identifier" is accurate.<br class=3D"">The<br =
class=3D""></blockquote>connection<br class=3D""><blockquote type=3D"cite"=
 class=3D"">is identified by the 4-uple (which is captured in the key<br =
class=3D"">statement). You<br class=3D""></blockquote>may<br =
class=3D""><blockquote type=3D"cite" class=3D"">consider simply: s/forms =
the connection identifier/of the<br =
class=3D""></blockquote></blockquote>connection.<br class=3D""><blockquote=
 type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D"">Another =
better approach would be, e.g.:<br class=3D""><br class=3D"">OLD:<br =
class=3D""> "Local address that forms the connection identifier.";<br =
class=3D"">NEW:<br class=3D""> "Identifies the address that is used =
locally by an endpoint to bind<br class=3D"">the connection.";<br =
class=3D""><br class=3D"">OLD:<br class=3D""> "Remote address that forms =
the connection identifier.";<br class=3D"">NEW:<br class=3D""> =
"Identifies the address that is used by the peer endpoint to bind<br =
class=3D"">the connection.";<br class=3D""><br class=3D"">etc.<br =
class=3D""><br class=3D"">- Make this change in 5.2: s/registrations =
are<br class=3D"">requested/registration is requested<br class=3D"">- =
Update 5.2 to make it explicit that the module is not maintained<br =
class=3D""></blockquote></blockquote>by IANA:<br class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""><br =
class=3D"">OLD:<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;name: =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;ietf-tcp<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;namespace: =
&nbsp;&nbsp;&nbsp;urn:ietf:params:xml:ns:yang:ietf-tcp<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;prefix: =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;tcp<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;reference: &nbsp;&nbsp;&nbsp;RFC XXXX<br =
class=3D""><br class=3D"">NEW:<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;name: =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;ietf-tcp<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;namespace: =
&nbsp;&nbsp;&nbsp;urn:ietf:params:xml:ns:yang:ietf-tcp<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;prefix: =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;tcp<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;maintained by IANA: N<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;reference: &nbsp;&nbsp;&nbsp;RFC XXXX<br =
class=3D""><br class=3D"">- I think you can remove the note in B.1<br =
class=3D"">- remove "\" right after "&lt;/description&gt;" in B.2<br =
class=3D"">- As you are referring to RFC 8792, you need to add the 8792 =
header<br class=3D"">to adhere with the following (B.2):<br class=3D""><br=
 class=3D"">=3D=3D=3D=3D=3D=3D<br class=3D"">...<br class=3D""> =
&nbsp;&nbsp;Text content that has been folded as specified by this =
strategy<br class=3D""></blockquote></blockquote>MUST<br =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D""> &nbsp;&nbsp;adhere to the following structure.<br =
class=3D""><br class=3D"">7.1.1. &nbsp;Header<br class=3D""><br =
class=3D""> &nbsp;&nbsp;The header is two lines long.<br class=3D""><br =
class=3D""> &nbsp;&nbsp;The first line is the following 36-character =
string; this string<br class=3D""></blockquote></blockquote>MAY<br =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D""> &nbsp;&nbsp;be surrounded by any number of printable =
characters. &nbsp;This first<br =
class=3D""></blockquote></blockquote>line<br class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""> =
&nbsp;&nbsp;cannot itself be folded.<br class=3D""><br class=3D""> =
&nbsp;&nbsp;NOTE: '\' line wrapping per RFC 8792<br class=3D""><br =
class=3D""> &nbsp;&nbsp;The second line is an empty line, containing =
only the end-of-line<br class=3D""> &nbsp;&nbsp;character sequence. =
&nbsp;This line provides visual separation for<br class=3D""> =
&nbsp;&nbsp;readability.<br class=3D"">=3D=3D=3D=3D=3D<br class=3D""><br =
class=3D"">- B.2:<br class=3D""><br class=3D"">&lt;!--<br class=3D"">This =
example sets TCP-AO configuration parameters as demonstrated by<br =
class=3D"">examples in draft-touch-tcpm-ao-test-vectors.<br =
class=3D"">--&gt;<br class=3D""><br class=3D"">Not sure why this I-D is =
cited here given the example does not echo<br class=3D"">an example from =
that I-D (e.g., same local/remote port, local<br =
class=3D""></blockquote></blockquote>address, etc.).<br =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D"">Also, I was expecting to see an example that illustrates the =
use of<br class=3D"">send-id and recv-id.<br class=3D""><br class=3D"">If =
the citation is maintained, please make this change:<br class=3D""><br =
class=3D"">s/demonstrated by examples in draft-touch-tcpm-ao-test-<br =
class=3D"">vectors/demonstrated by examples in =
I-D.ietf-tcpm-ao-test-vectors<br class=3D""><br class=3D"">Cheers,<br =
class=3D"">Med<br class=3D""><br class=3D""><blockquote type=3D"cite" =
class=3D"">-----Message d'origine-----<br class=3D"">De&nbsp;: Scharf, =
Michael &lt;<a href=3D"mailto:Michael.Scharf@hs-esslingen.de" =
class=3D"">Michael.Scharf@hs-esslingen.de</a>&gt; Envoy=C3=A9&nbsp;:<br =
class=3D"">jeudi 30 d=C3=A9cembre 2021 15:19 =C3=80&nbsp;: BOUCADAIR =
Mohamed INNOV/NET<br class=3D""></blockquote>&lt;<a =
href=3D"mailto:mohamed.boucadair@orange.com" =
class=3D"">mohamed.boucadair@orange.com</a>&gt;<br class=3D""><blockquote =
type=3D"cite" class=3D"">Cc&nbsp;: <a href=3D"mailto:tcpm@ietf.org" =
class=3D"">tcpm@ietf.org</a><br class=3D"">Objet&nbsp;: RE: [tcpm] I-D =
Action: draft-ietf-tcpm-yang-tcp-04.txt<br class=3D""><br class=3D"">Hi =
Med,<br class=3D""><br class=3D"">Thanks a lot for the detailed review. =
I went through all your<br =
class=3D""></blockquote></blockquote></blockquote>comments.<br =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D""><blockquote type=3D"cite" class=3D"">Most were =
straightforward to address.<br class=3D""><br class=3D"">Our suggestion =
is to clearly state that MPTCP is outside the scope<br class=3D"">of =
this model. MPTCP may need a different YANG model. Yet, the<br =
class=3D"">most obvious<br class=3D""></blockquote>use<br =
class=3D""><blockquote type=3D"cite" class=3D"">case of the YANG model =
are TCP-based control plane protocols on<br =
class=3D""></blockquote></blockquote>routers<br class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D"">(BGP, LDP, =
...). At least I am not aware of widespread use of BGP<br class=3D"">over =
MPTCP.<br class=3D""><br class=3D"">Can you please have a look at -05 =
and let me know if -05 works for<br =
class=3D""></blockquote></blockquote></blockquote>you?<br =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D""><blockquote type=3D"cite" class=3D""><br class=3D"">The full =
diff is available at:<br class=3D""><a =
href=3D"https://www.ietf.org/rfcdiff?url2=3Ddraft-" =
class=3D"">https://www.ietf.org/rfcdiff?url2=3Ddraft-</a><br =
class=3D"">ietf-tcpm-yang-tcp-05<br class=3D""><br class=3D"">Thanks<br =
class=3D""><br class=3D"">Michael<br class=3D""><br class=3D""><blockquote=
 type=3D"cite" class=3D"">-----Original Message-----<br class=3D"">From: =
mohamed.boucadair@orange.com<br =
class=3D"">&lt;mohamed.boucadair@orange.com&gt;<br class=3D"">Sent: =
Friday, November 12, 2021 11:17 AM<br class=3D"">To: Scharf, Michael =
&lt;Michael.Scharf@hs-esslingen.de&gt;<br class=3D"">Cc: =
tcpm@ietf.org<br class=3D"">Subject: RE: [tcpm] I-D Action: =
draft-ietf-tcpm-yang-tcp-04.txt<br class=3D""><br class=3D"">Hi =
Michael,<br class=3D""><br class=3D"">FWIW, please find below some =
comments to the latest version of<br class=3D"">the<br =
class=3D""></blockquote>draft:<br class=3D""><blockquote type=3D"cite" =
class=3D""><br class=3D"">* pdf: =
https://raw.githubusercontent.com/boucadair/IETF-Drafts-<br =
class=3D"">Reviews/master/draft-ietf-tcpm-yang-tcp-04-rev%20Med.pdf<br =
class=3D"">* doc: https://github.com/boucadair/IETF-Drafts-<br =
class=3D"">Reviews/raw/master/draft-ietf-tcpm-yang-tcp-04-rev%20Med.doc<br=
 class=3D""><br class=3D"">I think that the document is almost stable. =
Thank you.<br class=3D""><br class=3D"">Cheers,<br class=3D"">Med<br =
class=3D""><br class=3D""><blockquote type=3D"cite" =
class=3D"">-----Message d'origine-----<br class=3D"">De&nbsp;: tcpm =
&lt;tcpm-bounces@ietf.org&gt; De la part de<br =
class=3D"">internet-drafts@ietf.org Envoy=C3=A9&nbsp;: lundi 25 octobre =
2021 20:22<br =
class=3D""></blockquote></blockquote></blockquote></blockquote></blockquot=
e>=C3=80&nbsp;:<br class=3D""><blockquote type=3D"cite" =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D""><a href=3D"mailto:i-d-announce@ietf.org" =
class=3D"">i-d-announce@ietf.org</a> Cc&nbsp;: <a =
href=3D"mailto:tcpm@ietf.org" class=3D"">tcpm@ietf.org</a> Objet&nbsp;: =
[tcpm] I-D<br =
class=3D""></blockquote></blockquote></blockquote></blockquote></blockquot=
e>Action:<br class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D"">draft-ietf-tcpm-yang-tcp-04.txt<br class=3D""><br =
class=3D""><br class=3D"">A New Internet-Draft is available from the =
on-line<br class=3D"">Internet-Drafts directories.<br class=3D"">This =
draft is a work item of the TCP Maintenance and Minor<br =
class=3D"">Extensions WG of the IETF.<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Title =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: YANG Model =
for Transmission Control<br =
class=3D""></blockquote></blockquote></blockquote></blockquote></blockquot=
e>Protocol<br class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D"">(TCP) =
Configuration<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Authors =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: Michael Scharf<br =
class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;Mahesh Jethanandani<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&n=
bsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbs=
p;Vishal Murgai<br class=3D""><span class=3D"Apple-tab-span" =
style=3D"white-space:pre">	</span>Filename =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: =
draft-ietf-tcpm-yang-tcp-04.txt<br class=3D""><span =
class=3D"Apple-tab-span" style=3D"white-space:pre">	</span>Pages =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: 23<br =
class=3D""><span class=3D"Apple-tab-span" style=3D"white-space:pre">	=
</span>Date =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;: =
2021-10-25<br class=3D""><br class=3D"">Abstract:<br class=3D""> =
&nbsp;&nbsp;This document specifies a minimal YANG model for TCP on<br =
class=3D"">devices<br class=3D""></blockquote></blockquote>that<br =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D""> &nbsp;&nbsp;are configured by network management protocols. =
&nbsp;The YANG<br =
class=3D""></blockquote></blockquote></blockquote></blockquote>model<br =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D""> &nbsp;&nbsp;defines a container for all TCP connections and =
groupings<br =
class=3D""></blockquote></blockquote></blockquote></blockquote></blockquot=
e>of<br class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""> =
&nbsp;&nbsp;authentication parameters that can be imported and used =
in<br =
class=3D""></blockquote></blockquote></blockquote></blockquote></blockquot=
e>TCP<br class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""> =
&nbsp;&nbsp;implementations or by other models that need to configure<br =
class=3D""></blockquote></blockquote></blockquote></blockquote></blockquot=
e>TCP<br class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""> =
&nbsp;&nbsp;parameters. &nbsp;The model also includes basic TCP =
statistics.<br class=3D"">The<br =
class=3D""></blockquote></blockquote>model<br class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""> =
&nbsp;&nbsp;is NMDA (RFC 8342) compliant.<br class=3D""><br class=3D""><br=
 class=3D"">The IETF datatracker status page for this draft is:<br =
class=3D""><a =
href=3D"https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/" =
class=3D"">https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/</a><=
br class=3D""><br class=3D"">There is also an htmlized version available =
at:<br =
class=3D"">https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-yang-tcp<=
br class=3D"">-04<br class=3D""><br class=3D"">A diff from the previous =
version is available at:<br =
class=3D"">https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-yang-tcp-04=
<br class=3D""><br class=3D""><br class=3D"">Internet-Drafts are also =
available by anonymous FTP at:<br =
class=3D"">ftp://ftp.ietf.org/internet-drafts/<br class=3D""><br =
class=3D""><br =
class=3D"">_______________________________________________<br =
class=3D"">tcpm mailing list<br class=3D"">tcpm@ietf.org<br =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm<br =
class=3D""></blockquote><br class=3D""><br =
class=3D""></blockquote></blockquote><br =
class=3D""></blockquote>__________________________________________________=
________<br class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D""><br =
class=3D""></blockquote></blockquote><br =
class=3D""></blockquote>__________________________________________________=
________<br class=3D""><blockquote type=3D"cite" class=3D""><blockquote =
type=3D"cite" class=3D""><blockquote type=3D"cite" class=3D"">_____<br =
class=3D""><br class=3D"">Ce message et ses pieces jointes peuvent =
contenir des<br class=3D"">informations confidentielles ou privilegiees =
et ne doivent donc<br class=3D"">pas etre diffuses, exploites ou copies =
sans autorisation. Si<br class=3D"">vous avez recu ce message par =
erreur, veuillez le signaler a<br class=3D"">l'expediteur et le detruire =
ainsi que les pieces jointes. Les<br class=3D"">messages electroniques =
etant susceptibles d'alteration, Orange<br class=3D"">decline toute =
responsabilite si ce message a ete altere, deforme<br =
class=3D""></blockquote></blockquote></blockquote></blockquote>ou =
falsifie. Merci.<br class=3D""><blockquote type=3D"cite" =
class=3D""><blockquote type=3D"cite" class=3D""><blockquote type=3D"cite" =
class=3D""><blockquote type=3D"cite" class=3D""><br class=3D"">This =
message and its attachments may contain confidential or<br =
class=3D"">privileged information that may be protected by law; they =
should<br class=3D"">not be distributed, used or copied without =
authorisation.<br class=3D"">If you have received this email in error, =
please notify the<br class=3D"">sender and delete this message and its =
attachments.<br class=3D"">As emails may be altered, Orange is not =
liable for messages that<br class=3D"">have been modified, changed or =
falsified.<br class=3D"">Thank you.<br =
class=3D""></blockquote></blockquote><br class=3D""><br class=3D""><br =
class=3D""></blockquote>__________________________________________________=
________<br class=3D""><blockquote type=3D"cite" class=3D""><br =
class=3D""></blockquote>__________________________________________________=
________<br class=3D""><blockquote type=3D"cite" class=3D"">_____<br =
class=3D""><br class=3D"">Ce message et ses pieces jointes peuvent =
contenir des informations<br class=3D"">confidentielles ou privilegiees =
et ne doivent donc pas etre<br class=3D"">diffuses, exploites ou copies =
sans autorisation. Si vous avez recu<br class=3D"">ce message par =
erreur, veuillez le signaler a l'expediteur et le<br class=3D"">detruire =
ainsi que les pieces jointes. Les messages electroniques<br =
class=3D"">etant susceptibles d'alteration, Orange decline toute =
responsabilite<br class=3D"">si ce message a ete altere, deforme ou =
falsifie. Merci.<br class=3D""><br class=3D"">This message and its =
attachments may contain confidential or<br class=3D"">privileged =
information that may be protected by law; they should not<br class=3D"">be=
 distributed, used or copied without authorisation.<br class=3D"">If you =
have received this email in error, please notify the sender<br =
class=3D"">and delete this message and its attachments.<br class=3D"">As =
emails may be altered, Orange is not liable for messages that<br =
class=3D"">have been modified, changed or falsified.<br class=3D"">Thank =
you.<br class=3D""></blockquote><br =
class=3D"">_______________________________________________<br =
class=3D"">tcpm mailing list<br class=3D""><a =
href=3D"mailto:tcpm@ietf.org" class=3D"">tcpm@ietf.org</a><br =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm<br =
class=3D""></blockquote></blockquote><br =
class=3D"">_______________________________________________________________=
__________________________________________________________<br =
class=3D""><br class=3D"">Ce message et ses pieces jointes peuvent =
contenir des informations confidentielles ou privilegiees et ne doivent =
donc<br class=3D"">pas etre diffuses, exploites ou copies sans =
autorisation. Si vous avez recu ce message par erreur, veuillez le =
signaler<br class=3D"">a l'expediteur et le detruire ainsi que les =
pieces jointes. Les messages electroniques etant susceptibles =
d'alteration,<br class=3D"">Orange decline toute responsabilite si ce =
message a ete altere, deforme ou falsifie. Merci.<br class=3D""><br =
class=3D"">This message and its attachments may contain confidential or =
privileged information that may be protected by law;<br class=3D"">they =
should not be distributed, used or copied without authorisation.<br =
class=3D"">If you have received this email in error, please notify the =
sender and delete this message and its attachments.<br class=3D"">As =
emails may be altered, Orange is not liable for messages that have been =
modified, changed or falsified.<br class=3D"">Thank you.<br class=3D""><br=
 class=3D"">_______________________________________________<br =
class=3D"">tcpm mailing list<br class=3D""><a =
href=3D"mailto:tcpm@ietf.org" class=3D"">tcpm@ietf.org</a><br =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm<br =
class=3D""></div></div></blockquote></div><br =
class=3D""></div></body></html>=

--Apple-Mail=_B8EEAA2F-6A8D-4449-9DC3-22E4D82DC951--


From nobody Tue Feb  8 10:56:43 2022
Return-Path: <noreply@ietf.org>
X-Original-To: tcpm@ietf.org
Delivered-To: tcpm@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 3BA023A10B2; Tue,  8 Feb 2022 10:56:39 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: Yoshifumi Nishida via Datatracker <noreply@ietf.org>
To: <martin.h.duke@gmail.com>
X-Test-IDTracker: no
X-IETF-IDTracker: 7.44.0
Auto-Submitted: auto-generated
Precedence: bulk
Cc: iesg-secretary@ietf.org, nsd.ietf@gmail.com, tcpm-chairs@ietf.org, tcpm@ietf.org
Message-ID: <164434659922.27235.9459544648717493656@ietfa.amsl.com>
Date: Tue, 08 Feb 2022 10:56:39 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/Scp0YMDwtbl31WGNhzXtSjGlWNg>
Subject: [tcpm] Publication has been requested for draft-ietf-tcpm-yang-tcp-06
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Feb 2022 18:56:40 -0000

Yoshifumi Nishida has requested publication of draft-ietf-tcpm-yang-tcp-06 as None on behalf of the TCPM working group.

Please verify the document's state at https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/



From nobody Tue Feb  8 13:01:11 2022
Return-Path: <internet-drafts@ietf.org>
X-Original-To: tcpm@ietf.org
Delivered-To: tcpm@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 4CEB33A0912; Tue,  8 Feb 2022 13:00:38 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <i-d-announce@ietf.org>
Cc: tcpm@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 7.44.0
Auto-Submitted: auto-generated
Precedence: bulk
Reply-To: tcpm@ietf.org
Message-ID: <164435403825.23361.17582972013699039226@ietfa.amsl.com>
Date: Tue, 08 Feb 2022 13:00:38 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/tar8KppdEOhPEfP61M5UErYwZtg>
Subject: [tcpm] I-D Action: draft-ietf-tcpm-rfc793bis-26.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Feb 2022 21:00:39 -0000

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the TCP Maintenance and Minor Extensions WG of the IETF.

        Title           : Transmission Control Protocol (TCP) Specification
        Author          : Wesley M. Eddy
	Filename        : draft-ietf-tcpm-rfc793bis-26.txt
	Pages           : 114
	Date            : 2022-02-08

Abstract:
   This document specifies the Transmission Control Protocol (TCP).  TCP
   is an important transport layer protocol in the Internet protocol
   stack, and has continuously evolved over decades of use and growth of
   the Internet.  Over this time, a number of changes have been made to
   TCP as it was specified in RFC 793, though these have only been
   documented in a piecemeal fashion.  This document collects and brings
   those changes together with the protocol specification from RFC 793.
   This document obsoletes RFC 793, as well as RFCs 879, 2873, 6093,
   6429, 6528, and 6691 that updated parts of RFC 793.  It updates RFCs
   1011 and 1122, and should be considered as a replacement for the
   portions of those document dealing with TCP requirements.  It also
   updates RFC 5961 by adding a small clarification in reset handling
   while in the SYN-RECEIVED state.  The TCP header control bits from
   RFC 793 have also been updated based on RFC 3168.

   RFC EDITOR NOTE: If approved for publication as an RFC, this should
   be marked additionally as "STD: 7" and replace RFC 793 in that role.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc793bis/

There is also an htmlized version available at:
https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-rfc793bis-26

A diff from the previous version is available at:
https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rfc793bis-26


Internet-Drafts are also available by rsync at rsync.ietf.org::internet-drafts



From nobody Wed Feb  9 02:01:32 2022
Return-Path: <nsd.ietf@gmail.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D12D13A1375 for <tcpm@ietfa.amsl.com>; Wed,  9 Feb 2022 02:01:29 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.097
X-Spam-Level: 
X-Spam-Status: No, score=-2.097 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id to6_r2iBvE4d for <tcpm@ietfa.amsl.com>; Wed,  9 Feb 2022 02:01:25 -0800 (PST)
Received: from mail-qv1-xf34.google.com (mail-qv1-xf34.google.com [IPv6:2607:f8b0:4864:20::f34]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 431FD3A127F for <tcpm@ietf.org>; Wed,  9 Feb 2022 02:01:25 -0800 (PST)
Received: by mail-qv1-xf34.google.com with SMTP id a28so1326008qvb.10 for <tcpm@ietf.org>; Wed, 09 Feb 2022 02:01:25 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112;  h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=q/Lv0bWb84NsoovzMiP+tLafQdybne4hUTPVoecexII=; b=nLdEKQmLIMyzDe1JNLL0uGIm18OjwSc8ZNz6Zlpeqb4GLgg+rm+0bNGD1mDT4KHdqn /KPyLt2uuX/WnTfN7+4sselunFnOK7P0WQ/FdTfwE7BEOr1Rj5eaRWf/rlQNeFWNje2X si75RnGTmgyWRTagy0pC81gM9WC6Zj0fBDBk0xPaf/bPnhagy3RJYQ8qp4kADljx5CMp Qeodc0QfOzjDF5Wu4xRRNSNsFJkHRg9XXSOLqGv2h1tEFUHjTtbgskaR8TUWswlIxN7C OlmLLuWgC99qgLuz9WS9eqjxdGO6LaklErakF9OAZjFsdvJwGwXu9UHgm8BFDdLfsblF Wn4w==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=q/Lv0bWb84NsoovzMiP+tLafQdybne4hUTPVoecexII=; b=M1K70/PuvonqXocIw3wQL4e1MlEVUSxn83mPi/ED97aysk8T4ZMUKdPzZymWDCvtlV OLVFtzvlBzfBv6V0AC/5/kIagUlFvN/xhtFg6jtqJeeVZzTzkMH6r0P9PTj4cpZIgMfD QSqC00FE6in036Ze7R8i1xXl7ThQTqB1J6GpO8pDvuisM5HGFq1SJl+SIzZvOYe3oslD k14OzfWV9Hq5Ta6eg00kg6D5Lsxkac2+W3QDwLiopFY+Yq5Ai6wkIoIgnEmLVDAn5fqS YoYHlVAyeBeMqrL3sJWEUinMQf0LGYRogXmGHNKa1tOE0azRW8Jsj6+lmIYAJ+KdWVRC YoxQ==
X-Gm-Message-State: AOAM533mUYPMzJAol4PmQhw/AgOEsUEr4jy4klDqB8KsTkaNTrOx2Vbs vW7ns+bcHV9jBA8JDJIRWrDXEVySkEVfr1MA5lRFio0g
X-Google-Smtp-Source: ABdhPJwnzhSXUriLrMp6tIa9mkzamAm3B6qIYum+VrwBVuIOhIfC3u/B2KX9MkSVG/eNwPEgkHmkGha304eiqZnAeZY=
X-Received: by 2002:a05:6214:5294:: with SMTP id kj20mr890729qvb.111.1644400883465;  Wed, 09 Feb 2022 02:01:23 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <CAAK044R6B7HT0mdSDwHU=FhvN=A=M-8pxYb8wGfwxczwtCT-yw@mail.gmail.com> <492FC264-2D3C-4D3D-B7C2-D547593E4A7D@apple.com>
In-Reply-To: <492FC264-2D3C-4D3D-B7C2-D547593E4A7D@apple.com>
From: Yoshifumi Nishida <nsd.ietf@gmail.com>
Date: Wed, 9 Feb 2022 02:01:12 -0800
Message-ID: <CAAK044QXt8oHQ5qkqX3dKEuXKuXQTjiuRO=DKZkkrD7DeH4EOQ@mail.gmail.com>
To: Vidhi Goel <vidhi_goel@apple.com>
Cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>
Content-Type: multipart/alternative; boundary="00000000000086feb805d792ea1b"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/E8FSjFBCOZSDkrRrMY-kTLzXvkk>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Feb 2022 10:01:30 -0000

--00000000000086feb805d792ea1b
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi Vidhi,

Thanks for the comments.

On Mon, Feb 7, 2022 at 3:32 PM Vidhi Goel <vidhi_goel@apple.com> wrote:

> Thanks Yoshi for reviewing the draft.
>
> So, it seems one PS doc mentions not to use cwnd, while another PS doc
> mentions using cwnd is fine (with RFC7661).
>
> RFC5681 is conservative in its approach to use FlightSize and will result
> in very low throughput in scenarios when the congestion event occurs at t=
he
> tail of send window (when flight size could be very small)
>

Yes. I agree.


>
> I am not sure which can cause this situation yet, but in any case, does
> this mean there is no concern on
> "some implementations may incidentally increase well beyond rwnd" in cubi=
c?
> Is this because this is an old implementation bug or are there other
> reasons?
> I am wondering if we might want some clarification on this point.
>
> If an implementation uses RFC7661, there is no concern about increasing
> cwnd beyond rwnd as RFC7661 tracks bytes acknowledged in a RTT which mean=
s
> that `rwnd` check has already been performed.
> Usually cwnd doesn=E2=80=99t grow beyond rwnd  (ACK-clocking increases cw=
nd which
> means the sender already checked `rwnd` before sending out the data).
>
> But there could be implementations that may want to keep `rwnd` even
> smaller than initial `cwnd`. In those cases, RFC7661 would make the cwnd
> invalidated when the number of acknowledged bytes is much lower than the
> current cwnd.
>
> Do you have any suggestions on how to make it more clear?
>

I think it will be fine as long as the draft clarifies we can safely
override what's written in RFC5681 when we have some mechanisms such as
RFC7661.

> Also, I am wondering if we might want to use "MAY be used" here instead o=
f
> "can be used"
>
> Yeah, we can make this change.
>

Thanks!
--=20
Yoshi


>
>
> Thanks,
> Vidhi
>
> On Feb 6, 2022, at 3:01 AM, Yoshifumi Nishida <nsd.ietf@gmail.com> wrote:
>
> Hi,
> I just have one more comment on the draft.
> In RFC5681, it mentions in Section 3:
> ''
>
>    Implementation Note: An easy mistake to make is to simply use cwnd,
>    rather than FlightSize, which in some implementations may
>    incidentally increase well beyond rwnd.
>
> "
>
> while this draft mentions in Section 4.6:
> "
>
>    To avoid suboptimal performance with such applications,
>
>    the mechanisms described in [RFC7661 <https://datatracker.ietf.org/doc=
/html/rfc7661>] can be used to mitigate this
>    issue as it would allow using a value between _cwnd_ and
>    _flight_size_ to calculate the new _ssthresh_ in Figure 5.
>
> "
>
> So, it seems one PS doc mentions not to use cwnd, while another PS doc
> mentions using cwnd is fine (with RFC7661).
>
> I am not sure which can cause this situation yet, but in any case, does
> this mean there is no concern on
> "some implementations may incidentally increase well beyond rwnd" in cubi=
c?
> Is this because this is an old implementation bug or are there other
> reasons?
> I am wondering if we might want some clarification on this point.
>
> Also, I am wondering if we might want to use "MAY be used" here instead o=
f
> "can be used"
>
> Thanks,
> --
> Yoshi
>
> On Mon, Jan 31, 2022 at 11:43 PM Yoshifumi Nishida <nsd.ietf@gmail.com>
> wrote:
>
>> Hello,
>>
>> After some discussions among chairs, we decided to run the 2nd WGLC on
>> draft-ietf-tcpm-rfc8312bis in consideration of the importance of the dra=
ft.
>> We'll be grateful if you could send your feedback to the ML. The WGLC
>> runs until *Feb 11*.
>>
>> If interested, you can check in-depth past discussions in the following
>> URL.
>> https://github.com/NTAP/rfc8312bis/
>>
>> Thank you so much!
>> --
>> tcpm co-chairs
>>
>>
>> On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <lars@eggert.org> wrote:
>>
>>> Hi,
>>>
>>> this -06 version rolls in all the changes requested during (and after)
>>> WGLC ended.
>>>
>>> I'll leave it up to the chairs to decide if another WGLC is warranted o=
r
>>> the document can progress as-is.
>>>
>>> Thanks,
>>> Lars
>>>
>>>
>>> > On 2022-1-26, at 11:12, 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 TCP Maintenance and Minor Extensions
>>> WG of the IETF.
>>> >
>>> >        Title           : CUBIC for Fast and Long-Distance Networks
>>> >        Authors         : Lisong Xu
>>> >                          Sangtae Ha
>>> >                          Injong Rhee
>>> >                          Vidhi Goel
>>> >                          Lars Eggert
>>> >       Filename        : draft-ietf-tcpm-rfc8312bis-06.txt
>>> >       Pages           : 35
>>> >       Date            : 2022-01-26
>>> >
>>> > Abstract:
>>> >   CUBIC is a standard TCP congestion control algorithm that uses a
>>> >   cubic function instead of a linear congestion window increase
>>> >   function to improve scalability and stability over fast and long-
>>> >   distance networks.  CUBIC has been adopted as the default TCP
>>> >   congestion control algorithm by the Linux, Windows, and Apple stack=
s.
>>> >
>>> >   This document updates the specification of CUBIC to include
>>> >   algorithmic improvements based on these implementations and recent
>>> >   academic work.  Based on the extensive deployment experience with
>>> >   CUBIC, it also moves the specification to the Standards Track,
>>> >   obsoleting RFC 8312.  This also requires updating RFC 5681, to allo=
w
>>> >   for CUBIC's occasionally more aggressive sending behavior.
>>> >
>>> >
>>> > The IETF datatracker status page for this draft is:
>>> > https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/
>>> >
>>> > There is also an HTML version available at:
>>> > https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html
>>> >
>>> > A diff from the previous version is available at:
>>> > https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06
>>> >
>>> >
>>> > Internet-Drafts are also available by rsync at rsync.ietf.org
>>> ::internet-drafts
>>> >
>>> >
>>> > _______________________________________________
>>> > tcpm mailing list
>>> > tcpm@ietf.org
>>> > https://www.ietf.org/mailman/listinfo/tcpm
>>>
>>> _______________________________________________
>>> tcpm mailing list
>>> tcpm@ietf.org
>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>
>> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm
>
>
>

--00000000000086feb805d792ea1b
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi=C2=A0Vidhi,</div><div><br></div><div>Thanks for th=
e comments.</div><br><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D"g=
mail_attr">On Mon, Feb 7, 2022 at 3:32 PM Vidhi Goel &lt;<a href=3D"mailto:=
vidhi_goel@apple.com">vidhi_goel@apple.com</a>&gt; wrote:<br></div><blockqu=
ote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px=
 solid rgb(204,204,204);padding-left:1ex"><div style=3D"overflow-wrap: brea=
k-word;">Thanks Yoshi for reviewing the draft.<div><br></div><div><blockquo=
te type=3D"cite"><div dir=3D"ltr"><div>So, it seems one PS doc mentions not=
 to use cwnd, while another PS doc mentions using cwnd is fine (with RFC766=
1).</div></div></blockquote>RFC5681 is conservative in its approach to use =
FlightSize and will result in very low throughput in scenarios when the con=
gestion event occurs at the tail of send window (when flight size could be =
very small)</div></div></blockquote><div><br></div><div>Yes. I agree.</div>=
<div>=C2=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px =
0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div sty=
le=3D"overflow-wrap: break-word;"><div><br></div><div><blockquote type=3D"c=
ite"><div dir=3D"ltr"><div><div>I am not sure which can cause this situatio=
n yet, but in any case, does this mean there is no concern on=C2=A0</div><d=
iv>&quot;some implementations may incidentally increase well beyond rwnd&qu=
ot; in cubic?</div></div><div>Is this because this is an old implementation=
 bug or are there other reasons?<br></div><div>I am wondering if we might w=
ant some clarification on this point.</div></div></blockquote><div>If an im=
plementation uses RFC7661, there is no concern about increasing cwnd beyond=
 rwnd as RFC7661 tracks bytes acknowledged in a RTT which means that `rwnd`=
 check has already been performed.</div><div>Usually cwnd doesn=E2=80=99t g=
row beyond rwnd =C2=A0(ACK-clocking increases cwnd which means the sender a=
lready checked <span style=3D"white-space:pre-wrap">	</span>`rwnd` before s=
ending out the data).</div><div><br></div><div>But there could be implement=
ations that may want to keep `rwnd` even smaller than initial `cwnd`. In th=
ose cases, RFC7661 would make the cwnd invalidated when the number of ackno=
wledged bytes is much lower than the current cwnd.</div><div><br></div><div=
>Do you have any suggestions on how to make it more clear?</div></div></div=
></blockquote><div><br></div><div>I think it will be fine as long as the dr=
aft clarifies we can safely override what&#39;s written in RFC5681 when we =
have some mechanisms such as RFC7661.</div><blockquote class=3D"gmail_quote=
" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);=
padding-left:1ex"><div style=3D"overflow-wrap: break-word;"><div><blockquot=
e type=3D"cite"><div dir=3D"ltr"><div>Also, I am wondering if we might want=
 to use &quot;MAY be used&quot; here instead of &quot;can be used&quot;=C2=
=A0=C2=A0</div></div></blockquote><div><div dir=3D"ltr"><div>Yeah, we can m=
ake this change.</div></div></div></div></div></blockquote><div><br></div><=
div>Thanks!</div><div>--=C2=A0</div><div>Yoshi</div><div>=C2=A0</div><block=
quote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1=
px solid rgb(204,204,204);padding-left:1ex"><div style=3D"overflow-wrap: br=
eak-word;"><div><div><div dir=3D"ltr"><div><br></div><div><br></div><div>Th=
anks,</div><div>Vidhi</div></div></div><div><br><blockquote type=3D"cite"><=
div>On Feb 6, 2022, at 3:01 AM, Yoshifumi Nishida &lt;<a href=3D"mailto:nsd=
.ietf@gmail.com" target=3D"_blank">nsd.ietf@gmail.com</a>&gt; wrote:</div><=
br><div><div dir=3D"ltr"><div>Hi,</div><div>I just have one more comment on=
 the draft.</div><div>In RFC5681, it mentions in Section 3:</div><div>&#39;=
&#39;</div><div><pre style=3D"font-size:13.3333px;margin-top:0px;margin-bot=
tom:0px;break-before:page">   Implementation Note: An easy mistake to make =
is to simply use cwnd,
   rather than FlightSize, which in some implementations may
   incidentally increase well beyond rwnd.</pre></div><div>&quot;</div><div=
><br></div><div>while this draft mentions in Section 4.6:</div><div>&quot;<=
/div><div><pre style=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0p=
x;break-before:page">   To avoid suboptimal performance with such applicati=
ons,</pre><pre style=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0p=
x;break-before:page">   the mechanisms described in [<a href=3D"https://dat=
atracker.ietf.org/doc/html/rfc7661" title=3D"&quot;Updating TCP to Support =
Rate-Limited Traffic&quot;" target=3D"_blank">RFC7661</a>] can be used to m=
itigate this
   issue as it would allow using a value between _cwnd_ and
   _flight_size_ to calculate the new _ssthresh_ in Figure 5. </pre><pre st=
yle=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0px;break-before:pa=
ge">&quot;</pre></div><div>So, it seems one PS doc mentions not to use cwnd=
, while another PS doc mentions using cwnd is fine (with RFC7661).</div><di=
v><br></div><div><div>I am not sure which can cause this situation yet, but=
 in any case, does this mean there is no concern on=C2=A0</div><div>&quot;s=
ome implementations may incidentally increase well beyond rwnd&quot; in cub=
ic?</div></div><div>Is this because this is an old implementation bug or ar=
e there other reasons?<br></div><div>I am wondering if we might want some c=
larification on this point.<br></div><div>=C2=A0</div><div>Also, I am wonde=
ring if we might want to use &quot;MAY be used&quot; here instead of &quot;=
can be used&quot;=C2=A0=C2=A0</div><div><br></div><div>Thanks,</div><div>--=
</div><div>Yoshi</div><div><br></div><div class=3D"gmail_quote"><div dir=3D=
"ltr" class=3D"gmail_attr">On Mon, Jan 31, 2022 at 11:43 PM Yoshifumi Nishi=
da &lt;<a href=3D"mailto:nsd.ietf@gmail.com" target=3D"_blank">nsd.ietf@gma=
il.com</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"m=
argin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left=
:1ex"><div dir=3D"ltr"><div>Hello,</div><div><br></div><div>After some disc=
ussions among chairs, we decided to run the 2nd WGLC on draft-ietf-tcpm-rfc=
8312bis in consideration of the importance of the draft.=C2=A0</div><div>We=
&#39;ll be grateful if you could send your feedback to the ML. The WGLC run=
s until *Feb 11*.</div><div><br></div><div>If interested, you can check in-=
depth past discussions in the following URL.</div><div><a href=3D"https://g=
ithub.com/NTAP/rfc8312bis/" target=3D"_blank">https://github.com/NTAP/rfc83=
12bis/</a><br></div><div><br></div><div>Thank you so much!</div><div>--</di=
v><div>tcpm co-chairs</div><div><br></div><br><div class=3D"gmail_quote"><d=
iv dir=3D"ltr" class=3D"gmail_attr">On Wed, Jan 26, 2022 at 2:50 AM Lars Eg=
gert &lt;<a href=3D"mailto:lars@eggert.org" target=3D"_blank">lars@eggert.o=
rg</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"margi=
n:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex=
">Hi,<br>
<br>
this -06 version rolls in all the changes requested during (and after) WGLC=
 ended.<br>
<br>
I&#39;ll leave it up to the chairs to decide if another WGLC is warranted o=
r the document can progress as-is.<br>
<br>
Thanks,<br>
Lars<br>
<br>
<br>
&gt; On 2022-1-26, at 11:12, <a href=3D"mailto:internet-drafts@ietf.org" ta=
rget=3D"_blank">internet-drafts@ietf.org</a> wrote:<br>
&gt; <br>
&gt; <br>
&gt; A New Internet-Draft is available from the on-line Internet-Drafts dir=
ectories.<br>
&gt; This draft is a work item of the TCP Maintenance and Minor Extensions =
WG of the IETF.<br>
&gt; <br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Title=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0: CUBIC for Fast and Long-Distance Networks<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Authors=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: =
Lisong Xu<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Sangtae Ha<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Injong Rhee<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Vidhi Goel<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Lars Eggert<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Filename=C2=A0 =C2=A0 =C2=A0 =C2=A0 : draft-=
ietf-tcpm-rfc8312bis-06.txt<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Pages=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0: 35<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Date=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 : 2022-01-26<br>
&gt; <br>
&gt; Abstract:<br>
&gt;=C2=A0 =C2=A0CUBIC is a standard TCP congestion control algorithm that =
uses a<br>
&gt;=C2=A0 =C2=A0cubic function instead of a linear congestion window incre=
ase<br>
&gt;=C2=A0 =C2=A0function to improve scalability and stability over fast an=
d long-<br>
&gt;=C2=A0 =C2=A0distance networks.=C2=A0 CUBIC has been adopted as the def=
ault TCP<br>
&gt;=C2=A0 =C2=A0congestion control algorithm by the Linux, Windows, and Ap=
ple stacks.<br>
&gt; <br>
&gt;=C2=A0 =C2=A0This document updates the specification of CUBIC to includ=
e<br>
&gt;=C2=A0 =C2=A0algorithmic improvements based on these implementations an=
d recent<br>
&gt;=C2=A0 =C2=A0academic work.=C2=A0 Based on the extensive deployment exp=
erience with<br>
&gt;=C2=A0 =C2=A0CUBIC, it also moves the specification to the Standards Tr=
ack,<br>
&gt;=C2=A0 =C2=A0obsoleting RFC 8312.=C2=A0 This also requires updating RFC=
 5681, to allow<br>
&gt;=C2=A0 =C2=A0for CUBIC&#39;s occasionally more aggressive sending behav=
ior.<br>
&gt; <br>
&gt; <br>
&gt; The IETF datatracker status page for this draft is:<br>
&gt; <a href=3D"https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis=
/" rel=3D"noreferrer" target=3D"_blank">https://datatracker.ietf.org/doc/dr=
aft-ietf-tcpm-rfc8312bis/</a><br>
&gt; <br>
&gt; There is also an HTML version available at:<br>
&gt; <a href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-=
06.html" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/archive/=
id/draft-ietf-tcpm-rfc8312bis-06.html</a><br>
&gt; <br>
&gt; A diff from the previous version is available at:<br>
&gt; <a href=3D"https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312=
bis-06" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/rfcdiff?u=
rl2=3Ddraft-ietf-tcpm-rfc8312bis-06</a><br>
&gt; <br>
&gt; <br>
&gt; Internet-Drafts are also available by rsync at <a href=3D"http://rsync=
.ietf.org" target=3D"_blank">rsync.ietf.org</a>::internet-drafts<br>
&gt; <br>
&gt; <br>
&gt; _______________________________________________<br>
&gt; tcpm mailing list<br>
&gt; <a href=3D"mailto:tcpm@ietf.org" target=3D"_blank">tcpm@ietf.org</a><b=
r>
&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/tcpm" rel=3D"noreferr=
er" target=3D"_blank">https://www.ietf.org/mailman/listinfo/tcpm</a><br>
<br>
_______________________________________________<br>
tcpm mailing list<br>
<a href=3D"mailto:tcpm@ietf.org" target=3D"_blank">tcpm@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/tcpm" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/listinfo/tcpm</a><br>
</blockquote></div></div>
</blockquote></div></div>
_______________________________________________<br>tcpm mailing list<br><a =
href=3D"mailto:tcpm@ietf.org" target=3D"_blank">tcpm@ietf.org</a><br><a hre=
f=3D"https://www.ietf.org/mailman/listinfo/tcpm" target=3D"_blank">https://=
www.ietf.org/mailman/listinfo/tcpm</a><br></div></blockquote></div><br></di=
v></div></blockquote></div></div>

--00000000000086feb805d792ea1b--


From nobody Wed Feb  9 02:42:41 2022
Return-Path: <Michael.Scharf@hs-esslingen.de>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9646B3A0DB1 for <tcpm@ietfa.amsl.com>; Wed,  9 Feb 2022 02:42:38 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.095
X-Spam-Level: 
X-Spam-Status: No, score=-2.095 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_NONE=0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=hs-esslingen.de
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id sEh39R5dMbrO for <tcpm@ietfa.amsl.com>; Wed,  9 Feb 2022 02:42:33 -0800 (PST)
Received: from mail.hs-esslingen.de (mail.hs-esslingen.de [134.108.32.78]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id F10003A0D7D for <tcpm@ietf.org>; Wed,  9 Feb 2022 02:42:32 -0800 (PST)
Received: from localhost (localhost.localdomain [127.0.0.1]) by mail.hs-esslingen.de (Postfix) with ESMTP id 93CB425A17; Wed,  9 Feb 2022 11:42:29 +0100 (CET)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=hs-esslingen.de; s=mail; t=1644403349; bh=FUvFQPrHpwNRGeEqwCb3bTPfqj2GtHY3cSQV/q2V4pA=; h=From:To:CC:Subject:Date:References:In-Reply-To:From; b=hZBVtfiULPCq60VmAhIoZoQ0AnvZkMYL2B9JhnuvH2OUIJc3SVaX7GDw90is4LUlu jBVGTHWWfJRqu9D5QBwvHet7pISJ5uYy0P2BvhMWexQR2kl4sU9d1rVfxJ1BtNDYiK +EG16fzsfXAZaja5QMg4RNTqa6eeZAsNNSsyWy+s=
X-Virus-Scanned: by amavisd-new-2.7.1 (20120429) (Debian) at hs-esslingen.de
Received: from mail.hs-esslingen.de ([127.0.0.1]) by localhost (hs-esslingen.de [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Vy7g9UBnccs5; Wed,  9 Feb 2022 11:42:26 +0100 (CET)
Received: from rznt8201.rznt.rzdir.fht-esslingen.de (rznt8201.hs-esslingen.de [134.108.48.164]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.hs-esslingen.de (Postfix) with ESMTPS; Wed,  9 Feb 2022 11:42:26 +0100 (CET)
Received: from rznt8202.rznt.rzdir.fht-esslingen.de (134.108.48.165) by rznt8201.rznt.rzdir.fht-esslingen.de (134.108.48.164) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.18; Wed, 9 Feb 2022 11:42:26 +0100
Received: from rznt8202.rznt.rzdir.fht-esslingen.de ([fe80::aca4:171a:3ee1:57e0]) by rznt8202.rznt.rzdir.fht-esslingen.de ([fe80::aca4:171a:3ee1:57e0%3]) with mapi id 15.01.2375.018; Wed, 9 Feb 2022 11:42:26 +0100
From: "Scharf, Michael" <Michael.Scharf@hs-esslingen.de>
To: "touch@strayalpha.com" <touch@strayalpha.com>, "mohamed.boucadair@orange.com" <mohamed.boucadair@orange.com>
CC: "tcpm@ietf.org" <tcpm@ietf.org>
Thread-Topic: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
Thread-Index: AQHXyc25qfTSowdLSEa6dKhesfqRYKv/uF0AgEvBUuCABenxUIAYa72QgBqz19CABXaAYIAAX6CAgAFVLrA=
Date: Wed, 9 Feb 2022 10:42:26 +0000
Message-ID: <c46c926c48254155a1c0e46de5eecfb4@hs-esslingen.de>
References: <163518613944.7645.2494692847367626249@ietfa.amsl.com> <18622_1636712195_618E3F03_18622_35_1_787AE7BB302AE849A7480A190F8B933035450859@OPEXCAUBMA2.corporate.adroot.infra.ftgroup> <bde902389b8a4c76aae776ea2d68d9ce@hs-esslingen.de> <3337_1641201204_61D2BE34_3337_132_1_787AE7BB302AE849A7480A190F8B93303546BDC5@OPEXCAUBMA2.corporate.adroot.infra.ftgroup> <706848a62fd74d18a34566046369d049@hs-esslingen.de> <bdc3831619dd442dbc5e2914fff7a2b5@hs-esslingen.de> <18668_1644309564_62022C3C_18668_151_1_787AE7BB302AE849A7480A190F8B93303548DD57@OPEXCAUBMA2.corporate.adroot.infra.ftgroup> <745AF8A3-5F02-4752-BA28-43C6F10FFCAB@strayalpha.com>
In-Reply-To: <745AF8A3-5F02-4752-BA28-43C6F10FFCAB@strayalpha.com>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [134.108.140.249]
Content-Type: multipart/alternative; boundary="_000_c46c926c48254155a1c0e46de5eecfb4hsesslingende_"
MIME-Version: 1.0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/gCMXDY3iWKyWK6d3dbqgVxwfe6M>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-yang-tcp-04.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Feb 2022 10:42:39 -0000

--_000_c46c926c48254155a1c0e46de5eecfb4hsesslingende_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_c46c926c48254155a1c0e46de5eecfb4hsesslingende_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_c46c926c48254155a1c0e46de5eecfb4hsesslingende_--


From nobody Wed Feb  9 08:30:42 2022
Return-Path: <nsd.ietf@gmail.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5D7EF3A059F for <tcpm@ietfa.amsl.com>; Wed,  9 Feb 2022 08:30:40 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.098
X-Spam-Level: 
X-Spam-Status: No, score=-2.098 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id izgW8yh2qyI1 for <tcpm@ietfa.amsl.com>; Wed,  9 Feb 2022 08:30:36 -0800 (PST)
Received: from mail-qt1-x832.google.com (mail-qt1-x832.google.com [IPv6:2607:f8b0:4864:20::832]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E36613A040B for <tcpm@ietf.org>; Wed,  9 Feb 2022 08:30:35 -0800 (PST)
Received: by mail-qt1-x832.google.com with SMTP id t1so2215541qtq.13 for <tcpm@ietf.org>; Wed, 09 Feb 2022 08:30:35 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112;  h=mime-version:from:date:message-id:subject:to; bh=MKsx35PUh45JEKcnaLZGgtbNoLgby7krimIrqrWDzxQ=; b=MTrMYeuhIvrsp/jMTWpCmTxV/lzaYtFcGpIaCaxmqT9P9hTwBZGokInPG7eYEqT0Rc CNhkDbZtEzD1Exqq05fZmuOBS6Qk1onnpzcTe7CxISI6xSJftz0K1OW4yDyM84KH6573 MUwWsN2nr8lFVsS56iLpMSr2HsKV3ewZNfL1/wDEcpwPZA5DhV0waiXU2XMIUvO/CWf4 X4dnHCZI/lMFpwepbh17DqxHDnz6QY+Xlqjye4BnmI/xH1SXOMYTzZpETc2xx2YEwnxP ziuPjHvWpEY5KwertyloAvHc9DdKVZgNC0UhphS7sTBcFWEpm20ciBXjQQn1mqxSQ2q4 o7Vw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:from:date:message-id:subject:to; bh=MKsx35PUh45JEKcnaLZGgtbNoLgby7krimIrqrWDzxQ=; b=Xo9S0MynXdJ2vv14lka3oZDZNZ0Rbr/SWi+owVa11339/xB2cyZ0GqwdUo3VxjOVtx lwdKOvDlFYbip1Li1YOBMs6dv/RVwv3B+gMU4CQFdkFzAJlcLAcK9ad/AvMqp3Ocd9ju nSXz8TgsQfMDPBtxwcG+wntVz6RdxIY/+s6CIHFFxJPpqxFGmj+ym4/KYB10k4KP9zx3 q1s4XZPUA93BQDWYdjtJmNO+BYMukOqISfCyZx6r6yw0dUwMwROmUq/DLq79Sfxf/iLm sKC7j+SJ2zaRmOWnWpoxDmtS6hMZoBtkfsWqZsoh/1Iz7/vcmIHqDP8wg6GfaMRl1+eM u3XQ==
X-Gm-Message-State: AOAM5319pjPD+xzd3eM7MOGsxi6yWiJU2oYCYaRvQ97Hsg99vDqxgbdf lM/7WHmixa0PRKcR9haNr7WW+R1lUP8JnwEA1m9GEOzCYf4=
X-Google-Smtp-Source: ABdhPJzHRUTrFtym6CkJz4Fc9p0dNGRDJf7obmYsb5ntcz3e4by19A/k1OfrOJvfDBVQ1s7MvKzzZ8cRf9GFQquDBOU=
X-Received: by 2002:a05:622a:174d:: with SMTP id l13mr1930428qtk.55.1644424233482;  Wed, 09 Feb 2022 08:30:33 -0800 (PST)
MIME-Version: 1.0
From: Yoshifumi Nishida <nsd.ietf@gmail.com>
Date: Wed, 9 Feb 2022 08:30:22 -0800
Message-ID: <CAAK044TS==yvSf+ve22XVQGU2s2os2P8cqckLykmMb9XEa9MTw@mail.gmail.com>
To: "tcpm@ietf.org Extensions" <tcpm@ietf.org>
Content-Type: multipart/alternative; boundary="0000000000004bfb7b05d7985a1c"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/i57MWRpzwFVGPSyHb4jRf1Aeab4>
Subject: [tcpm] comments on draft-ietf-tcpm-prr-rfc6937bis-01
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Feb 2022 16:30:41 -0000

--0000000000004bfb7b05d7985a1c
Content-Type: text/plain; charset="UTF-8"

Hi,

I have the following comments on draft-ietf-tcpm-prr-rfc6937bis-01

1: Page 5
    "but recommended to use BBR-SSRB"
      -> PRR-SSRB?

2: Page 7
    "and do what? @@@@"
      -> Please add more texts or remove this.

3: Page 8:
   """
      prr_delivered = 0         // Total bytes delivered during recovery

      prr_out = 0               // Total bytes sent during recovery

   """
     -> I personally think "during recovery" might be a bit ambiguous. I
think it would be better to clarify
        whether this includes a packet sent by fast retransmit or not.

4: Page 8:
  " pipe = (RFC 6675 <https://datatracker.ietf.org/doc/html/rfc6675> pipe
algorithm) "

     -> The algorithm here seems to depend on SACK, but the draft also
states:
        "It is most accurate and more easily implemented with SACK [RFC2018
<https://datatracker.ietf.org/doc/html/rfc2018>], but does not require
SACK."
         I think It is not clear how this algorithm works without SACK.

5: Section 7 Page 9
     Why all examples shown here use Limited Transmit even though PRR
doesn't require it?
     I think there should be some explanations for it even if there's no
strong reason.


6: Page 10:
     It seems that the pipe size in the figure is different from what
RFC6675 calculates because Section 5 (4.2) in RFC6675 mentions
    "

      note that [RFC5681
<https://datatracker.ietf.org/doc/html/rfc5681>] requires that any
      segments sent as part of the Limited Transmit mechanism not
      be counted in FlightSize for the purpose of the above
      equation.

        Hence, I think the pipe size would be something like this if
it follows RFC6675.

    Please let me know if I miss something.


6675
   ack#   X  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19
   cwnd:    20 20 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10
   pipe:    19 19 19 18 17 16 15 14 13 12 11 10 09 09 09 09 09 09 09
   sent:     N  N  R                             N  N  N  N  N  N  N


PRR

   ack#   X  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19
   pipe:    19 19 18 18 18 17 17 16 16 15 15 14 14 13 13 12 12 11 10
   sent:     N  N  R     N     N     N     N     N     N     N



6675

   ack#   X  X  X  X  X  X  X  X  X  X  X  X  X  X  X 15 16 17 18 19
   cwnd:                                              20 20 10 10 10
   pipe:                                              19 19  4 10 10
   sent:                                               N  N 6R  R  R


Thanks,
--
Yoshi

--0000000000004bfb7b05d7985a1c
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hi,<div><br></div><div>I have the following comments on dr=
aft-ietf-tcpm-prr-rfc6937bis-01</div><div><br></div><div>1: Page 5</div><di=
v>=C2=A0 =C2=A0 &quot;but recommended to use BBR-SSRB&quot;</div><div>=C2=
=A0 =C2=A0 =C2=A0 -&gt; PRR-SSRB?</div><div><br></div><div>2: Page 7</div><=
div>=C2=A0 =C2=A0 &quot;and do what?=C2=A0@@@@&quot;</div><div>=C2=A0 =C2=
=A0 =C2=A0 -&gt; Please add more texts or remove this.</div><div><br></div>=
<div>3: Page 8:</div><div>=C2=A0 =C2=A0&quot;&quot;&quot;</div><div><font f=
ace=3D"monospace"><span style=3D"color:rgb(0,0,0);font-size:13.3333px">=C2=
=A0 =C2=A0 =C2=A0 prr_delivered =3D 0=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0// T=
otal bytes delivered during recovery</span><br></font></div><div><pre style=
=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0px;break-before:page;=
color:rgb(0,0,0)">      prr_out =3D 0               // Total bytes sent dur=
ing recovery</pre></div><div>=C2=A0 =C2=A0&quot;&quot;&quot;</div><div>=C2=
=A0 =C2=A0 =C2=A0-&gt; I personally think &quot;during recovery&quot; might=
 be a bit ambiguous. I think it would be better to clarify=C2=A0</div><div>=
=C2=A0 =C2=A0 =C2=A0 =C2=A0 whether this includes a packet sent by fast ret=
ransmit or not.=C2=A0</div><div><br></div><div>4: Page 8:</div><div>=C2=A0 =
&quot;<span style=3D"color:rgb(0,0,0);font-size:13.3333px">      pipe =3D (=
</span><a href=3D"https://datatracker.ietf.org/doc/html/rfc6675" style=3D"f=
ont-size:13.3333px" target=3D"_blank">RFC 6675</a><span style=3D"color:rgb(=
0,0,0);font-size:13.3333px"> pipe algorithm) &quot;</span></div><div><span =
style=3D"color:rgb(0,0,0);font-size:13.3333px">=C2=A0=C2=A0</span></div><di=
v><span style=3D"color:rgb(0,0,0);font-size:13.3333px">=C2=A0 =C2=A0 =C2=A0=
-&gt; The algorithm here seems to depend on SACK, but the=C2=A0draft also s=
tates:</span></div><div><span style=3D"color:rgb(0,0,0);font-size:13.3333px=
">=C2=A0 =C2=A0 =C2=A0 =C2=A0 &quot;</span><span style=3D"color:rgb(0,0,0);=
font-size:13.3333px">I<font face=3D"arial, sans-serif">t is most=C2=A0</fon=
t></span><span style=3D"font-family:arial,sans-serif;color:rgb(0,0,0);font-=
size:13.3333px">accurate and more easily implemented with SACK [</span><a h=
ref=3D"https://datatracker.ietf.org/doc/html/rfc2018" title=3D"&quot;TCP Se=
lective Acknowledgment Options&quot;" style=3D"font-family:arial,sans-serif=
;font-size:13.3333px" target=3D"_blank">RFC2018</a><span style=3D"font-fami=
ly:arial,sans-serif;color:rgb(0,0,0);font-size:13.3333px">], but does=C2=A0=
</span><span style=3D"font-family:arial,sans-serif;color:rgb(0,0,0);font-si=
ze:13.3333px">not require SACK.&quot;</span></div>=C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0I think It is not clear how this algorithm works without SACK.=C2=
=A0<div><br></div><div>5: Section 7 Page 9</div><div>=C2=A0 =C2=A0 =C2=A0Wh=
y all examples shown here use Limited Transmit even though PRR doesn&#39;t =
require it?=C2=A0</div><div>=C2=A0 =C2=A0 =C2=A0I think there should be som=
e explanations for it even if there&#39;s no strong reason.</div><div><br><=
/div><div><br></div><div>6: Page 10:</div><div>=C2=A0 =C2=A0 =C2=A0It seems=
 that the pipe size in the figure is different from what RFC6675 calculates=
 because Section 5 (4.2) in RFC6675 mentions</div><div>=C2=A0 =C2=A0 &quot;=
</div><div><pre style=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0=
px;break-before:page;color:rgb(0,0,0)">      note that [<a href=3D"https://=
datatracker.ietf.org/doc/html/rfc5681" title=3D"&quot;TCP Congestion Contro=
l&quot;" target=3D"_blank">RFC5681</a>] requires that any
      segments sent as part of the Limited Transmit mechanism not
      be counted in FlightSize for the purpose of the above
      equation.</pre><pre style=3D"font-size:13.3333px;margin-top:0px;margi=
n-bottom:0px;break-before:page;color:rgb(0,0,0)"><font face=3D"arial, sans-=
serif">    </font></pre><pre style=3D"font-size:13.3333px;margin-top:0px;ma=
rgin-bottom:0px;break-before:page;color:rgb(0,0,0)"><font face=3D"arial, sa=
ns-serif">    Hence, I think the pipe size would be something like this if =
it follows RFC6675.</font></pre><pre style=3D"font-size:13.3333px;margin-to=
p:0px;margin-bottom:0px;break-before:page;color:rgb(0,0,0)"><font face=3D"a=
rial, sans-serif">    Please let me know if I miss something.</font></pre><=
pre style=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0px;break-bef=
ore:page;color:rgb(0,0,0)"><font face=3D"arial, sans-serif"><br></font></pr=
e><pre style=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0px;break-=
before:page;color:rgb(0,0,0)">6675<br>=C2=A0 =C2=A0ack# =C2=A0 X =C2=A01 =
=C2=A02 =C2=A03 =C2=A04 =C2=A05 =C2=A06 =C2=A07 =C2=A08 =C2=A09 10 11 12 13=
 14 15 16 17 18 19<br>=C2=A0 =C2=A0cwnd: =C2=A0 =C2=A020 20 10 10 10 10 10 =
10 10 10 10 10 10 10 10 10 10 10 10<br>=C2=A0 =C2=A0pipe: =C2=A0 =C2=A019 1=
9 19 18 17 16 15 14 13 12 11 10 09 09 09 09 09 09 09<br>=C2=A0 =C2=A0sent: =
=C2=A0 =C2=A0 N =C2=A0N =C2=A0R =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 N =C2=A0N =C2=A0N =
=C2=A0N =C2=A0N =C2=A0N =C2=A0N<br><br><br>PRR<br><br>=C2=A0 =C2=A0ack# =C2=
=A0 X =C2=A01 =C2=A02 =C2=A03 =C2=A04 =C2=A05 =C2=A06 =C2=A07 =C2=A08 =C2=
=A09 10 11 12 13 14 15 16 17 18 19<br>=C2=A0 =C2=A0pipe: =C2=A0 =C2=A019 19=
 18 18 18 17 17 16 16 15 15 14 14 13 13 12 12 11 10<br>=C2=A0 =C2=A0sent: =
=C2=A0 =C2=A0 N =C2=A0N =C2=A0R =C2=A0 =C2=A0 N =C2=A0 =C2=A0 N =C2=A0 =C2=
=A0 N =C2=A0 =C2=A0 N =C2=A0 =C2=A0 N =C2=A0 =C2=A0 N =C2=A0 =C2=A0 N<br></=
pre><pre style=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0px;brea=
k-before:page;color:rgb(0,0,0)"><br></pre><pre style=3D"font-size:13.3333px=
;margin-top:0px;margin-bottom:0px;break-before:page;color:rgb(0,0,0)"><br><=
/pre><pre style=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0px;bre=
ak-before:page;color:rgb(0,0,0)">6675</pre><pre style=3D"font-size:13.3333p=
x;margin-top:0px;margin-bottom:0px;break-before:page;color:rgb(0,0,0)">=C2=
=A0 =C2=A0ack# =C2=A0 X =C2=A0X =C2=A0X =C2=A0X =C2=A0X =C2=A0X =C2=A0X =C2=
=A0X =C2=A0X =C2=A0X =C2=A0X =C2=A0X =C2=A0X =C2=A0X =C2=A0X 15 16 17 18 19=
<br>=C2=A0 =C2=A0cwnd: =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A020 20 10 10 10<br>=C2=A0 =C2=A0pipe: =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A019 19 =C2=A04 10 10<br>=C2=A0 =C2=A0sent: =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 N =C2=A0=
N 6R =C2=A0R =C2=A0R<br><br></pre></div><div>=C2=A0<br></div><div>Thanks,</=
div><div>--</div><div>Yoshi</div><div><br><div><br></div><div><br></div><di=
v><br></div></div></div>

--0000000000004bfb7b05d7985a1c--


From nobody Wed Feb  9 14:19:30 2022
Return-Path: <csmiley@amsl.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8CF423A0122 for <tcpm@ietfa.amsl.com>; Wed,  9 Feb 2022 14:19:27 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.898
X-Spam-Level: 
X-Spam-Status: No, score=-1.898 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6O9GyJorOCKS for <tcpm@ietfa.amsl.com>; Wed,  9 Feb 2022 14:19:23 -0800 (PST)
Received: from c8a.amsl.com (c8a.amsl.com [4.31.198.40]) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 336F43A0A10 for <tcpm@ietf.org>; Wed,  9 Feb 2022 14:19:23 -0800 (PST)
Received: from localhost (localhost [127.0.0.1]) by c8a.amsl.com (Postfix) with ESMTP id 14097425A344; Wed,  9 Feb 2022 14:19:23 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
Received: from c8a.amsl.com ([127.0.0.1]) by localhost (c8a.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Sp2tQkMbcWzQ; Wed,  9 Feb 2022 14:19:23 -0800 (PST)
Received: from [192.168.1.16] (cpe-76-95-228-63.socal.res.rr.com [76.95.228.63]) by c8a.amsl.com (Postfix) with ESMTPSA id B566D424B435; Wed,  9 Feb 2022 14:19:22 -0800 (PST)
Content-Type: text/plain; charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 12.4 \(3445.104.11\))
From: Chris Smiley <csmiley@amsl.com>
In-Reply-To: <20211226085938.97471F0F1F@rfc-editor.org>
Date: Wed, 9 Feb 2022 14:19:22 -0800
Cc: yaakovjstein@gmail.com, david.borman@quantum.com, braden@isi.edu, vanj@google.com, rs@netapp.com, tcpm@ietf.org, RFC Errata System <rfc-editor@rfc-editor.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <CC30262C-B376-44B8-9993-E9D492187103@amsl.com>
References: <20211226085938.97471F0F1F@rfc-editor.org>
To: martin.h.duke@gmail.com, Zaheduzzaman Sarker <Zaheduzzaman.Sarker@ericsson.com>
X-Mailer: Apple Mail (2.3445.104.11)
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/_PqiSr1RcdSmSs7XV7t4m4qEJRA>
Subject: Re: [tcpm] [Editorial Errata Reported] RFC7323 (6798)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Feb 2022 22:19:28 -0000

Greetings,

We are unable to verify this erratum that the submitter marked as =
editorial. =20
Please note that we have changed the =E2=80=9CType=E2=80=9D of the =
following errata=20
report to =E2=80=9CTechnical=E2=80=9D.  As Stream Approver, please =
review and set the=20
Status and Type accordingly (see the definitions at=20
https://www.rfc-editor.org/errata-definitions/).

You may review the report at:=20
https://www.rfc-editor.org/errata/eid6798

Please see https://www.rfc-editor.org/how-to-verify/ for further=20
information on how to verify errata reports.

Further information on errata can be found at:=20
https://www.rfc-editor.org/errata.php.

Thank you.

RFC Editor/cs

> On Dec 26, 2021, at 12:59 AM, RFC Errata System =
<rfc-editor@rfc-editor.org> wrote:
>=20
> The following errata report has been submitted for RFC7323,
> "TCP Extensions for High Performance".
>=20
> --------------------------------------
> You may review the report below and at:
> https://www.rfc-editor.org/errata/eid6798
>=20
> --------------------------------------
> Type: Editorial
> Reported by: Yaakov Stein <yaakovjstein@gmail.com>
>=20
> Section: GLOBAL
>=20
> Original Text
> -------------
> The Timestamps option may appear in any data or <ACK> segment, adding
>=20
> The timestamp clock may be derived from a system clock that is
>=20
> A random offset may be added to the timestamp clock on a per-
>=20
> Corrected Text
> --------------
> The Timestamps option MAY appear in any data or <ACK> segment, adding
>=20
> The timestamp clock MAY be derived from a system clock that is
>=20
> A random offset MAY be added to the timestamp clock on a per-
>=20
> Notes
> -----
> several "MAY"s were incorrectly written as non RFC 2119 "may"s
>=20
> Instructions:
> -------------
> This erratum is currently posted as "Reported". If necessary, please
> use "Reply All" to discuss whether it should be verified or
> rejected. When a decision is reached, the verifying party =20
> can log in to change the status and edit the report, if necessary.=20
>=20
> --------------------------------------
> RFC7323 (draft-ietf-tcpm-1323bis-21)
> --------------------------------------
> Title               : TCP Extensions for High Performance
> Publication Date    : September 2014
> Author(s)           : D. Borman, B. Braden, V. Jacobson, R. =
Scheffenegger, Ed.
> Category            : PROPOSED STANDARD
> Source              : TCP Maintenance and Minor Extensions
> Area                : Transport
> Stream              : IETF
> Verifying Party     : IESG
>=20


From nobody Thu Feb 10 08:53:54 2022
Return-Path: <ietf@bobbriscoe.net>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EA4FD3A0E0F for <tcpm@ietfa.amsl.com>; Thu, 10 Feb 2022 08:53:51 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -7.813
X-Spam-Level: 
X-Spam-Status: No, score=-7.813 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, NICE_REPLY_A=-0.714, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=bobbriscoe.net
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jBZZIQl3cySC for <tcpm@ietfa.amsl.com>; Thu, 10 Feb 2022 08:53:46 -0800 (PST)
Received: from mail-ssdrsserver2.hostinginterface.eu (mail-ssdrsserver2.hostinginterface.eu [185.185.85.90]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 64E7B3A07F1 for <tcpm@ietf.org>; Thu, 10 Feb 2022 08:53:46 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=bobbriscoe.net; s=default; h=In-Reply-To:From:References:Cc:To:Subject: MIME-Version:Date:Message-ID:Content-Type:Sender:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id: List-Help:List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=zJcBpUiZse7NfH8BiVDvc3qlWF+m2XIxsMyQaXBauHE=; b=LF7QTE85hXvAmXo6AaranqqB5X CLwSBqkaNhkwZ4YwW/bn7RE69+g26qBEqi4BxHQvMwq1kQtMuFDq9NxsSyQv6EUqNO1Zt0fGgLjur NoYkje7FWqxlhj/8+SwRCnhCxmQs7q4e4nb/5zcbJIbJpOybdhmSD/2er5s5BjDSWW+z7H1CBGAre M4u5OFCBMHJg1fzCB4ovmUwy0/5jOsWMZZ1Wn1TGCMg4qqEydDH7/Jc+9D4L1dta5OO50CWfpLxbE GBDN+i24bqhvoRTMzh1Qjna0a0YKOhX9tZfzYA8NpI13lnflrrtdJqXuyyX0AuXstGE4B5f7+CvVC Zbk/YLIw==;
Received: from 67.153.238.178.in-addr.arpa ([178.238.153.67]:55354 helo=[192.168.1.11]) by ssdrsserver2.hostinginterface.eu with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 (Exim 4.94.2) (envelope-from <ietf@bobbriscoe.net>) id 1nIChV-00075s-Ij; Thu, 10 Feb 2022 16:53:44 +0000
Content-Type: multipart/alternative; boundary="------------V8LckoMr5emQ54ffsERkSpct"
Message-ID: <00d7ff67-968c-1593-2bb3-ad485756fd80@bobbriscoe.net>
Date: Thu, 10 Feb 2022 16:53:42 +0000
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.5.0
Content-Language: en-GB
To: "Scheffenegger, Richard" <rs.ietf@gmx.at>, Vidhi Goel <vidhi_goel=40apple.com@dmarc.ietf.org>
Cc: Richard Scheffenegger <rscheff@gmx.at>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, Mirja Kuehlewind <ietf@kuehlewind.net>
References: <164389942954.13556.7754569279919863072@ietfa.amsl.com> <c2c5bdf9-6b34-fe19-3296-5f1cd831ed8e@bobbriscoe.net> <2FB9199D-4D31-418B-81DE-2E2D4358DEF6@apple.com> <e57355ae-5a9d-0a14-9478-29453d25622c@bobbriscoe.net> <BB23E172-C995-41FE-A47E-E02D5586B67D@apple.com> <1cbe3dbe-241b-3f24-096f-f1a2ffe17d98@gmx.at> <615549a4-6d09-8a7b-f892-42a1ea52c2e2@bobbriscoe.net> <24dd6cbd-6668-23e2-6902-9772d0a1ef9a@gmx.at>
From: Bob Briscoe <ietf@bobbriscoe.net>
In-Reply-To: <24dd6cbd-6668-23e2-6902-9772d0a1ef9a@gmx.at>
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - ssdrsserver2.hostinginterface.eu
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - bobbriscoe.net
X-Get-Message-Sender-Via: ssdrsserver2.hostinginterface.eu: authenticated_id: in@bobbriscoe.net
X-Authenticated-Sender: ssdrsserver2.hostinginterface.eu: in@bobbriscoe.net
X-Source: 
X-Source-Args: 
X-Source-Dir: 
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/wy2jeUXSszGfu3mQrnSDZvfTdIg>
Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-accurate-ecn-16.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Feb 2022 16:53:52 -0000

This is a multi-part message in MIME format.
--------------V8LckoMr5emQ54ffsERkSpct
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Richard,see [BB] inline.

On 06/02/2022 23:00, Scheffenegger, Richard wrote:
> Hi Bob,
>
> could it be, that the first "data" row and column in this table are
> parially transposed?
>
> In the case of sending a Non-ECT SYN, but getting an indication of a
> received CE on the SYN-ACK, why would you respond to that indication of
> an incorrect CE marking? (With the response being, that the cwnd would
> probably never grow beyond 1 or 2 packets per RTT, or possibly diminish
> to n RTTs per packet, one a stack supports such low rates).
>
> I am thinking of the most common probable cause - a middlebox
> unconditionally overriding the full TOS byte, and simply setting both
> rightmost bits continously.
>
> Such a path would be rendered utterly unusable for an AccECN session, 
> not?
>
> So at least that case would be a ND in your nomenclature.

[BB] The full story for the first row (Not-ECT to ECT/CE transitions) 
was pointed out by Vidhi, and discussed in my previous response to her:

  * Not-ECT to CE might have been two transitions
      o Not-ECT to ECT at a broken ToS remarker (prob first hop);
      o ECT to CE at a subsequent bottleneck;
  * We can only assume that a ToS remarking transition will be
    consistent. The congestion part could be transient.
      o So /initially/ all cases where Not-ECT changes to anything else
        need to be treated as a change to ECT;
      o But then there's /another/ test, which disables response to CE
        if there's solid CE marking;
      o I.e. NR, that potentially becomes ND later.


>
> In the column - if the IP ECN is bleached, why disable response to CE
> (if that is unlike to come back, as long as the session traverses the
> bleaching path)? A "NR" seems more logical here in case of path changes?

[BB] The thinking was that, if an ECT SYN becomes Not-ECT, then 
subsequently the sender only sends Not-ECT data, if that becomes CE, it 
seems pretty likely that is not a real congestion marking, whether 
there's a path change or not (because Not-ECT should remain Not-ECT).

>
> But again, all these ins-and-outs should be discussed in a different 
> draft.

[BB] Yes, deployment experience is needed before anything definitive can 
be said about fallback, because there will be cases where the 
recommendation depends on which of a number of possible odd behaviours 
is most likely.

>
> I fully agree, that only the normative response of the AccECN feedback
> mechanism has to be fully defined - the two bullet point is exactly what
> I expect to be stated.

[BB] Great. I'll write that up.


Bob

>
> Regards,
>
>    Richard
>
>
> Am 06.02.2022 um 23:27 schrieb Bob Briscoe:
>> Richard,
>>
>> Indeed, I recently discovered I had missed an item that I added to my
>> ToDo list in Mar 2021, based on a conversation between Gorry and Mirja
>> where they both made the same point - that fall-back behaviour after
>> discovering ECN mangling ought to be non-normative in this draft,
>> because it's beyond the scope of the new feedback mechanism:
>> https://mailarchive.ietf.org/arch/msg/tcpm/00URm1uwptgrAzKq96T6KSh8dQ4/
>>
>> We do need to say the following normatively in "Testing for Mangling of
>> the IP/ECN Field" because they are relevant to feedback, the Data 
>> Sender:
>> * remains in AccECN mode
>> * MUST continue to feed back any ECN markings on arriving packets.
>>
>> BTW, Vidhi, the text you quoted was from §2 on "Protocol Overview and
>> Rationale". That was meant to be an example of why it is useful for the
>> handshake to give feedback on exactly which IP-ECN codepoint arrived. It
>> wasn't meant to be a full spec. So I'll make it clearer that it's only
>> an example, rather than try to make it comprehensive, which isn't
>> appropriate at that point in the doc.
>>
>> Nonetheless, your points are correct, but if they are dealt with anyway,
>> they should be dealt with in §3.2.2.3. "Testing for Mangling of the
>> IP/ECN Field" where these scenarios are already covered (except for the
>> congestion response, which was silently left out of scope):
>> https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3 
>>
>>
>>
>> We could just say that the questions of whether to set ECT for
>> subsequent packets, and whether to respond to feedback of CE markings
>> are beyond the scope of the draft. I think we should, but I think we
>> should also give some non-normative advice on what to do. It can only be
>> interim advice, not normative requirements. Both because it's out of
>> scope, and because the best behaviour and the alternatives will depend
>> on deployment experience. So they shouldn't be written normatively in a
>> Proposed Standard yet.
>>
>> We will do what the WG wants, but assuming it agrees to including
>> non-normative advice, following from your (Richard's) list of cases,
>> here's a full matrix. I don't propose this matrix should go in the
>> AccECN draft - it's just for this email, to check every case is covered.
>>
>>
>>     Feedback in SYN/ACK of SYN's IP-ECN
>> IP-ECN on SYN     Not-ECT     ECT(0)     ECT(1)     CE
>> Not ECT     √     NR     NR     NR
>> ECT(0)     ND     √     √     √
>> ECT(1)     ND     CU
>>     √     √
>> CE     n/a     n/a     n/a     n/a
>>
>>
>> Key:
>> √       Valid response
>> NR    Non-normative advice: Set *N*ot-ECT for subsequent packets, but
>> continue to *R*espond to CE feedback
>> ND    Non-normative advice: Set *N*ot-ECT for subsequent packets, and
>> *D*isable response to CE feedback
>> CU    Currently unused transition (not listed as invalid in the draft)
>> n/a    not applicable in the AccECN spec, in which there is never a case
>> where the sender sets CE when sending a packet.
>>
>> In the draft we don't need to mention any cases where CE is sent (but a
>> draft on fall-back probably would, because the sender might set CE to
>> test the path). We can cover the other cases with two groups of invalid
>> transitions: those where the SYN (or SYN-ACK) arriving at the receiver
>> appears to have been
>> i) ECN-capable
>> ii) Not-ECN-capable.
>>
>> I propose to also say that details (like what 'subsequent packets'
>> implies about how long this behaviour continues) are for future study.
>>
>> I'll try to write some text and propose it to this list. Perhaps it
>> should be in a new appendix?
>>
>>
>> Bob
>>
>> On 05/02/2022 12:01, Scheffenegger, Richard wrote:
>>> Vidhi,
>>>
>>> Thanks for reading the draft very closely!
>>>
>>> Well, in the non-ECT SYN case, where the SYNACK indicates an ECT (ect0,
>>> ect1, ce) is the obvious - and fully semantically correct - use case of
>>> that possible use.
>>>
>>> For the set of three codepoints (ECT0, ECT1, CE) the only other 
>>> possible
>>> valid detection scenario would be if a SYN+CE is sent (deliberately,
>>> althought one can argue that a sender should not originate CE-marked
>>> packets), but a SYNACK without CE is returned (full or partial
>>> bleaching). At least under the current RFC3168 definition of IP ECN
>>> codepoints.
>>>
>>> allowed transitions:
>>> nonECT -> nonECT
>>>
>>> ECT0 -> ECT0
>>> ECT0 -> ECT1
>>> ECT0 -> CE
>>>
>>> ECT1 -> ECT1
>>> ECT1 -> ECT0 (possibly)
>>> ECT1 -> CE
>>>
>>> CE -> CE
>>>
>>>
>>> Thinking about this aspect more, I am now unsure, if the higher level
>>> functions, such as detection of IP ECN codepoint mangling, or 
>>> validating
>>> across multiple RTTs, should be in this document - or not rather be 
>>> in a
>>> separate draft, in order to facilitate independent implementation and
>>> improvement of mechanisms in that space.
>>>
>>> However, the above table of allowed IP ECN transistions may be updated
>>> independently of AccECN - and performing extensive checks as part of a
>>> signal protocol ossifies this IMHO.
>>>
>>> As a middle ground, perhaps having these descriptions in a 
>>> non-normative
>>> section, clearly stating these to be examples of what an impemented
>>> could choose to do?
>>>
>>> Richard
>>>
>>> Am 05.02.2022 um 02:14 schrieb Vidhi Goel:
>>>> I like your response for all the points. I can review the diff for the
>>>> proposed changes, if you’d like before committing them to the draft.
>>>>
>>>> Sorry, I want to add a new comment for this text,
>>>>
>>>> If a TCP client has set the SYN to Not-ECT, but
>>>>     receives feedback that the IP-ECN field on the SYN arrived with a
>>>>     different codepoint, it can detect such middlebox interference and
>>>>     send Not-ECT for the rest of the connection.
>>>>
>>>>
>>>> This statement holds good for Not-ECT but doesn’t if lets say the SYN
>>>> was ECT0 or ECT1 as those could be marked CE and still be 
>>>> valid. Should
>>>> we add a statement for ECT marked SYN as well?
>>>>
>>>> Thanks,
>>>> Vidhi
>>>>
>>>>> On Feb 4, 2022, at 5:44 AM, Bob Briscoe <ietf@bobbriscoe.net
>>>>> <mailto:ietf@bobbriscoe.net>> wrote:
>>>>>
>>>>> Vidhi,
>>>>>
>>>>> I've increased the table of contents to tocdepth = "4" which might
>>>>> help find the relevant sections better, because a lot of the meat in
>>>>> this draft is under sections 3.2.2 (ACE) and 3.2.3 (Option).
>>>>>
>>>>> I've responded below, but I'd prefer it if you would say these things
>>>>> on the list, so we have a better justification for changing the 
>>>>> draft.
>>>>> Reviewing a draft that has just be re-posted shouldn't imply anything
>>>>> about whether Apple is implementing the draft or not.
>>>>>
>>>>> (BTW, regarding the proposed change to the initial value of r.e1b, I
>>>>> was going to send that  to the list once co-authors agreed to it. But
>>>>> it would be nice if some of the edits were not self-generated.)
>>>>>
>>>>> See [BB] inline for responses (which I'll repeat if you send to the
>>>>> list).
>>>>>
>>>>> On 04/02/2022 07:37, Vidhi Goel wrote:
>>>>>> Hello Bob and authors,
>>>>>>
>>>>>> I have read the draft a few times and will send my extensive review
>>>>>> at a later point. But right now, I want to ask some critical things
>>>>>> that an implementation could benefit from.
>>>>>>
>>>>>> The draft says,
>>>>>> /If a TCP client has set the SYN to Not-ECT, but/
>>>>>> /   receives feedback that the IP-ECN field on the SYN arrived 
>>>>>> with a/
>>>>>> /   different codepoint, it can detect such middlebox interference
>>>>>> and/
>>>>>> /   send Not-ECT for the rest of the connection/
>>>>>>
>>>>>> 1. On the forward path from client to server, the client will revert
>>>>>> to Not-ECT when it sees for example, that it sent a Non-ECT SYN but
>>>>>> received ACE encoding other than 0 1 0. What does the client do in
>>>>>> the last ACK of the 3WHS -
>>>>>>  a. does it stay in AccECN mode and still send AccECN encoding based
>>>>>> on the IP code point of SYN-ACK (Table 4)? This would mean that
>>>>>> client won’t participate in ECN on the sender half of the connection
>>>>>> and only provide AccECN feedback to the server as a receiver.
>>>>>>  b.  does it disable AccECN mode and set ACE= 0 0 0 so that a server
>>>>>> in AccECN mode can disable ECN based on Table4?
>>>>>>
>>>>>> While writing this, I realized that the intention is probably a.
>>>>>> Could you confirm? Also, when the sender sets Not-ECT in its data
>>>>>> packets, it should also disable acting upon any ACE feedback as we
>>>>>> could still receive false ACE feedback from the server if the
>>>>>> network, lets say, changed 00 (not ECT) to 01(ECT1). If you 
>>>>>> agree, we
>>>>>> should add some text around this. Based on the current text, the
>>>>>> sender will always respond to the ACE feedback even if it sends
>>>>>> Not-ECT.
>>>>>>
>>>>>> TBH, this is a complicated scenario, where sender said to the 
>>>>>> network
>>>>>> - I don’t trust you so I can’t use ECN. Feel free to drop my 
>>>>>> packets.
>>>>>> And the network mangles the IP to ECT1 and then set CE (when
>>>>>> congested) which would be feedback’ed from the receiver. Now, the
>>>>>> packet wasn’t dropped which it should have been. So, is it better to
>>>>>> just ignore this feedback because sender doesn’t trust the 
>>>>>> network or
>>>>>> just act on it and reduce cwnd in order to reduce congestion in the
>>>>>> network somewhere.
>>>>>
>>>>> [BB] Your points are all good ones.
>>>>> I'll address this last complex scenario first, because it has
>>>>> implications for the earlier questions. First I'll define some
>>>>> terminology:
>>>>>
>>>>> Simple mangling scenario:
>>>>>
>>>>>   * some network function, e.g. broken Diffserv ToS-byte remarking,
>>>>>     illegally remarks a Not-ECT  SYN to CE.
>>>>>
>>>>> Your complex mangling scenario, repeated here:
>>>>>
>>>>>   * some network function, e.g. broken Diffserv ToS-byte remarking,
>>>>>     illegally remarks a Not-ECT SYN to ECT0 or ECT1,
>>>>>   * then congestion at a subsequent bottleneck is marking some
>>>>> packets CE
>>>>>
>>>>> Then potentially the client could tailor its behaviour after 
>>>>> sending a
>>>>> Not-ECT SYN, If the SYN/ACK feedback is
>>>>>
>>>>>  1. ECT: disable sending ECT, but continue responding to ECN feedback
>>>>>  2. CE, disable sending ECT, and disable response to ECN feedback
>>>>>
>>>>> However, this is uncertain, because CE feedback on the SYN/ACK could
>>>>> indicate either the simple or the complex mangling scenario.
>>>>>
>>>>> A simpler alternative would be to always continue responding to ECN
>>>>> feedback. Rationale:
>>>>>
>>>>>   * Whether case #1 or #2, assume that mangling of the SYN might have
>>>>>     been Not-ECT to ECT, even if the feedback off the SYN is CE.
>>>>>   * Then as the connection progresses, if /all/ feedback is CE,
>>>>>     there's already a recommendation to fall-back to disabling
>>>>>     congestion response.
>>>>>
>>>>> If we do this, I think we ought to say "SHOULD continue to respond to
>>>>> ECN feedback", not "MUST".
>>>>>
>>>>> And we'll need to put this all to the WG.
>>>>>
>>>>> 1.a) Now back to the beginning of your point. The text you quote is
>>>>> from §2.5 which is in the non-normative "Overview and Rationale"
>>>>> section (§2).
>>>>> You really need the normative text from:
>>>>>
>>>>>
>>>>>           3.2.2.3.
>>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3>Testing 
>>>>>
>>>>>           for Mangling of the IP/ECN Field
>>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the> 
>>>>>
>>>>>
>>>>>
>>>>> where it says (end of 1st para):
>>>>>
>>>>>     " ...for the remainder of the connection the client MUST NOT send
>>>>>     ECN-capable packets, but it MUST continue to feed back any ECN
>>>>>     markings on arriving packets."
>>>>>
>>>>> So, you're right, the answer is (a).
>>>>> It can't be (b), because, once in AccECN mode, 000 on the ACE field
>>>>> just becomes a counter value, and no longer negotiates the feedback
>>>>> mode.
>>>>>
>>>>> 1.b) As per the above, let's conservatively assume complex mangling.
>>>>> So, at the end of the first para quoted above I suggest we add:
>>>>>
>>>>>     " ...for the remainder of the connection the client MUST NOT send
>>>>>     ECN-capable packets *but**it**SHOULD continue to respond to ECN
>>>>>     feedback even though **it **is no longer sending ECN-capable
>>>>>     packets (see reasoning below)**. T**he client**MUST remain in the
>>>>>     same feedback mode and***it MUST continue to feed back any ECN
>>>>>     markings on arriving packets."
>>>>>
>>>>> I'll do the same for next para about the server.
>>>>> I'll work out some text for the reasoning, but I won't give it here.
>>>>>
>>>>> There are two other cases where it disables sending ECT, which don't
>>>>> say whether it continues to respond to congestion:
>>>>>
>>>>>   * "Testing for Mangling" section, penultimate para, where it's
>>>>>     receiving solid CE:
>>>>>     "Once a Data Sender has entered AccECN mode it SHOULD check
>>>>>     whether all feedback received for the first three or four rounds
>>>>>     indicated that every packet it sent was CE-marked. If so, for the
>>>>>     remainder of the connection, the Data Sender SHOULD NOT send
>>>>>     ECN-capable packets*and it SHOULD NOT respond to ECN feedback*,
>>>>>     but *it MUST remain in the same feedback mode and *it MUST
>>>>>     continue to feed back any ECN markings on arriving packets*(in 
>>>>> its
>>>>>     role as Data Receiver)*."
>>>>>   * Next section "Zeroing of the ACE Field", 2nd para:
>>>>>     "If the value of this ACE field is zero (0b000), the Data Sender
>>>>>     disables sending ECN-capable packets for the remainder of the
>>>>>     half-connection by setting the IP/ECN field in all subsequent
>>>>>     packets to Not-ECT, *but**it**SHOULD continue to respond to ECN
>>>>>     feedback even though **it **is no longer sending ECN-capable
>>>>>     packets.* *It MUST also remain in the same feedback mode and it
>>>>>     MUST continue to feed back any ECN markings on arriving packets
>>>>>     (in its role as Data Receiver).*"
>>>>>
>>>>> I'll also add reasoning (in the zeroing section, it already says that
>>>>> ACE=0b000 is not an unambiguous indication of mangling).
>>>>>
>>>>> And this prompts me to edit the bullet in an earlier section about 
>>>>> the
>>>>> obligation of a sender to respond to congestion feedback:
>>>>>
>>>>>
>>>>>         3.1.5.
>>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.1.5>Implications 
>>>>>
>>>>>         of AccECN Mode
>>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-implications-of-accecn-mode> 
>>>>>
>>>>>
>>>>>
>>>>>     "It is still obliged to respond appropriately to AccECN feedback
>>>>>     that indicates there were ECN marks on *ECN-capable *packets it
>>>>>     had previously sent."
>>>>>
>>>>> And I'll add a bullet:
>>>>>
>>>>>     "*If the sender chooses not to send ECN-capable packets (e.g.
>>>>>     because path traversal of the ECN field is suspect), it can 
>>>>> ignore
>>>>>     any **ECN **feedback about those packets if it is certain that it
>>>>>     cannot be valid (see Section 3.2.2, which gives normative
>>>>>     requirements for certain specific cases).*"
>>>>>
>>>>> I've said 'can' rather than 'MUST' because it's hard to cover all
>>>>> cases, e.g. single packets without ECT when it's not clear whether 
>>>>> the
>>>>> feedback covered another packet that was ECT.
>>>>>
>>>>> How does all this sound?
>>>>>
>>>>>> 2. On the reverse path from server to client, if a server sends a
>>>>>> Not-ECT SYN-ACK and receives ACE handshake encoding on last 
>>>>>> ACK other
>>>>>> than 0 1 0, there is no text like above that says server should send
>>>>>> Not-ECT for the rest of the connection (or at least I didn’t find
>>>>>> it). I think the server should also do same as client as the two
>>>>>> paths could be different. One could make it more complicated by
>>>>>> saying, if both client and server decide to not use ECN on their
>>>>>> corresponding sender half, then ECN should be disabled but let’s 
>>>>>> talk
>>>>>> about that later.
>>>>>
>>>>> [BB] It's in the second para of:
>>>>>
>>>>>
>>>>>           3.2.2.3.
>>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3>Testing 
>>>>>
>>>>>           for Mangling of the IP/ECN Field
>>>>> <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the> 
>>>>>
>>>>>
>>>>>
>>>>>> 3. In general, not setting the ECT (ECT0 or ECT1) code point on an
>>>>>> outgoing packet is different from supporting AccECN right as in the
>>>>>> host can still provide AccECN feedback on the receive path.
>>>>>
>>>>> [BB] Yup.
>>>>>
>>>>> Do you think any further explanation is needed in the above sections?
>>>>>
>>>>>
>>>>> Bob
>>>>>
>>>>>> ….
>>>>>>
>>>>>> To be continued if more questions come to my mind.
>>>>>>
>>>>>>
>>>>>> Thanks,
>>>>>> Vidhi
>>>>>>
>>>>>>> On Feb 3, 2022, at 7:24 AM, Bob Briscoe <ietf@bobbriscoe.net> 
>>>>>>> wrote:
>>>>>>>
>>>>>>> tcpm folks,
>>>>>>>
>>>>>>> This rev to accurate-ecn is the first of two. The second will
>>>>>>> hopefully follow on its heels in the next couple of days.
>>>>>>>
>>>>>>> Diffs between this first rev (-16) and -15:
>>>>>>>
>>>>>>> 1.  switches round two fairly large sections, so I've deferred 
>>>>>>> other
>>>>>>> changes to a second rev so the diffs won't be masked by the switch
>>>>>>> round of sections.
>>>>>>> Suggested by Ilpo to match the order in which the tests in these
>>>>>>> sections will be executed:
>>>>>>> * Test for mangling the IP-ECN field (now 3.2.2.3),
>>>>>>> * Then test for zeroing the ACE field (now 3.2.2.4).
>>>>>>>
>>>>>>> 2. Ilpo suggested some clarifications in "3.2.3.2.5. Consistency
>>>>>>> between AccECN Feedback Fields", which is about the receiver of
>>>>>>> feedback ensuring consistency between the mandatory 3-bit ACE field
>>>>>>> and the optional 24-bit counters. In brief (paraphrasing) it
>>>>>>> previously only said "MUST consider both fields", when it is now
>>>>>>> clearer that it actually meant "MUST reconcile both fields", so 
>>>>>>> that
>>>>>>> there is always a consistent baseline for subsequent ACKs.
>>>>>>>
>>>>>>> 3. A minor point is added in an appendix about the details that the
>>>>>>> pseudocode doesn't cover.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Bob
>>>>>>>
>>>>>>> PS. #2 & #3 were added to the XML ages ago (Jul '21), so you will
>>>>>>> have seen them in the HTML. However, prob due to my clumsiness, the
>>>>>>> posted TXT didn't include them whereas the posted XML did (ironic
>>>>>>> for a section about consistency). In turn, inconsistency was only
>>>>>>> possible because I am having to manually post the TXT for this
>>>>>>> draft, due to an unresolved issue with v3 RFC XML tables.
>>>>>>>
>>>>>>>
>>>>>>> On 03/02/2022 14:43, 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 TCP Maintenance and Minor
>>>>>>>> Extensions WG of the IETF.
>>>>>>>>
>>>>>>>>         Title           : More Accurate ECN Feedback in TCP
>>>>>>>>         Authors         : Bob Briscoe
>>>>>>>>                           Mirja Kühlewind
>>>>>>>>                           Richard Scheffenegger
>>>>>>>> Filename        : draft-ietf-tcpm-accurate-ecn-16.txt
>>>>>>>> Pages           : 60
>>>>>>>> Date            : 2022-02-03
>>>>>>>>
>>>>>>>> Abstract:
>>>>>>>>    Explicit Congestion Notification (ECN) is a mechanism where
>>>>>>>> network
>>>>>>>>    nodes can mark IP packets instead of dropping them to indicate
>>>>>>>>    incipient congestion to the end-points.  Receivers with an ECN-
>>>>>>>>    capable transport protocol feed back this information to the
>>>>>>>> sender.
>>>>>>>>    ECN was originally specified for TCP in such a way that only 
>>>>>>>> one
>>>>>>>>    feedback signal can be transmitted per Round-Trip Time (RTT).
>>>>>>>>  Recent
>>>>>>>>    new TCP mechanisms like Congestion Exposure (ConEx), Data
>>>>>>>> Center TCP
>>>>>>>>    (DCTCP) or Low Latency Low Loss Scalable Throughput (L4S) need
>>>>>>>> more
>>>>>>>>    accurate ECN feedback information whenever more than one
>>>>>>>> marking is
>>>>>>>>    received in one RTT.  This document specifies a scheme to 
>>>>>>>> provide
>>>>>>>>    more than one feedback signal per RTT in the TCP header.
>>>>>>>>  Given TCP
>>>>>>>>    header space is scarce, it allocates a reserved header bit
>>>>>>>> previously
>>>>>>>>    assigned to the ECN-Nonce.  It also overloads the two existing
>>>>>>>> ECN
>>>>>>>>    flags in the TCP header.  The resulting extra space is
>>>>>>>> exploited to
>>>>>>>>    feed back the IP-ECN field received during the 3-way 
>>>>>>>> handshake as
>>>>>>>>    well.  Supplementary feedback information can optionally be
>>>>>>>> provided
>>>>>>>>    in a new TCP option, which is never used on the TCP SYN.  The
>>>>>>>>    document also specifies the treatment of this updated TCP wire
>>>>>>>>    protocol by middleboxes.
>>>>>>>>
>>>>>>>>
>>>>>>>> The IETF datatracker status page for this draft is:
>>>>>>>> https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/
>>>>>>>>
>>>>>>>> There is also an htmlized version available at:
>>>>>>>> https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16 
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> A diff from the previous version is available at:
>>>>>>>> https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16
>>>>>>>>
>>>>>>>>
>>>>>>>> Internet-Drafts are also available by rsync at rsync.ietf.org
>>>>>>>> <http://rsync.ietf.org>::internet-drafts
>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> tcpm mailing list
>>>>>>>> tcpm@ietf.org
>>>>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>>>>
>>>>>>> -- 
>>>>>>> ________________________________________________________________
>>>>>>> Bob Briscoe http://bobbriscoe.net/
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> tcpm mailing list
>>>>>>> tcpm@ietf.org
>>>>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>>>>>
>>>>>
>>>>> -- 
>>>>> ________________________________________________________________
>>>>> Bob Briscoehttp://bobbriscoe.net/
>>>>
>>>>
>>>> _______________________________________________
>>>> tcpm mailing list
>>>> tcpm@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/tcpm
>>
>> -- 
>> ________________________________________________________________
>> Bob Briscoehttp://bobbriscoe.net/
>>

-- 
________________________________________________________________
Bob Briscoehttp://bobbriscoe.net/

--------------V8LckoMr5emQ54ffsERkSpct
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    Richard,see [BB] inline.<br>
    <br>
    <div class="moz-cite-prefix">On 06/02/2022 23:00, Scheffenegger,
      Richard wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:24dd6cbd-6668-23e2-6902-9772d0a1ef9a@gmx.at">Hi Bob,
      <br>
      <br>
      could it be, that the first "data" row and column in this table
      are
      <br>
      parially transposed?
      <br>
      <br>
      In the case of sending a Non-ECT SYN, but getting an indication of
      a
      <br>
      received CE on the SYN-ACK, why would you respond to that
      indication of
      <br>
      an incorrect CE marking? (With the response being, that the cwnd
      would
      <br>
      probably never grow beyond 1 or 2 packets per RTT, or possibly
      diminish
      <br>
      to n RTTs per packet, one a stack supports such low rates).
      <br>
      <br>
      I am thinking of the most common probable cause - a middlebox
      <br>
      unconditionally overriding the full TOS byte, and simply setting
      both
      <br>
      rightmost bits continously.
      <br>
      <br>
      Such a path would be rendered utterly unusable for an AccECN
      session, not?
      <br>
      <br>
      So at least that case would be a ND in your nomenclature.
      <br>
    </blockquote>
    <br>
    [BB] The full story for the first row (Not-ECT to ECT/CE
    transitions) was pointed out by Vidhi, and discussed in my previous
    response to her:<br>
    <ul>
      <li>Not-ECT to CE might have been two transitions</li>
      <ul>
        <li>Not-ECT to ECT at a broken ToS remarker (prob first hop); <br>
        </li>
        <li>ECT to CE at a subsequent bottleneck;</li>
      </ul>
      <li>We can only assume that a ToS remarking transition will be
        consistent. The congestion part could be transient.</li>
      <ul>
        <li>So /initially/ all cases where Not-ECT changes to anything
          else need to be treated as a change to ECT;</li>
        <li>But then there's /another/ test, which disables response to
          CE if there's solid CE marking;</li>
        <li>I.e. NR, that potentially becomes ND later.</li>
      </ul>
    </ul>
    <br>
    <blockquote type="cite"
      cite="mid:24dd6cbd-6668-23e2-6902-9772d0a1ef9a@gmx.at">
      <br>
      In the column - if the IP ECN is bleached, why disable response to
      CE
      <br>
      (if that is unlike to come back, as long as the session traverses
      the
      <br>
      bleaching path)? A "NR" seems more logical here in case of path
      changes?
      <br>
    </blockquote>
    <br>
    [BB] The thinking was that, if an ECT SYN becomes Not-ECT, then
    subsequently the sender only sends Not-ECT data, if that becomes CE,
    it seems pretty likely that is not a real congestion marking,
    whether there's a path change or not (because Not-ECT should remain
    Not-ECT).<br>
    <br>
    <blockquote type="cite"
      cite="mid:24dd6cbd-6668-23e2-6902-9772d0a1ef9a@gmx.at">
      <br>
      But again, all these ins-and-outs should be discussed in a
      different draft.
      <br>
    </blockquote>
    <br>
    [BB] Yes, deployment experience is needed before anything definitive
    can be said about fallback, because there will be cases where the
    recommendation depends on which of a number of possible odd
    behaviours is most likely.<br>
    <br>
    <blockquote type="cite"
      cite="mid:24dd6cbd-6668-23e2-6902-9772d0a1ef9a@gmx.at">
      <br>
      I fully agree, that only the normative response of the AccECN
      feedback
      <br>
      mechanism has to be fully defined - the two bullet point is
      exactly what
      <br>
      I expect to be stated.
      <br>
    </blockquote>
    <br>
    [BB] Great. I'll write that up.<br>
    <br>
    <br>
    Bob<br>
    <br>
    <blockquote type="cite"
      cite="mid:24dd6cbd-6668-23e2-6902-9772d0a1ef9a@gmx.at">
      <br>
      Regards,
      <br>
      <br>
         Richard
      <br>
      <br>
      <br>
      Am 06.02.2022 um 23:27 schrieb Bob Briscoe:
      <br>
      <blockquote type="cite">Richard,
        <br>
        <br>
        Indeed, I recently discovered I had missed an item that I added
        to my
        <br>
        ToDo list in Mar 2021, based on a conversation between Gorry and
        Mirja
        <br>
        where they both made the same point - that fall-back behaviour
        after
        <br>
        discovering ECN mangling ought to be non-normative in this
        draft,
        <br>
        because it's beyond the scope of the new feedback mechanism:
        <br>
<a class="moz-txt-link-freetext" href="https://mailarchive.ietf.org/arch/msg/tcpm/00URm1uwptgrAzKq96T6KSh8dQ4/">https://mailarchive.ietf.org/arch/msg/tcpm/00URm1uwptgrAzKq96T6KSh8dQ4/</a>
        <br>
        <br>
        We do need to say the following normatively in "Testing for
        Mangling of
        <br>
        the IP/ECN Field" because they are relevant to feedback, the
        Data Sender:
        <br>
        * remains in AccECN mode
        <br>
        * MUST continue to feed back any ECN markings on arriving
        packets.
        <br>
        <br>
        BTW, Vidhi, the text you quoted was from §2 on "Protocol
        Overview and
        <br>
        Rationale". That was meant to be an example of why it is useful
        for the
        <br>
        handshake to give feedback on exactly which IP-ECN codepoint
        arrived. It
        <br>
        wasn't meant to be a full spec. So I'll make it clearer that
        it's only
        <br>
        an example, rather than try to make it comprehensive, which
        isn't
        <br>
        appropriate at that point in the doc.
        <br>
        <br>
        Nonetheless, your points are correct, but if they are dealt with
        anyway,
        <br>
        they should be dealt with in §3.2.2.3. "Testing for Mangling of
        the
        <br>
        IP/ECN Field" where these scenarios are already covered (except
        for the
        <br>
        congestion response, which was silently left out of scope):
        <br>
<a class="moz-txt-link-freetext" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3">https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3</a>
        <br>
        <br>
        <br>
        We could just say that the questions of whether to set ECT for
        <br>
        subsequent packets, and whether to respond to feedback of CE
        markings
        <br>
        are beyond the scope of the draft. I think we should, but I
        think we
        <br>
        should also give some non-normative advice on what to do. It can
        only be
        <br>
        interim advice, not normative requirements. Both because it's
        out of
        <br>
        scope, and because the best behaviour and the alternatives will
        depend
        <br>
        on deployment experience. So they shouldn't be written
        normatively in a
        <br>
        Proposed Standard yet.
        <br>
        <br>
        We will do what the WG wants, but assuming it agrees to
        including
        <br>
        non-normative advice, following from your (Richard's) list of
        cases,
        <br>
        here's a full matrix. I don't propose this matrix should go in
        the
        <br>
        AccECN draft - it's just for this email, to check every case is
        covered.
        <br>
        <br>
        <br>
            Feedback in SYN/ACK of SYN's IP-ECN
        <br>
        IP-ECN on SYN     Not-ECT     ECT(0)     ECT(1)     CE
        <br>
        Not ECT     √     NR     NR     NR
        <br>
        ECT(0)     ND     √     √     √
        <br>
        ECT(1)     ND     CU
        <br>
            √     √
        <br>
        CE     n/a     n/a     n/a     n/a
        <br>
        <br>
        <br>
        Key:
        <br>
        √       Valid response
        <br>
        NR    Non-normative advice: Set *N*ot-ECT for subsequent
        packets, but
        <br>
        continue to *R*espond to CE feedback
        <br>
        ND    Non-normative advice: Set *N*ot-ECT for subsequent
        packets, and
        <br>
        *D*isable response to CE feedback
        <br>
        CU    Currently unused transition (not listed as invalid in the
        draft)
        <br>
        n/a    not applicable in the AccECN spec, in which there is
        never a case
        <br>
        where the sender sets CE when sending a packet.
        <br>
        <br>
        In the draft we don't need to mention any cases where CE is sent
        (but a
        <br>
        draft on fall-back probably would, because the sender might set
        CE to
        <br>
        test the path). We can cover the other cases with two groups of
        invalid
        <br>
        transitions: those where the SYN (or SYN-ACK) arriving at the
        receiver
        <br>
        appears to have been
        <br>
        i) ECN-capable
        <br>
        ii) Not-ECN-capable.
        <br>
        <br>
        I propose to also say that details (like what 'subsequent
        packets'
        <br>
        implies about how long this behaviour continues) are for future
        study.
        <br>
        <br>
        I'll try to write some text and propose it to this list. Perhaps
        it
        <br>
        should be in a new appendix?
        <br>
        <br>
        <br>
        Bob
        <br>
        <br>
        On 05/02/2022 12:01, Scheffenegger, Richard wrote:
        <br>
        <blockquote type="cite">Vidhi,
          <br>
          <br>
          Thanks for reading the draft very closely!
          <br>
          <br>
          Well, in the non-ECT SYN case, where the SYNACK indicates an
          ECT (ect0,
          <br>
          ect1, ce) is the obvious - and fully semantically correct -
          use case of
          <br>
          that possible use.
          <br>
          <br>
          For the set of three codepoints (ECT0, ECT1, CE) the only
          other possible
          <br>
          valid detection scenario would be if a SYN+CE is sent
          (deliberately,
          <br>
          althought one can argue that a sender should not originate
          CE-marked
          <br>
          packets), but a SYNACK without CE is returned (full or partial
          <br>
          bleaching). At least under the current RFC3168 definition of
          IP ECN
          <br>
          codepoints.
          <br>
          <br>
          allowed transitions:
          <br>
          nonECT -&gt; nonECT
          <br>
          <br>
          ECT0 -&gt; ECT0
          <br>
          ECT0 -&gt; ECT1
          <br>
          ECT0 -&gt; CE
          <br>
          <br>
          ECT1 -&gt; ECT1
          <br>
          ECT1 -&gt; ECT0 (possibly)
          <br>
          ECT1 -&gt; CE
          <br>
          <br>
          CE -&gt; CE
          <br>
          <br>
          <br>
          Thinking about this aspect more, I am now unsure, if the
          higher level
          <br>
          functions, such as detection of IP ECN codepoint mangling, or
          validating
          <br>
          across multiple RTTs, should be in this document - or not
          rather be in a
          <br>
          separate draft, in order to facilitate independent
          implementation and
          <br>
          improvement of mechanisms in that space.
          <br>
          <br>
          However, the above table of allowed IP ECN transistions may be
          updated
          <br>
          independently of AccECN - and performing extensive checks as
          part of a
          <br>
          signal protocol ossifies this IMHO.
          <br>
          <br>
          As a middle ground, perhaps having these descriptions in a
          non-normative
          <br>
          section, clearly stating these to be examples of what an
          impemented
          <br>
          could choose to do?
          <br>
          <br>
          Richard
          <br>
          <br>
          Am 05.02.2022 um 02:14 schrieb Vidhi Goel:
          <br>
          <blockquote type="cite">I like your response for all the
            points. I can review the diff for the
            <br>
            proposed changes, if you’d like before committing them to
            the draft.
            <br>
            <br>
            Sorry, I want to add a new comment for this text,
            <br>
            <br>
            If a TCP client has set the SYN to Not-ECT, but
            <br>
                receives feedback that the IP-ECN field on the SYN
            arrived with a
            <br>
                different codepoint, it can detect such middlebox
            interference and
            <br>
                send Not-ECT for the rest of the connection.
            <br>
            <br>
            <br>
            This statement holds good for Not-ECT but doesn’t if lets
            say the SYN
            <br>
            was ECT0 or ECT1 as those could be marked CE and still be
            valid. Should
            <br>
            we add a statement for ECT marked SYN as well?
            <br>
            <br>
            Thanks,
            <br>
            Vidhi
            <br>
            <br>
            <blockquote type="cite">On Feb 4, 2022, at 5:44 AM, Bob
              Briscoe &lt;<a class="moz-txt-link-abbreviated" href="mailto:ietf@bobbriscoe.net">ietf@bobbriscoe.net</a>
              <br>
              <a class="moz-txt-link-rfc2396E" href="mailto:ietf@bobbriscoe.net">&lt;mailto:ietf@bobbriscoe.net&gt;</a>&gt; wrote:
              <br>
              <br>
              Vidhi,
              <br>
              <br>
              I've increased the table of contents to tocdepth = "4"
              which might
              <br>
              help find the relevant sections better, because a lot of
              the meat in
              <br>
              this draft is under sections 3.2.2 (ACE) and 3.2.3
              (Option).
              <br>
              <br>
              I've responded below, but I'd prefer it if you would say
              these things
              <br>
              on the list, so we have a better justification for
              changing the draft.
              <br>
              Reviewing a draft that has just be re-posted shouldn't
              imply anything
              <br>
              about whether Apple is implementing the draft or not.
              <br>
              <br>
              (BTW, regarding the proposed change to the initial value
              of r.e1b, I
              <br>
              was going to send that  to the list once co-authors agreed
              to it. But
              <br>
              it would be nice if some of the edits were not
              self-generated.)
              <br>
              <br>
              See [BB] inline for responses (which I'll repeat if you
              send to the
              <br>
              list).
              <br>
              <br>
              On 04/02/2022 07:37, Vidhi Goel wrote:
              <br>
              <blockquote type="cite">Hello Bob and authors,
                <br>
                <br>
                I have read the draft a few times and will send my
                extensive review
                <br>
                at a later point. But right now, I want to ask some
                critical things
                <br>
                that an implementation could benefit from.
                <br>
                <br>
                The draft says,
                <br>
                /If a TCP client has set the SYN to Not-ECT, but/
                <br>
                /   receives feedback that the IP-ECN field on the SYN
                arrived with a/
                <br>
                /   different codepoint, it can detect such middlebox
                interference
                <br>
                and/
                <br>
                /   send Not-ECT for the rest of the connection/
                <br>
                <br>
                1. On the forward path from client to server, the client
                will revert
                <br>
                to Not-ECT when it sees for example, that it sent a
                Non-ECT SYN but
                <br>
                received ACE encoding other than 0 1 0. What does the
                client do in
                <br>
                the last ACK of the 3WHS -
                <br>
                 a. does it stay in AccECN mode and still send AccECN
                encoding based
                <br>
                on the IP code point of SYN-ACK (Table 4)? This would
                mean that
                <br>
                client won’t participate in ECN on the sender half of
                the connection
                <br>
                and only provide AccECN feedback to the server as a
                receiver.
                <br>
                 b.  does it disable AccECN mode and set ACE= 0 0 0 so
                that a server
                <br>
                in AccECN mode can disable ECN based on Table4?
                <br>
                <br>
                While writing this, I realized that the intention
                is probably a.
                <br>
                Could you confirm? Also, when the sender sets Not-ECT in
                its data
                <br>
                packets, it should also disable acting upon any ACE
                feedback as we
                <br>
                could still receive false ACE feedback from the server
                if the
                <br>
                network, lets say, changed 00 (not ECT) to 01(ECT1). If
                you agree, we
                <br>
                should add some text around this. Based on the current
                text, the
                <br>
                sender will always respond to the ACE feedback even if
                it sends
                <br>
                Not-ECT.
                <br>
                <br>
                TBH, this is a complicated scenario, where sender said
                to the network
                <br>
                - I don’t trust you so I can’t use ECN. Feel free to
                drop my packets.
                <br>
                And the network mangles the IP to ECT1 and then set CE
                (when
                <br>
                congested) which would be feedback’ed from the receiver.
                Now, the
                <br>
                packet wasn’t dropped which it should have been. So, is
                it better to
                <br>
                just ignore this feedback because sender doesn’t trust
                the network or
                <br>
                just act on it and reduce cwnd in order to reduce
                congestion in the
                <br>
                network somewhere.
                <br>
              </blockquote>
              <br>
              [BB] Your points are all good ones.
              <br>
              I'll address this last complex scenario first, because it
              has
              <br>
              implications for the earlier questions. First I'll define
              some
              <br>
              terminology:
              <br>
              <br>
              Simple mangling scenario:
              <br>
              <br>
                * some network function, e.g. broken Diffserv ToS-byte
              remarking,
              <br>
                  illegally remarks a Not-ECT  SYN to CE.
              <br>
              <br>
              Your complex mangling scenario, repeated here:
              <br>
              <br>
                * some network function, e.g. broken Diffserv ToS-byte
              remarking,
              <br>
                  illegally remarks a Not-ECT SYN to ECT0 or ECT1,
              <br>
                * then congestion at a subsequent bottleneck is marking
              some
              <br>
              packets CE
              <br>
              <br>
              Then potentially the client could tailor its behaviour
              after sending a
              <br>
              Not-ECT SYN, If the SYN/ACK feedback is
              <br>
              <br>
               1. ECT: disable sending ECT, but continue responding to
              ECN feedback
              <br>
               2. CE, disable sending ECT, and disable response to ECN
              feedback
              <br>
              <br>
              However, this is uncertain, because CE feedback on the
              SYN/ACK could
              <br>
              indicate either the simple or the complex mangling
              scenario.
              <br>
              <br>
              A simpler alternative would be to always continue
              responding to ECN
              <br>
              feedback. Rationale:
              <br>
              <br>
                * Whether case #1 or #2, assume that mangling of the SYN
              might have
              <br>
                  been Not-ECT to ECT, even if the feedback off the SYN
              is CE.
              <br>
                * Then as the connection progresses, if /all/ feedback
              is CE,
              <br>
                  there's already a recommendation to fall-back to
              disabling
              <br>
                  congestion response.
              <br>
              <br>
              If we do this, I think we ought to say "SHOULD continue to
              respond to
              <br>
              ECN feedback", not "MUST".
              <br>
              <br>
              And we'll need to put this all to the WG.
              <br>
              <br>
              1.a) Now back to the beginning of your point. The text you
              quote is
              <br>
              from §2.5 which is in the non-normative "Overview and
              Rationale"
              <br>
              section (§2).
              <br>
              You really need the normative text from:
              <br>
              <br>
              <br>
                        3.2.2.3.
              <br>
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3&gt;</a>Testing
              <br>
                        for Mangling of the IP/ECN Field
              <br>
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the&gt;</a>
              <br>
              <br>
              <br>
              where it says (end of 1st para):
              <br>
              <br>
                  " ...for the remainder of the connection the client
              MUST NOT send
              <br>
                  ECN-capable packets, but it MUST continue to feed back
              any ECN
              <br>
                  markings on arriving packets."
              <br>
              <br>
              So, you're right, the answer is (a).
              <br>
              It can't be (b), because, once in AccECN mode, 000 on the
              ACE field
              <br>
              just becomes a counter value, and no longer negotiates the
              feedback
              <br>
              mode.
              <br>
              <br>
              1.b) As per the above, let's conservatively assume complex
              mangling.
              <br>
              So, at the end of the first para quoted above I suggest we
              add:
              <br>
              <br>
                  " ...for the remainder of the connection the client
              MUST NOT send
              <br>
                  ECN-capable packets *but**it**SHOULD continue to
              respond to ECN
              <br>
                  feedback even though **it **is no longer sending
              ECN-capable
              <br>
                  packets (see reasoning below)**. T**he client**MUST
              remain in the
              <br>
                  same feedback mode and***it MUST continue to feed back
              any ECN
              <br>
                  markings on arriving packets."
              <br>
              <br>
              I'll do the same for next para about the server.
              <br>
              I'll work out some text for the reasoning, but I won't
              give it here.
              <br>
              <br>
              There are two other cases where it disables sending ECT,
              which don't
              <br>
              say whether it continues to respond to congestion:
              <br>
              <br>
                * "Testing for Mangling" section, penultimate para, 
              where it's
              <br>
                  receiving solid CE:
              <br>
                  "Once a Data Sender has entered AccECN mode it SHOULD
              check
              <br>
                  whether all feedback received for the first three or
              four rounds
              <br>
                  indicated that every packet it sent was CE-marked. If
              so, for the
              <br>
                  remainder of the connection, the Data Sender SHOULD
              NOT send
              <br>
                  ECN-capable packets*and it SHOULD NOT respond to ECN
              feedback*,
              <br>
                  but *it MUST remain in the same feedback mode and *it
              MUST
              <br>
                  continue to feed back any ECN markings on arriving
              packets*(in its
              <br>
                  role as Data Receiver)*."
              <br>
                * Next section "Zeroing of the ACE Field", 2nd para:
              <br>
                  "If the value of this ACE field is zero (0b000), the
              Data Sender
              <br>
                  disables sending ECN-capable packets for the remainder
              of the
              <br>
                  half-connection by setting the IP/ECN field in all
              subsequent
              <br>
                  packets to Not-ECT, *but**it**SHOULD continue to
              respond to ECN
              <br>
                  feedback even though **it **is no longer sending
              ECN-capable
              <br>
                  packets.* *It MUST also remain in the same feedback
              mode and it
              <br>
                  MUST continue to feed back any ECN markings on
              arriving packets
              <br>
                  (in its role as Data Receiver).*"
              <br>
              <br>
              I'll also add reasoning (in the zeroing section, it
              already says that
              <br>
              ACE=0b000 is not an unambiguous indication of mangling).
              <br>
              <br>
              And this prompts me to edit the bullet in an earlier
              section about the
              <br>
              obligation of a sender to respond to congestion feedback:
              <br>
              <br>
              <br>
                      3.1.5.
              <br>
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.1.5">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.1.5&gt;</a>Implications
              <br>
                      of AccECN Mode
              <br>
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-implications-of-accecn-mode">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-implications-of-accecn-mode&gt;</a>
              <br>
              <br>
              <br>
                  "It is still obliged to respond appropriately to
              AccECN feedback
              <br>
                  that indicates there were ECN marks on *ECN-capable
              *packets it
              <br>
                  had previously sent."
              <br>
              <br>
              And I'll add a bullet:
              <br>
              <br>
                  "*If the sender chooses not to send ECN-capable
              packets (e.g.
              <br>
                  because path traversal of the ECN field is suspect),
              it can ignore
              <br>
                  any **ECN **feedback about those packets if it is
              certain that it
              <br>
                  cannot be valid (see Section 3.2.2, which gives
              normative
              <br>
                  requirements for certain specific cases).*"
              <br>
              <br>
              I've said 'can' rather than 'MUST' because it's hard to
              cover all
              <br>
              cases, e.g. single packets without ECT when it's not clear
              whether the
              <br>
              feedback covered another packet that was ECT.
              <br>
              <br>
              How does all this sound?
              <br>
              <br>
              <blockquote type="cite">2. On the reverse path from server
                to client, if a server sends a
                <br>
                Not-ECT SYN-ACK and receives ACE handshake encoding on
                last ACK other
                <br>
                than 0 1 0, there is no text like above that says server
                should send
                <br>
                Not-ECT for the rest of the connection (or at least
                I didn’t find
                <br>
                it). I think the server should also do same as client
                as the two
                <br>
                paths could be different. One could make it more
                complicated by
                <br>
                saying, if both client and server decide to not use ECN
                on their
                <br>
                corresponding sender half, then ECN should be disabled
                but let’s talk
                <br>
                about that later.
                <br>
              </blockquote>
              <br>
              [BB] It's in the second para of:
              <br>
              <br>
              <br>
                        3.2.2.3.
              <br>
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#section-3.2.2.3&gt;</a>Testing
              <br>
                        for Mangling of the IP/ECN Field
              <br>
<a class="moz-txt-link-rfc2396E" href="https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the">&lt;https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-16.html#name-testing-for-mangling-of-the&gt;</a>
              <br>
              <br>
              <br>
              <blockquote type="cite">3. In general, not setting the ECT
                (ECT0 or ECT1) code point on an
                <br>
                outgoing packet is different from supporting AccECN
                right as in the
                <br>
                host can still provide AccECN feedback on the receive
                path.
                <br>
              </blockquote>
              <br>
              [BB] Yup.
              <br>
              <br>
              Do you think any further explanation is needed in the
              above sections?
              <br>
              <br>
              <br>
              Bob
              <br>
              <br>
              <blockquote type="cite">….
                <br>
                <br>
                To be continued if more questions come to my mind.
                <br>
                <br>
                <br>
                Thanks,
                <br>
                Vidhi
                <br>
                <br>
                <blockquote type="cite">On Feb 3, 2022, at 7:24 AM, Bob
                  Briscoe <a class="moz-txt-link-rfc2396E" href="mailto:ietf@bobbriscoe.net">&lt;ietf@bobbriscoe.net&gt;</a> wrote:
                  <br>
                  <br>
                  tcpm folks,
                  <br>
                  <br>
                  This rev to accurate-ecn is the first of two. The
                  second will
                  <br>
                  hopefully follow on its heels in the next couple of
                  days.
                  <br>
                  <br>
                  Diffs between this first rev (-16) and -15:
                  <br>
                  <br>
                  1.  switches round two fairly large sections, so I've
                  deferred other
                  <br>
                  changes to a second rev so the diffs won't be masked
                  by the switch
                  <br>
                  round of sections.
                  <br>
                  Suggested by Ilpo to match the order in which the
                  tests in these
                  <br>
                  sections will be executed:
                  <br>
                  * Test for mangling the IP-ECN field (now 3.2.2.3),
                  <br>
                  * Then test for zeroing the ACE field (now 3.2.2.4).
                  <br>
                  <br>
                  2. Ilpo suggested some clarifications in "3.2.3.2.5.
                  Consistency
                  <br>
                  between AccECN Feedback Fields", which is about the
                  receiver of
                  <br>
                  feedback ensuring consistency between the mandatory
                  3-bit ACE field
                  <br>
                  and the optional 24-bit counters. In brief
                  (paraphrasing) it
                  <br>
                  previously only said "MUST consider both fields", when
                  it is now
                  <br>
                  clearer that it actually meant "MUST reconcile both
                  fields", so that
                  <br>
                  there is always a consistent baseline for subsequent
                  ACKs.
                  <br>
                  <br>
                  3. A minor point is added in an appendix about the
                  details that the
                  <br>
                  pseudocode doesn't cover.
                  <br>
                  <br>
                  <br>
                  <br>
                  Bob
                  <br>
                  <br>
                  PS. #2 &amp; #3 were added to the XML ages ago (Jul
                  '21), so you will
                  <br>
                  have seen them in the HTML. However, prob due to my
                  clumsiness, the
                  <br>
                  posted TXT didn't include them whereas the posted XML
                  did (ironic
                  <br>
                  for a section about consistency). In turn,
                  inconsistency was only
                  <br>
                  possible because I am having to manually post the TXT
                  for this
                  <br>
                  draft, due to an unresolved issue with v3 RFC XML
                  tables.
                  <br>
                  <br>
                  <br>
                  On 03/02/2022 14:43, <a class="moz-txt-link-abbreviated" href="mailto:internet-drafts@ietf.org">internet-drafts@ietf.org</a> wrote:
                  <br>
                  <blockquote type="cite">A New Internet-Draft is
                    available from the on-line Internet-Drafts
                    <br>
                    directories.
                    <br>
                    This draft is a work item of the TCP Maintenance and
                    Minor
                    <br>
                    Extensions WG of the IETF.
                    <br>
                    <br>
                            Title           : More Accurate ECN Feedback
                    in TCP
                    <br>
                            Authors         : Bob Briscoe
                    <br>
                                              Mirja Kühlewind
                    <br>
                                              Richard Scheffenegger
                    <br>
                    Filename        :
                    draft-ietf-tcpm-accurate-ecn-16.txt
                    <br>
                    Pages           : 60
                    <br>
                    Date            : 2022-02-03
                    <br>
                    <br>
                    Abstract:
                    <br>
                       Explicit Congestion Notification (ECN) is a
                    mechanism where
                    <br>
                    network
                    <br>
                       nodes can mark IP packets instead of dropping
                    them to indicate
                    <br>
                       incipient congestion to the end-points.
                     Receivers with an ECN-
                    <br>
                       capable transport protocol feed back this
                    information to the
                    <br>
                    sender.
                    <br>
                       ECN was originally specified for TCP in such a
                    way that only one
                    <br>
                       feedback signal can be transmitted per Round-Trip
                    Time (RTT).
                    <br>
                     Recent
                    <br>
                       new TCP mechanisms like Congestion Exposure
                    (ConEx), Data
                    <br>
                    Center TCP
                    <br>
                       (DCTCP) or Low Latency Low Loss Scalable
                    Throughput (L4S) need
                    <br>
                    more
                    <br>
                       accurate ECN feedback information whenever more
                    than one
                    <br>
                    marking is
                    <br>
                       received in one RTT.  This document specifies a
                    scheme to provide
                    <br>
                       more than one feedback signal per RTT in the TCP
                    header.
                    <br>
                     Given TCP
                    <br>
                       header space is scarce, it allocates a reserved
                    header bit
                    <br>
                    previously
                    <br>
                       assigned to the ECN-Nonce.  It also overloads the
                    two existing
                    <br>
                    ECN
                    <br>
                       flags in the TCP header.  The resulting extra
                    space is
                    <br>
                    exploited to
                    <br>
                       feed back the IP-ECN field received during the
                    3-way handshake as
                    <br>
                       well.  Supplementary feedback information can
                    optionally be
                    <br>
                    provided
                    <br>
                       in a new TCP option, which is never used on the
                    TCP SYN.  The
                    <br>
                       document also specifies the treatment of this
                    updated TCP wire
                    <br>
                       protocol by middleboxes.
                    <br>
                    <br>
                    <br>
                    The IETF datatracker status page for this draft is:
                    <br>
<a class="moz-txt-link-freetext" href="https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/">https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/</a>
                    <br>
                    <br>
                    There is also an htmlized version available at:
                    <br>
<a class="moz-txt-link-freetext" href="https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16">https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-16</a>
                    <br>
                    <br>
                    <br>
                    A diff from the previous version is available at:
                    <br>
<a class="moz-txt-link-freetext" href="https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16">https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-16</a>
                    <br>
                    <br>
                    <br>
                    Internet-Drafts are also available by rsync at
                    rsync.ietf.org
                    <br>
                    <a class="moz-txt-link-rfc2396E" href="http://rsync.ietf.org">&lt;http://rsync.ietf.org&gt;</a>::internet-drafts
                    <br>
                    <br>
                    <br>
                    _______________________________________________
                    <br>
                    tcpm mailing list
                    <br>
                    <a class="moz-txt-link-abbreviated" href="mailto:tcpm@ietf.org">tcpm@ietf.org</a>
                    <br>
                    <a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/tcpm">https://www.ietf.org/mailman/listinfo/tcpm</a>
                    <br>
                  </blockquote>
                  <br>
                  --
                  <br>
________________________________________________________________
                  <br>
                  Bob Briscoe <a class="moz-txt-link-freetext" href="http://bobbriscoe.net/">http://bobbriscoe.net/</a>
                  <br>
                  <br>
                  _______________________________________________
                  <br>
                  tcpm mailing list
                  <br>
                  <a class="moz-txt-link-abbreviated" href="mailto:tcpm@ietf.org">tcpm@ietf.org</a>
                  <br>
                  <a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/tcpm">https://www.ietf.org/mailman/listinfo/tcpm</a>
                  <br>
                </blockquote>
                <br>
              </blockquote>
              <br>
              --
              <br>
________________________________________________________________
              <br>
              Bob Briscoehttp://bobbriscoe.net/
              <br>
            </blockquote>
            <br>
            <br>
            _______________________________________________
            <br>
            tcpm mailing list
            <br>
            <a class="moz-txt-link-abbreviated" href="mailto:tcpm@ietf.org">tcpm@ietf.org</a>
            <br>
            <a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/tcpm">https://www.ietf.org/mailman/listinfo/tcpm</a>
            <br>
          </blockquote>
        </blockquote>
        <br>
        --
        <br>
        ________________________________________________________________
        <br>
        Bob Briscoehttp://bobbriscoe.net/
        <br>
        <br>
      </blockquote>
    </blockquote>
    <br>
    <pre class="moz-signature" cols="72">-- 
________________________________________________________________
Bob Briscoe                               <a class="moz-txt-link-freetext" href="http://bobbriscoe.net/">http://bobbriscoe.net/</a></pre>
  </body>
</html>

--------------V8LckoMr5emQ54ffsERkSpct--


From nobody Fri Feb 11 21:34:43 2022
Return-Path: <kojo@cs.helsinki.fi>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7C1B53A1371; Fri, 11 Feb 2022 21:34:42 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, SPF_PASS=-0.001,  URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cs.helsinki.fi
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qFlC8sgtLhNG; Fri, 11 Feb 2022 21:34:37 -0800 (PST)
Received: from script.cs.helsinki.fi (script.cs.helsinki.fi [128.214.11.1]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8A2203A136E; Fri, 11 Feb 2022 21:34:35 -0800 (PST)
X-DKIM: Courier DKIM Filter v0.50+pk-2017-10-25 mail.cs.helsinki.fi Sat, 12 Feb 2022 07:34:25 +0200
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cs.helsinki.fi; h=date:from:to:cc:subject:in-reply-to:message-id:references :mime-version:content-type:content-id; s=dkim20130528; bh=TD2mmH G4GPTZhaCG9Qw7MebrIwmlbVn/3nHqVthxbfw=; b=BaRqyGiMHx01zwbo/vhN7E X/wtfCd+bkUwzOIZFszsZsrv8ZcdjCrnVzGb4S7Bh/31AsxGqCnezZgoC0WUo4q7 9Rkh8DsICkSgJeglV9cNk/yCht/pV/82fJA6lqofIbFdShOKpmojZlqgTtCPbKUv 8DRxU9HBN7FSsy8QI8uMg=
Received: from hp8x-60 (88-113-49-197.elisa-laajakaista.fi [88.113.49.197]) (AUTH: PLAIN kojo, TLS: TLSv1/SSLv3,256bits,AES256-GCM-SHA384) by mail.cs.helsinki.fi with ESMTPSA; Sat, 12 Feb 2022 07:34:25 +0200 id 00000000005A1C65.00000000620746E1.00005D7E
Date: Sat, 12 Feb 2022 07:34:24 +0200 (EET)
From: Markku Kojo <kojo@cs.helsinki.fi>
To: Yoshifumi Nishida <nsd.ietf@gmail.com>
cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
In-Reply-To: <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com>
Message-ID: <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com>
User-Agent: Alpine 2.21 (DEB 202 2017-01-01)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="=_script-23958-1644644065-0001-2"
Content-ID: <alpine.DEB.2.21.2202120716540.4019@hp8x-60.cs.helsinki.fi>
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/fc0mrMdX0I3cglZwCa4crqc-01g>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 12 Feb 2022 05:34:43 -0000

This is a MIME-formatted message.  If you see this text it means that your
E-mail software does not support MIME-formatted messages.

--=_script-23958-1644644065-0001-2
Content-Type: text/plain; charset="iso-8859-15"; format=flowed
Content-Transfer-Encoding: quoted-printable
Content-ID: <alpine.DEB.2.21.2202120410011.4019@hp8x-60.cs.helsinki.fi>

Hi Yoshi, all,

It seems to me that many issues that I raised have been solved, thanks.=20
However, there are still a number of important issues that have not been=20
addressed adequately. At least the following:

#135 on W_max. Yoshi's observation was correct that this is not resolved:=20
the co-authors and original developpers of CUBIC (@lisongxu and=20
@sangtaeha) agreed in their last message that Wmax needs different=20
treatment for slow start and congestion avoidance and plan comprehensive=20
(new) evaluation of it. This is obviously an open issue but the issue=20
was closed?

#85 (& #86 with basically same issue and these two were combined) This=20
(#85) is about ECN but the major issue is on using the same MD=20
factor in slow start and in congestion avoidance when using=20
loss-based CC. This (#85) remained closed even though I provided a=20
thorough explanation why it is wrong and against the original theory and=20
design by Van Jacobson, against the congestion control principles (RFC=20
2914) and two co-authors agreed on this in their same last message to=20
#135 when they agreed on Wmax needing rework. This is an important issue=20
that the wg should consider very carefully because it is not only=20
updating RFC 5681 but also in conflict with RFC 2914. How can=20
tcpm (and IETF) suggest and allow one CC algo to not follow congestion=20
control principles as set in RFC 2914 while requiring all other CCs to=20
follow RFC 2914 guidelines?
The current draft does not provide any justification for using the same=20
MD factor in slow start as in congestion avoidance. Nor am I=20
awere of any experimental data that would support this change.=20
The fact thet CUBIC has been long deployed does not alone provide any=20
supporting evidence because CUBIC is likely to give good performance as=20
it is overagressive and thereby unfair to competing traffic and users=20
tend to be happy when measuring the performance of the sending CUBIC=20
only, not the competing traffic that is badly impacted. HyStart++ is=20
suggested as mitigation to the problem but it cannot; HyStart++ is only=20
applicable during initial slow start, not during slow start after RTO!=20
That is, the "SHOULD use HYStart++" text in Sec 4.10 is impossible=20
to implement as I have pointed out in my comments earlier. Using a proper=20
MD factor in slow start is even more important if loss is detectected=20
during a RTO recovery because the sender is likely to face heavy=20
congestion in such a case and it is very bad if the sender continues=20
sending with overaggressive rate, stealing the capacity from and causing=20
harm to coexisting flows. In addition, as I have explained, HyStart++=20
does not remove the problem even for the initial slow start as it is not=20
shown to work always. Instead, the results with the HyStart++ draft show=20
that it reduces 50% of rexmits and only 36% RTOs, meaning that there is=20
likely to be a notable percentage of cases when a sender is still=20
in slow start when first loss is detected (i.e., HyStart++ had no effect)=20
and a significant number of cases where a CUBIC sender is=20
overaggressive continuing with a 1-40% larger cwnd than what is the=20
available capacity. Note also that any delay-based heuristics like=20
HyStart++ are known to work poorly in various wireless environmens where=20
link delay tends to vary a lot. We may come up with some other MD factor=20
that 0.5 when in slow start and HyStart++ is in use, but that is=20
experimental, if not research, and definitely not ready for stds track.

#114, #132, and #143 w.r.t flightsize vs. cwnd. The current text
      does not quite correcly reflect what stacks that use cwnd do.
      I'll comment in #143 separately.

#96 & #98: The text added does not address the problems raised which are=20
also evidenced in the paper pointed by Bob in #96. Even though CUBIC has=20
been modified a bit after the paper was published, it does not=20
automatically mean that the problem has been shown resolved: experimental=20
evidence is required but not provided. The fact that CUBIC does not=20
change MD factor for fast convergence is the root of the problem=20
evidenced in the paper and remains so in the algo specified in this=20
draft. This is also a significant problem when competing with Reno CC=20
because CUBIC behaves much more aggressive than Reno CC when there is=20
sudden congestion and all competing flows must converge fast down to a=20
small fraction of the current cwnd to be fair to each other. This again=20
cannot be evidenced not to be a problem by long deployment experience=20
unless experimental data that measures the impact on competing traffic is=20
presented to back the claims. Adjusting just Wmax for fast convergence is=20
not enough and is even likely to be ineffective because there tend to be=20
several losses when sudden congestion is hit, and particularly if NewReno=20
is in use the sender stays several RTTs in Fast Recovery being=20
overaggressive and then possibly continues at the same rate in CA which=20
is unlikely to reach evan close to Wmax before a new loss hits=20
the sender again. That is, lower Wmax and lower additive increase factor=20
do not compensate the use of larger MD factor when sudden congestion is=20
encountered.

#93 & #94 & (#89) Sec 5.3 still does not address any difficult=20
environments, in particular buffer-bloated paths (nor does Sec 5.4).
We need evidence (results) that show CUBIC is fair towards other CCs=20
(Reno) also in such environments. Note that CUBIC's decision to leave=20
Reno-friendly region is based on the size of cwnd which tends to be=20
incorrect with buffer-bloated bottlenecks because with huge buffers the=20
cwnd can be many times larger than what is actually needed to fully=20
utilize the available network bit-rate. Therefore, Reno CC has no problem=20
in fully utilizing such bottleneck links and CUBIC must stay in=20
Reno-friendly region longer but it leaves it too early because the same C=20
is used as with non-bloated environments. We lack experiments showing=20
CUBIC follows the congestion control principles and is fair to current=20
standard TCP CC; to my understanding no experiments with buffer-bloated
bottlenecks are cited to back up the claims even though buffer bloat is=20
very well known to be a common (difficult) environment in today's=20
Internet.

#90 The current text on applying undo (a response to detected false=20
fastrexmit) does not provide correct result if someone implements it.
I have explained the problems there in github but seem to have not=20
replied to latest comments by Neal. I'll reply and try to explain more.
Again one major problem here is that the draft suggest a new response=20
algo for false fast rexmits but does not provide any experimental data to=20
support it. Long deployment experience has been suggested as=20
justification but again without any carefully evaluated experimental=20
data and evidence there is no meat. The issue is important to solve but i=
s=20
not specific to CUBIC. Instead, it is general problem for all TCP CC=20
variants. IMHO, this is not ready for standards track but deserves a draf=
t=20
of its own so that it can be carefully evaluated and discussed ion the=20
list. AFAIK there has been no discussion on this on the tcpm list, so=20
those probably interested and having experience are likely to be unaware=20
that this is part of CUBIC draft.

#88 The problem with correctness of the AIMD model and setting alpha for=20
CUBIC requires further consideration. Bob provided an analysis that=20
leaves things still open. It seems that I never had time to review and=20
comment the analysis and clarify why the model does not work. I'll do tha=
t=20
separately as it is important to ensure CUBIC behaves fairly as intended=20
for the Reno-friendly region.

Best regards,

/Markku

On Mon, 31 Jan 2022, Yoshifumi Nishida wrote:

> Hello,
>=20
> After some discussions among chairs, we decided to run the 2nd WGLC on =
draft-ietf-tcpm-rfc8312bis in
> consideration of the importance of the draft.=A0
> We'll be grateful if you could send your feedback to the ML. The WGLC r=
uns until *Feb 11*.
>=20
> If interested, you can check in-depth past discussions in the following =
URL.
> https://github.com/NTAP/rfc8312bis/
>=20
> Thank you so much!
> --
> tcpm co-chairs
>=20
>=20
> On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <lars@eggert.org> wrote:
>       Hi,
>
>       this -06 version rolls in all the changes requested during (and a=
fter) WGLC ended.
>
>       I'll leave it up to the chairs to decide if another WGLC is warra=
nted or the document can
>       progress as-is.
>
>       Thanks,
>       Lars
>=20
>
>       > On 2022-1-26, at 11:12, internet-drafts@ietf.org wrote:
>       >
>       >
>       > A New Internet-Draft is available from the on-line Internet-Dra=
fts directories.
>       > This draft is a work item of the TCP Maintenance and Minor Exte=
nsions WG of the IETF.
>       >
>       >=A0 =A0 =A0 =A0 Title=A0 =A0 =A0 =A0 =A0 =A0: CUBIC for Fast and =
Long-Distance Networks
>       >=A0 =A0 =A0 =A0 Authors=A0 =A0 =A0 =A0 =A0: Lisong Xu
>       >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Sangtae Ha
>       >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Injong Rhee
>       >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Vidhi Goel
>       >=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Lars Eggert
>       >=A0 =A0 =A0 =A0Filename=A0 =A0 =A0 =A0 : draft-ietf-tcpm-rfc8312=
bis-06.txt
>       >=A0 =A0 =A0 =A0Pages=A0 =A0 =A0 =A0 =A0 =A0: 35
>       >=A0 =A0 =A0 =A0Date=A0 =A0 =A0 =A0 =A0 =A0 : 2022-01-26
>       >
>       > Abstract:
>       >=A0 =A0CUBIC is a standard TCP congestion control algorithm that =
uses a
>       >=A0 =A0cubic function instead of a linear congestion window incr=
ease
>       >=A0 =A0function to improve scalability and stability over fast a=
nd long-
>       >=A0 =A0distance networks.=A0 CUBIC has been adopted as the defau=
lt TCP
>       >=A0 =A0congestion control algorithm by the Linux, Windows, and A=
pple stacks.
>       >
>       >=A0 =A0This document updates the specification of CUBIC to inclu=
de
>       >=A0 =A0algorithmic improvements based on these implementations a=
nd recent
>       >=A0 =A0academic work.=A0 Based on the extensive deployment exper=
ience with
>       >=A0 =A0CUBIC, it also moves the specification to the Standards T=
rack,
>       >=A0 =A0obsoleting RFC 8312.=A0 This also requires updating RFC 5=
681, to allow
>       >=A0 =A0for CUBIC's occasionally more aggressive sending behavior=
.
>       >
>       >
>       > The IETF datatracker status page for this draft is:
>       > https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/
>       >
>       > There is also an HTML version available at:
>       > https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.h=
tml
>       >
>       > A diff from the previous version is available at:
>       > https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-=
06
>       >
>       >
>       > Internet-Drafts are also available by rsync at rsync.ietf.org::=
internet-drafts
>       >
>       >
>       > _______________________________________________
>       > tcpm mailing list
>       > tcpm@ietf.org
>       > https://www.ietf.org/mailman/listinfo/tcpm
>
>       _______________________________________________
>       tcpm mailing list
>       tcpm@ietf.org
>       https://www.ietf.org/mailman/listinfo/tcpm
>=20
>=20
>
--=_script-23958-1644644065-0001-2--


From nobody Tue Feb 15 23:43:19 2022
Return-Path: <nsd.ietf@gmail.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E0FF63A0E77; Tue, 15 Feb 2022 23:43:15 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.097
X-Spam-Level: 
X-Spam-Status: No, score=-2.097 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ju8s-sg3wpob; Tue, 15 Feb 2022 23:43:10 -0800 (PST)
Received: from mail-qt1-x829.google.com (mail-qt1-x829.google.com [IPv6:2607:f8b0:4864:20::829]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B3C013A0E69; Tue, 15 Feb 2022 23:43:10 -0800 (PST)
Received: by mail-qt1-x829.google.com with SMTP id q20so1325621qtw.8; Tue, 15 Feb 2022 23:43:10 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112;  h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=196VElSHLu53wtn0KWlbc9fi0bOR0aAYt6PXwu9Q4oM=; b=eglRFgyL/4Tdg5W7/IcmI85kW4n4CqPlN9Gggfd76qKFEAwPWPqSpB5tf9CIux2+a9 fUVTTLDUwwQTFXzTYt/Tg1J1liKrkhC7+xmdsjK6SvowDgkUK6aWBKGrL1Xha8KA68uV K6m7+n1BLAImYRkZ8wlk1bXyZntMm/KnWqypXInV0KyVr2dRyAknOQGLAODweR/DOo58 ktg5Ancr6d4Fl9bgg/NyV9c4pf7DIyycE+L6g0HNQpToOzAKUDXYGE+OpKIUZJ5d3th1 QWO7rd+bFMApSTDU7blvnCW5LDO/k2b6Mxa+ZWpntzhBPtYd74nNYGAK8fPXYIalx5Wb bqCA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=196VElSHLu53wtn0KWlbc9fi0bOR0aAYt6PXwu9Q4oM=; b=2V4F9I952amyqV92vGuXUhQDAciqk8HlybBR3Px5t0LtBJKJi7JVUy0BflbNBCpZWE YvVg/N1TXn4FXA7Vfgr8qjtP7RyiAdgfoq6blDpK7VXmdSAU8y5f1YR808LtAzdZ/J/p UvaFAeUjK8s2q68her0Xze7bnp4gp99wG5mXxWRTKp/czwhRGNIcoVp6MArQql59CI8z qqwO4/HNbidR0IuDh1VGm14ZToJioaKV5d9qkwjnJJCFbWRAz7sMMoKeobhZicby0it/ tilLeThu2L6YpFpvTe1QLC8M9jsKLnjRRRQszUzdQliQiuh7+h6mOhX5R4wPfCt/rwzE tKuA==
X-Gm-Message-State: AOAM533XuJqQZGvDMMcJC6VeNipviHqKRbDoScHRuJPyFnHqaugwT9lg nXxzC7Ka5XRsE3041RORnTmZ4QHw3kP0a7Ke5adw2WzY
X-Google-Smtp-Source: ABdhPJwzTwXLY2uMiop3LEyr0TngwxVnVgqpkDgamNGtY6nLzqjyUDRNNMHXbdo7OSHD/pjMqfSxIJPP7GjGWhFTk/c=
X-Received: by 2002:ac8:5b91:0:b0:2db:acb7:2412 with SMTP id a17-20020ac85b91000000b002dbacb72412mr1090022qta.475.1644997388634; Tue, 15 Feb 2022 23:43:08 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi>
In-Reply-To: <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi>
From: Yoshifumi Nishida <nsd.ietf@gmail.com>
Date: Tue, 15 Feb 2022 23:42:57 -0800
Message-ID: <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com>
To: Markku Kojo <kojo@cs.helsinki.fi>
Cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: multipart/alternative; boundary="000000000000018d5b05d81dcdfb"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/4o5f7xLM4QqNvixd7hpM63usm00>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Feb 2022 07:43:16 -0000

--000000000000018d5b05d81dcdfb
Content-Type: text/plain; charset="UTF-8"

Hi Markku,

Thanks for the comments. I think these are very valid points.
However, I would like to check several things as a co-chair and a doc
shepherd before we discuss the points you've raised.

In my understanding (please correct me if I'm wrong), when this draft was
adopted as an WG item, I think the goal of the doc was some minor updates
from RFC8312 which include more clarifications, minor changes and bug
fixes.
However, if we try to address your concerns, I think we'll need to invent a
new version of CUBIC something like CUBIC++ or NewCUBIC in the end.
I won't deny the value of such doc, but, this seems not to be what we
agreed on beforehand.
if we proceed in this direction, I think we will need to check the WG
consensus whether this should be a new goal for the doc.

So, I would like to check if this is what you intend for the doc or you
think we can address your points while aligning with the original goal.
Also, if someone has opinions on this, please share.

Thanks,
--
Yoshi


On Fri, Feb 11, 2022 at 9:34 PM Markku Kojo <kojo@cs.helsinki.fi> wrote:

> Hi Yoshi, all,
>
> It seems to me that many issues that I raised have been solved, thanks.
> However, there are still a number of important issues that have not been
> addressed adequately. At least the following:
>
> #135 on W_max. Yoshi's observation was correct that this is not resolved:
> the co-authors and original developpers of CUBIC (@lisongxu and
> @sangtaeha) agreed in their last message that Wmax needs different
> treatment for slow start and congestion avoidance and plan comprehensive
> (new) evaluation of it. This is obviously an open issue but the issue
> was closed?
>
> #85 (& #86 with basically same issue and these two were combined) This
> (#85) is about ECN but the major issue is on using the same MD
> factor in slow start and in congestion avoidance when using
> loss-based CC. This (#85) remained closed even though I provided a
> thorough explanation why it is wrong and against the original theory and
> design by Van Jacobson, against the congestion control principles (RFC
> 2914) and two co-authors agreed on this in their same last message to
> #135 when they agreed on Wmax needing rework. This is an important issue
> that the wg should consider very carefully because it is not only
> updating RFC 5681 but also in conflict with RFC 2914. How can
> tcpm (and IETF) suggest and allow one CC algo to not follow congestion
> control principles as set in RFC 2914 while requiring all other CCs to
> follow RFC 2914 guidelines?
> The current draft does not provide any justification for using the same
> MD factor in slow start as in congestion avoidance. Nor am I
> awere of any experimental data that would support this change.
> The fact thet CUBIC has been long deployed does not alone provide any
> supporting evidence because CUBIC is likely to give good performance as
> it is overagressive and thereby unfair to competing traffic and users
> tend to be happy when measuring the performance of the sending CUBIC
> only, not the competing traffic that is badly impacted. HyStart++ is
> suggested as mitigation to the problem but it cannot; HyStart++ is only
> applicable during initial slow start, not during slow start after RTO!
> That is, the "SHOULD use HYStart++" text in Sec 4.10 is impossible
> to implement as I have pointed out in my comments earlier. Using a proper
> MD factor in slow start is even more important if loss is detectected
> during a RTO recovery because the sender is likely to face heavy
> congestion in such a case and it is very bad if the sender continues
> sending with overaggressive rate, stealing the capacity from and causing
> harm to coexisting flows. In addition, as I have explained, HyStart++
> does not remove the problem even for the initial slow start as it is not
> shown to work always. Instead, the results with the HyStart++ draft show
> that it reduces 50% of rexmits and only 36% RTOs, meaning that there is
> likely to be a notable percentage of cases when a sender is still
> in slow start when first loss is detected (i.e., HyStart++ had no effect)
> and a significant number of cases where a CUBIC sender is
> overaggressive continuing with a 1-40% larger cwnd than what is the
> available capacity. Note also that any delay-based heuristics like
> HyStart++ are known to work poorly in various wireless environmens where
> link delay tends to vary a lot. We may come up with some other MD factor
> that 0.5 when in slow start and HyStart++ is in use, but that is
> experimental, if not research, and definitely not ready for stds track.
>
> #114, #132, and #143 w.r.t flightsize vs. cwnd. The current text
>       does not quite correcly reflect what stacks that use cwnd do.
>       I'll comment in #143 separately.
>
> #96 & #98: The text added does not address the problems raised which are
> also evidenced in the paper pointed by Bob in #96. Even though CUBIC has
> been modified a bit after the paper was published, it does not
> automatically mean that the problem has been shown resolved: experimental
> evidence is required but not provided. The fact that CUBIC does not
> change MD factor for fast convergence is the root of the problem
> evidenced in the paper and remains so in the algo specified in this
> draft. This is also a significant problem when competing with Reno CC
> because CUBIC behaves much more aggressive than Reno CC when there is
> sudden congestion and all competing flows must converge fast down to a
> small fraction of the current cwnd to be fair to each other. This again
> cannot be evidenced not to be a problem by long deployment experience
> unless experimental data that measures the impact on competing traffic is
> presented to back the claims. Adjusting just Wmax for fast convergence is
> not enough and is even likely to be ineffective because there tend to be
> several losses when sudden congestion is hit, and particularly if NewReno
> is in use the sender stays several RTTs in Fast Recovery being
> overaggressive and then possibly continues at the same rate in CA which
> is unlikely to reach evan close to Wmax before a new loss hits
> the sender again. That is, lower Wmax and lower additive increase factor
> do not compensate the use of larger MD factor when sudden congestion is
> encountered.
>
> #93 & #94 & (#89) Sec 5.3 still does not address any difficult
> environments, in particular buffer-bloated paths (nor does Sec 5.4).
> We need evidence (results) that show CUBIC is fair towards other CCs
> (Reno) also in such environments. Note that CUBIC's decision to leave
> Reno-friendly region is based on the size of cwnd which tends to be
> incorrect with buffer-bloated bottlenecks because with huge buffers the
> cwnd can be many times larger than what is actually needed to fully
> utilize the available network bit-rate. Therefore, Reno CC has no problem
> in fully utilizing such bottleneck links and CUBIC must stay in
> Reno-friendly region longer but it leaves it too early because the same C
> is used as with non-bloated environments. We lack experiments showing
> CUBIC follows the congestion control principles and is fair to current
> standard TCP CC; to my understanding no experiments with buffer-bloated
> bottlenecks are cited to back up the claims even though buffer bloat is
> very well known to be a common (difficult) environment in today's
> Internet.
>
> #90 The current text on applying undo (a response to detected false
> fastrexmit) does not provide correct result if someone implements it.
> I have explained the problems there in github but seem to have not
> replied to latest comments by Neal. I'll reply and try to explain more.
> Again one major problem here is that the draft suggest a new response
> algo for false fast rexmits but does not provide any experimental data to
> support it. Long deployment experience has been suggested as
> justification but again without any carefully evaluated experimental
> data and evidence there is no meat. The issue is important to solve but is
> not specific to CUBIC. Instead, it is general problem for all TCP CC
> variants. IMHO, this is not ready for standards track but deserves a draft
> of its own so that it can be carefully evaluated and discussed ion the
> list. AFAIK there has been no discussion on this on the tcpm list, so
> those probably interested and having experience are likely to be unaware
> that this is part of CUBIC draft.
>
> #88 The problem with correctness of the AIMD model and setting alpha for
> CUBIC requires further consideration. Bob provided an analysis that
> leaves things still open. It seems that I never had time to review and
> comment the analysis and clarify why the model does not work. I'll do that
> separately as it is important to ensure CUBIC behaves fairly as intended
> for the Reno-friendly region.
>
> Best regards,
>
> /Markku
>
> On Mon, 31 Jan 2022, Yoshifumi Nishida wrote:
>
> > Hello,
> >
> > After some discussions among chairs, we decided to run the 2nd WGLC on
> draft-ietf-tcpm-rfc8312bis in
> > consideration of the importance of the draft.
> > We'll be grateful if you could send your feedback to the ML. The WGLC
> runs until *Feb 11*.
> >
> > If interested, you can check in-depth past discussions in the following
> URL.
> > https://github.com/NTAP/rfc8312bis/
> >
> > Thank you so much!
> > --
> > tcpm co-chairs
> >
> >
> > On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <lars@eggert.org> wrote:
> >       Hi,
> >
> >       this -06 version rolls in all the changes requested during (and
> after) WGLC ended.
> >
> >       I'll leave it up to the chairs to decide if another WGLC is
> warranted or the document can
> >       progress as-is.
> >
> >       Thanks,
> >       Lars
> >
> >
> >       > On 2022-1-26, at 11:12, 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 TCP Maintenance and Minor
> Extensions WG of the IETF.
> >       >
> >       >        Title           : CUBIC for Fast and Long-Distance
> Networks
> >       >        Authors         : Lisong Xu
> >       >                          Sangtae Ha
> >       >                          Injong Rhee
> >       >                          Vidhi Goel
> >       >                          Lars Eggert
> >       >       Filename        : draft-ietf-tcpm-rfc8312bis-06.txt
> >       >       Pages           : 35
> >       >       Date            : 2022-01-26
> >       >
> >       > Abstract:
> >       >   CUBIC is a standard TCP congestion control algorithm that uses
> a
> >       >   cubic function instead of a linear congestion window increase
> >       >   function to improve scalability and stability over fast and
> long-
> >       >   distance networks.  CUBIC has been adopted as the default TCP
> >       >   congestion control algorithm by the Linux, Windows, and Apple
> stacks.
> >       >
> >       >   This document updates the specification of CUBIC to include
> >       >   algorithmic improvements based on these implementations and
> recent
> >       >   academic work.  Based on the extensive deployment experience
> with
> >       >   CUBIC, it also moves the specification to the Standards Track,
> >       >   obsoleting RFC 8312.  This also requires updating RFC 5681, to
> allow
> >       >   for CUBIC's occasionally more aggressive sending behavior.
> >       >
> >       >
> >       > The IETF datatracker status page for this draft is:
> >       > https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/
> >       >
> >       > There is also an HTML version available at:
> >       >
> https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html
> >       >
> >       > A diff from the previous version is available at:
> >       > https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rfc8312bis-06
> >       >
> >       >
> >       > Internet-Drafts are also available by rsync at rsync.ietf.org:
> :internet-drafts
> >       >
> >       >
> >       > _______________________________________________
> >       > tcpm mailing list
> >       > tcpm@ietf.org
> >       > https://www.ietf.org/mailman/listinfo/tcpm
> >
> >       _______________________________________________
> >       tcpm mailing list
> >       tcpm@ietf.org
> >       https://www.ietf.org/mailman/listinfo/tcpm
> >
> >
> >

--000000000000018d5b05d81dcdfb
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi=C2=A0Markku,</div><div><br></div><div>Thanks for t=
he comments. I think these are very valid points.=C2=A0</div><div>However, =
I would like to check several things as a co-chair and a doc shepherd befor=
e we discuss=C2=A0the points you&#39;ve=C2=A0raised.</div><div><br></div><d=
iv>In my understanding (please correct me if I&#39;m wrong), when this draf=
t was adopted as an WG item, I think the goal of the doc was some minor upd=
ates from RFC8312 which include more clarifications, minor=C2=A0changes and=
 bug fixes.=C2=A0</div><div>However, if we try to address your=C2=A0concern=
s, I think we&#39;ll need to invent a new version of CUBIC something like C=
UBIC++ or NewCUBIC in the end.=C2=A0</div><div>I won&#39;t deny the value o=
f such doc, but, this seems not to be what we agreed on beforehand.=C2=A0=
=C2=A0</div><div>if we proceed in this direction, I think we will need to c=
heck the WG consensus whether this should be a new goal for the doc.</div><=
div><br></div><div>So, I would like to check if this is what you intend for=
 the doc or you think we can address your points while aligning with the or=
iginal goal.</div><div>Also, if someone has opinions on this, please share.=
</div><div><br></div><div>Thanks,</div><div>--</div><div>Yoshi</div><div><b=
r></div><br><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr=
">On Fri, Feb 11, 2022 at 9:34 PM Markku Kojo &lt;<a href=3D"mailto:kojo@cs=
.helsinki.fi">kojo@cs.helsinki.fi</a>&gt; wrote:<br></div><blockquote class=
=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rg=
b(204,204,204);padding-left:1ex">Hi Yoshi, all,<br>
<br>
It seems to me that many issues that I raised have been solved, thanks. <br=
>
However, there are still a number of important issues that have not been <b=
r>
addressed adequately. At least the following:<br>
<br>
#135 on W_max. Yoshi&#39;s observation was correct that this is not resolve=
d: <br>
the co-authors and original developpers of CUBIC (@lisongxu and <br>
@sangtaeha) agreed in their last message that Wmax needs different <br>
treatment for slow start and congestion avoidance and plan comprehensive <b=
r>
(new) evaluation of it. This is obviously an open issue but the issue <br>
was closed?<br>
<br>
#85 (&amp; #86 with basically same issue and these two were combined) This =
<br>
(#85) is about ECN but the major issue is on using the same MD <br>
factor in slow start and in congestion avoidance when using <br>
loss-based CC. This (#85) remained closed even though I provided a <br>
thorough explanation why it is wrong and against the original theory and <b=
r>
design by Van Jacobson, against the congestion control principles (RFC <br>
2914) and two co-authors agreed on this in their same last message to <br>
#135 when they agreed on Wmax needing rework. This is an important issue <b=
r>
that the wg should consider very carefully because it is not only <br>
updating RFC 5681 but also in conflict with RFC 2914. How can <br>
tcpm (and IETF) suggest and allow one CC algo to not follow congestion <br>
control principles as set in RFC 2914 while requiring all other CCs to <br>
follow RFC 2914 guidelines?<br>
The current draft does not provide any justification for using the same <br=
>
MD factor in slow start as in congestion avoidance. Nor am I <br>
awere of any experimental data that would support this change. <br>
The fact thet CUBIC has been long deployed does not alone provide any <br>
supporting evidence because CUBIC is likely to give good performance as <br=
>
it is overagressive and thereby unfair to competing traffic and users <br>
tend to be happy when measuring the performance of the sending CUBIC <br>
only, not the competing traffic that is badly impacted. HyStart++ is <br>
suggested as mitigation to the problem but it cannot; HyStart++ is only <br=
>
applicable during initial slow start, not during slow start after RTO! <br>
That is, the &quot;SHOULD use HYStart++&quot; text in Sec 4.10 is impossibl=
e <br>
to implement as I have pointed out in my comments earlier. Using a proper <=
br>
MD factor in slow start is even more important if loss is detectected <br>
during a RTO recovery because the sender is likely to face heavy <br>
congestion in such a case and it is very bad if the sender continues <br>
sending with overaggressive rate, stealing the capacity from and causing <b=
r>
harm to coexisting flows. In addition, as I have explained, HyStart++ <br>
does not remove the problem even for the initial slow start as it is not <b=
r>
shown to work always. Instead, the results with the HyStart++ draft show <b=
r>
that it reduces 50% of rexmits and only 36% RTOs, meaning that there is <br=
>
likely to be a notable percentage of cases when a sender is still <br>
in slow start when first loss is detected (i.e., HyStart++ had no effect) <=
br>
and a significant number of cases where a CUBIC sender is <br>
overaggressive continuing with a 1-40% larger cwnd than what is the <br>
available capacity. Note also that any delay-based heuristics like <br>
HyStart++ are known to work poorly in various wireless environmens where <b=
r>
link delay tends to vary a lot. We may come up with some other MD factor <b=
r>
that 0.5 when in slow start and HyStart++ is in use, but that is <br>
experimental, if not research, and definitely not ready for stds track.<br>
<br>
#114, #132, and #143 w.r.t flightsize vs. cwnd. The current text<br>
=C2=A0 =C2=A0 =C2=A0 does not quite correcly reflect what stacks that use c=
wnd do.<br>
=C2=A0 =C2=A0 =C2=A0 I&#39;ll comment in #143 separately.<br>
<br>
#96 &amp; #98: The text added does not address the problems raised which ar=
e <br>
also evidenced in the paper pointed by Bob in #96. Even though CUBIC has <b=
r>
been modified a bit after the paper was published, it does not <br>
automatically mean that the problem has been shown resolved: experimental <=
br>
evidence is required but not provided. The fact that CUBIC does not <br>
change MD factor for fast convergence is the root of the problem <br>
evidenced in the paper and remains so in the algo specified in this <br>
draft. This is also a significant problem when competing with Reno CC <br>
because CUBIC behaves much more aggressive than Reno CC when there is <br>
sudden congestion and all competing flows must converge fast down to a <br>
small fraction of the current cwnd to be fair to each other. This again <br=
>
cannot be evidenced not to be a problem by long deployment experience <br>
unless experimental data that measures the impact on competing traffic is <=
br>
presented to back the claims. Adjusting just Wmax for fast convergence is <=
br>
not enough and is even likely to be ineffective because there tend to be <b=
r>
several losses when sudden congestion is hit, and particularly if NewReno <=
br>
is in use the sender stays several RTTs in Fast Recovery being <br>
overaggressive and then possibly continues at the same rate in CA which <br=
>
is unlikely to reach evan close to Wmax before a new loss hits <br>
the sender again. That is, lower Wmax and lower additive increase factor <b=
r>
do not compensate the use of larger MD factor when sudden congestion is <br=
>
encountered.<br>
<br>
#93 &amp; #94 &amp; (#89) Sec 5.3 still does not address any difficult <br>
environments, in particular buffer-bloated paths (nor does Sec 5.4).<br>
We need evidence (results) that show CUBIC is fair towards other CCs <br>
(Reno) also in such environments. Note that CUBIC&#39;s decision to leave <=
br>
Reno-friendly region is based on the size of cwnd which tends to be <br>
incorrect with buffer-bloated bottlenecks because with huge buffers the <br=
>
cwnd can be many times larger than what is actually needed to fully <br>
utilize the available network bit-rate. Therefore, Reno CC has no problem <=
br>
in fully utilizing such bottleneck links and CUBIC must stay in <br>
Reno-friendly region longer but it leaves it too early because the same C <=
br>
is used as with non-bloated environments. We lack experiments showing <br>
CUBIC follows the congestion control principles and is fair to current <br>
standard TCP CC; to my understanding no experiments with buffer-bloated<br>
bottlenecks are cited to back up the claims even though buffer bloat is <br=
>
very well known to be a common (difficult) environment in today&#39;s <br>
Internet.<br>
<br>
#90 The current text on applying undo (a response to detected false <br>
fastrexmit) does not provide correct result if someone implements it.<br>
I have explained the problems there in github but seem to have not <br>
replied to latest comments by Neal. I&#39;ll reply and try to explain more.=
<br>
Again one major problem here is that the draft suggest a new response <br>
algo for false fast rexmits but does not provide any experimental data to <=
br>
support it. Long deployment experience has been suggested as <br>
justification but again without any carefully evaluated experimental <br>
data and evidence there is no meat. The issue is important to solve but is =
<br>
not specific to CUBIC. Instead, it is general problem for all TCP CC <br>
variants. IMHO, this is not ready for standards track but deserves a draft =
<br>
of its own so that it can be carefully evaluated and discussed ion the <br>
list. AFAIK there has been no discussion on this on the tcpm list, so <br>
those probably interested and having experience are likely to be unaware <b=
r>
that this is part of CUBIC draft.<br>
<br>
#88 The problem with correctness of the AIMD model and setting alpha for <b=
r>
CUBIC requires further consideration. Bob provided an analysis that <br>
leaves things still open. It seems that I never had time to review and <br>
comment the analysis and clarify why the model does not work. I&#39;ll do t=
hat <br>
separately as it is important to ensure CUBIC behaves fairly as intended <b=
r>
for the Reno-friendly region.<br>
<br>
Best regards,<br>
<br>
/Markku<br>
<br>
On Mon, 31 Jan 2022, Yoshifumi Nishida wrote:<br>
<br>
&gt; Hello,<br>
&gt; <br>
&gt; After some discussions among chairs, we decided to run the 2nd WGLC on=
 draft-ietf-tcpm-rfc8312bis in<br>
&gt; consideration of the importance of the draft.=C2=A0<br>
&gt; We&#39;ll be grateful if you could send your feedback to the ML. The W=
GLC runs until *Feb 11*.<br>
&gt; <br>
&gt; If interested, you can check in-depth past discussions in the followin=
g URL.<br>
&gt; <a href=3D"https://github.com/NTAP/rfc8312bis/" rel=3D"noreferrer" tar=
get=3D"_blank">https://github.com/NTAP/rfc8312bis/</a><br>
&gt; <br>
&gt; Thank you so much!<br>
&gt; --<br>
&gt; tcpm co-chairs<br>
&gt; <br>
&gt; <br>
&gt; On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert &lt;<a href=3D"mailto:lars=
@eggert.org" target=3D"_blank">lars@eggert.org</a>&gt; wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Hi,<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0this -06 version rolls in all the changes re=
quested during (and after) WGLC ended.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0I&#39;ll leave it up to the chairs to decide=
 if another WGLC is warranted or the document can<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0progress as-is.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Thanks,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Lars<br>
&gt; <br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; On 2022-1-26, at 11:12, <a href=3D"mail=
to:internet-drafts@ietf.org" target=3D"_blank">internet-drafts@ietf.org</a>=
 wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; A New Internet-Draft is available from =
the on-line Internet-Drafts directories.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; This draft is a work item of the TCP Ma=
intenance and Minor Extensions WG of the IETF.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Title=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: CUBIC for Fast and Long-Distance Networ=
ks<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Authors=C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: Lisong Xu<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Sangtae Ha<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Injong Rhee<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Vidhi Goel<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Lars Eggert<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Filename=C2=
=A0 =C2=A0 =C2=A0 =C2=A0 : draft-ietf-tcpm-rfc8312bis-06.txt<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Pages=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: 35<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Date=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 : 2022-01-26<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Abstract:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0CUBIC is a standard TCP con=
gestion control algorithm that uses a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0cubic function instead of a=
 linear congestion window increase<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0function to improve scalabi=
lity and stability over fast and long-<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0distance networks.=C2=A0 CU=
BIC has been adopted as the default TCP<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0congestion control algorith=
m by the Linux, Windows, and Apple stacks.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0This document updates the s=
pecification of CUBIC to include<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0algorithmic improvements ba=
sed on these implementations and recent<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0academic work.=C2=A0 Based =
on the extensive deployment experience with<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0CUBIC, it also moves the sp=
ecification to the Standards Track,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0obsoleting RFC 8312.=C2=A0 =
This also requires updating RFC 5681, to allow<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0for CUBIC&#39;s occasionall=
y more aggressive sending behavior.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; The IETF datatracker status page for th=
is draft is:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://datatracker.ietf.org=
/doc/draft-ietf-tcpm-rfc8312bis/" rel=3D"noreferrer" target=3D"_blank">http=
s://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; There is also an HTML version available=
 at:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://www.ietf.org/archive=
/id/draft-ietf-tcpm-rfc8312bis-06.html" rel=3D"noreferrer" target=3D"_blank=
">https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html</a><br=
>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; A diff from the previous version is ava=
ilable at:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://www.ietf.org/rfcdiff=
?url2=3Ddraft-ietf-tcpm-rfc8312bis-06" rel=3D"noreferrer" target=3D"_blank"=
>https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Internet-Drafts are also available by r=
sync at rsync.ietf.org::internet-drafts<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; _______________________________________=
________<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; tcpm mailing list<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"mailto:tcpm@ietf.org" target=
=3D"_blank">tcpm@ietf.org</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://www.ietf.org/mailman=
/listinfo/tcpm" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/m=
ailman/listinfo/tcpm</a><br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0____________________________________________=
___<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0tcpm mailing list<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"mailto:tcpm@ietf.org" target=3D"_=
blank">tcpm@ietf.org</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"https://www.ietf.org/mailman/list=
info/tcpm" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/mailma=
n/listinfo/tcpm</a><br>
&gt; <br>
&gt; <br>
&gt;</blockquote></div></div>

--000000000000018d5b05d81dcdfb--


From nobody Wed Feb 16 14:42:22 2022
Return-Path: <iesg-secretary@ietf.org>
X-Original-To: tcpm@ietf.org
Delivered-To: tcpm@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id D57D43A1769; Wed, 16 Feb 2022 14:42:19 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: The IESG <iesg-secretary@ietf.org>
To: "IETF-Announce" <ietf-announce@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 7.45.0
Auto-Submitted: auto-generated
Precedence: bulk
CC: draft-ietf-tcpm-yang-tcp@ietf.org, martin.h.duke@gmail.com, nsd.ietf@gmail.com, tcpm-chairs@ietf.org, tcpm@ietf.org
Reply-To: last-call@ietf.org
Sender: <iesg-secretary@ietf.org>
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
Message-ID: <164505133983.5800.270812007978080248@ietfa.amsl.com>
Date: Wed, 16 Feb 2022 14:42:19 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/4yEjDIXaRwzndDnYSezqdeWsvY0>
Subject: [tcpm] Last Call: <draft-ietf-tcpm-yang-tcp-06.txt> (A YANG Model for Transmission Control Protocol (TCP) Configuration) to Proposed Standard
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Feb 2022 22:42:20 -0000

The IESG has received a request from the TCP Maintenance and Minor Extensions
WG (tcpm) to consider the following document: - 'A YANG Model for
Transmission Control Protocol (TCP) Configuration'
  <draft-ietf-tcpm-yang-tcp-06.txt> as Proposed Standard

The IESG plans to make a decision in the next few weeks, and solicits final
comments on this action. Please send substantive comments to the
last-call@ietf.org mailing lists by 2022-03-02. Exceptionally, comments may
be sent to iesg@ietf.org instead. In either case, please retain the beginning
of the Subject line to allow automated sorting.

Abstract


   This document specifies a minimal YANG model for TCP on devices that
   are configured by network management protocols.  The YANG model
   defines a container for all TCP connections and groupings of
   authentication parameters that can be imported and used in TCP
   implementations or by other models that need to configure TCP
   parameters.  The model also includes basic TCP statistics.  The model
   is compliant with Network Management Datastore Architecture (NMDA)
   (RFC 8342).




The file can be obtained via
https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/



No IPR declarations have been submitted directly on this I-D.






From nobody Wed Feb 16 15:40:46 2022
Return-Path: <ycheng@google.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DCE513A17FE for <tcpm@ietfa.amsl.com>; Wed, 16 Feb 2022 15:40:42 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -17.6
X-Spam-Level: 
X-Spam-Status: No, score=-17.6 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_MED=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, ENV_AND_HDR_SPF_MATCH=-0.5, HTML_MESSAGE=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5, USER_IN_DEF_SPF_WL=-7.5] autolearn=unavailable autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=google.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id A9bEvJuykO2y for <tcpm@ietfa.amsl.com>; Wed, 16 Feb 2022 15:40:38 -0800 (PST)
Received: from mail-wm1-x332.google.com (mail-wm1-x332.google.com [IPv6:2a00:1450:4864:20::332]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CF2B93A17FF for <tcpm@ietf.org>; Wed, 16 Feb 2022 15:40:37 -0800 (PST)
Received: by mail-wm1-x332.google.com with SMTP id n8so2070415wms.3 for <tcpm@ietf.org>; Wed, 16 Feb 2022 15:40:37 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=0NgP3GATlT9833Vsw0uoND+JEXZKyxz4MqTe8Wg/9Is=; b=iNNn++rFgVsf6OfGMlvA1Qty2Y79vAXPbkfIGEXISATsZP64DOSNUTBCwRkQKD1UFp ruvYEYOu9UhxXs+kB0KJou0D3qC/po3emmr51DcaoqMBMi3RAr99k/QFmOy/BUX5DIw+ R7tx9oSUCFMB5MQXrKKApdvREe5zcQ0S4cN14qrsucxIk0xVD9V+bjyf64cfs4eKIEcN 2i/uWqKztHWGph6tf3USjrUynI18SzQG0eqr7OIIBJJE9vSetkgXTtBL9YoS3jlT1iX1 th4eaPW2g83wjJLYyBS0dNCM0EOj2GhH0NiPrsoWi8CB+sJr+sW8npSKK4yB7nIed31f /WFQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=0NgP3GATlT9833Vsw0uoND+JEXZKyxz4MqTe8Wg/9Is=; b=DWZvUmXTBF8Za2UTfZWdBTaXRSH5j8kIqN/L09AcQRayJ/XOq49TqGC4v3wAsLbgFb wdrA1Iumv/C094gDUeUsSfzzoHL9TEzxbNA9u1Se55waDYOnMY6VzL/gRYmN2UVQ1AJo CotMhAWxCvvQ2BBntZSgKPWdAI9Miq/XVg9erx/r2VekqijwpMGs2NhvjYUjvdYMDexQ 0VvkJW16FWnOZdeE4pMBq08sZ00X7ssu9hvpdFdPemL9c0H3CL2csKzaQrYTi/x57SNY X0aKtXucQrdUW9IcInM//eDMI3JAL5WOxwsZXD2OhY+fNvBtChHEDiSkPSvnHvM1pY8M eMPw==
X-Gm-Message-State: AOAM533RcvpRVsdatUVjKdfyhoaRFUd6TZfMWQvEWmGXStSyHzu/ekXE swqLnB6D7L6r/hOlicEGb3BryHeP44+LPryhy5yWD78E/qkOCA==
X-Google-Smtp-Source: ABdhPJzWBYq8XX3uCrIZQymoM6GB2oMepOVTQ+Z4e54VNiUhxhIIZXVMOTZrWL5LaatqJrMV4xuX1hmTnUvcrnon6P0=
X-Received: by 2002:a1c:f605:0:b0:37b:b5de:89a0 with SMTP id w5-20020a1cf605000000b0037bb5de89a0mr259264wmc.88.1645054835525; Wed, 16 Feb 2022 15:40:35 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com>
In-Reply-To: <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com>
From: Yuchung Cheng <ycheng@google.com>
Date: Wed, 16 Feb 2022 15:39:58 -0800
Message-ID: <CAK6E8=cWgLVi6TrBbvYr+2Jad5ZraC6iL=P=8jsT8wVtiLYCig@mail.gmail.com>
To: Yoshifumi Nishida <nsd.ietf@gmail.com>
Cc: Markku Kojo <kojo@cs.helsinki.fi>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>,  tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: multipart/alternative; boundary="0000000000001be2ff05d82b2d7a"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/YIvjsfdW86zdrKCm0-dwkyJeZvM>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Feb 2022 23:40:43 -0000

--0000000000001be2ff05d82b2d7a
Content-Type: text/plain; charset="UTF-8"

+1 to what Yoshifumi said for a minor update.

On Tue, Feb 15, 2022 at 11:43 PM Yoshifumi Nishida <nsd.ietf@gmail.com>
wrote:

> Hi Markku,
>
> Thanks for the comments. I think these are very valid points.
> However, I would like to check several things as a co-chair and a doc
> shepherd before we discuss the points you've raised.
>
> In my understanding (please correct me if I'm wrong), when this draft was
> adopted as an WG item, I think the goal of the doc was some minor updates
> from RFC8312 which include more clarifications, minor changes and bug
> fixes.
> However, if we try to address your concerns, I think we'll need to invent
> a new version of CUBIC something like CUBIC++ or NewCUBIC in the end.
> I won't deny the value of such doc, but, this seems not to be what we
> agreed on beforehand.
> if we proceed in this direction, I think we will need to check the WG
> consensus whether this should be a new goal for the doc.
>
> So, I would like to check if this is what you intend for the doc or you
> think we can address your points while aligning with the original goal.
> Also, if someone has opinions on this, please share.
>
> Thanks,
> --
> Yoshi
>
>
> On Fri, Feb 11, 2022 at 9:34 PM Markku Kojo <kojo@cs.helsinki.fi> wrote:
>
>> Hi Yoshi, all,
>>
>> It seems to me that many issues that I raised have been solved, thanks.
>> However, there are still a number of important issues that have not been
>> addressed adequately. At least the following:
>>
>> #135 on W_max. Yoshi's observation was correct that this is not resolved:
>> the co-authors and original developpers of CUBIC (@lisongxu and
>> @sangtaeha) agreed in their last message that Wmax needs different
>> treatment for slow start and congestion avoidance and plan comprehensive
>> (new) evaluation of it. This is obviously an open issue but the issue
>> was closed?
>>
>> #85 (& #86 with basically same issue and these two were combined) This
>> (#85) is about ECN but the major issue is on using the same MD
>> factor in slow start and in congestion avoidance when using
>> loss-based CC. This (#85) remained closed even though I provided a
>> thorough explanation why it is wrong and against the original theory and
>> design by Van Jacobson, against the congestion control principles (RFC
>> 2914) and two co-authors agreed on this in their same last message to
>> #135 when they agreed on Wmax needing rework. This is an important issue
>> that the wg should consider very carefully because it is not only
>> updating RFC 5681 but also in conflict with RFC 2914. How can
>> tcpm (and IETF) suggest and allow one CC algo to not follow congestion
>> control principles as set in RFC 2914 while requiring all other CCs to
>> follow RFC 2914 guidelines?
>> The current draft does not provide any justification for using the same
>> MD factor in slow start as in congestion avoidance. Nor am I
>> awere of any experimental data that would support this change.
>> The fact thet CUBIC has been long deployed does not alone provide any
>> supporting evidence because CUBIC is likely to give good performance as
>> it is overagressive and thereby unfair to competing traffic and users
>> tend to be happy when measuring the performance of the sending CUBIC
>> only, not the competing traffic that is badly impacted. HyStart++ is
>> suggested as mitigation to the problem but it cannot; HyStart++ is only
>> applicable during initial slow start, not during slow start after RTO!
>> That is, the "SHOULD use HYStart++" text in Sec 4.10 is impossible
>> to implement as I have pointed out in my comments earlier. Using a proper
>> MD factor in slow start is even more important if loss is detectected
>> during a RTO recovery because the sender is likely to face heavy
>> congestion in such a case and it is very bad if the sender continues
>> sending with overaggressive rate, stealing the capacity from and causing
>> harm to coexisting flows. In addition, as I have explained, HyStart++
>> does not remove the problem even for the initial slow start as it is not
>> shown to work always. Instead, the results with the HyStart++ draft show
>> that it reduces 50% of rexmits and only 36% RTOs, meaning that there is
>> likely to be a notable percentage of cases when a sender is still
>> in slow start when first loss is detected (i.e., HyStart++ had no effect)
>> and a significant number of cases where a CUBIC sender is
>> overaggressive continuing with a 1-40% larger cwnd than what is the
>> available capacity. Note also that any delay-based heuristics like
>> HyStart++ are known to work poorly in various wireless environmens where
>> link delay tends to vary a lot. We may come up with some other MD factor
>> that 0.5 when in slow start and HyStart++ is in use, but that is
>> experimental, if not research, and definitely not ready for stds track.
>>
>> #114, #132, and #143 w.r.t flightsize vs. cwnd. The current text
>>       does not quite correcly reflect what stacks that use cwnd do.
>>       I'll comment in #143 separately.
>>
>> #96 & #98: The text added does not address the problems raised which are
>> also evidenced in the paper pointed by Bob in #96. Even though CUBIC has
>> been modified a bit after the paper was published, it does not
>> automatically mean that the problem has been shown resolved: experimental
>> evidence is required but not provided. The fact that CUBIC does not
>> change MD factor for fast convergence is the root of the problem
>> evidenced in the paper and remains so in the algo specified in this
>> draft. This is also a significant problem when competing with Reno CC
>> because CUBIC behaves much more aggressive than Reno CC when there is
>> sudden congestion and all competing flows must converge fast down to a
>> small fraction of the current cwnd to be fair to each other. This again
>> cannot be evidenced not to be a problem by long deployment experience
>> unless experimental data that measures the impact on competing traffic is
>> presented to back the claims. Adjusting just Wmax for fast convergence is
>> not enough and is even likely to be ineffective because there tend to be
>> several losses when sudden congestion is hit, and particularly if NewReno
>> is in use the sender stays several RTTs in Fast Recovery being
>> overaggressive and then possibly continues at the same rate in CA which
>> is unlikely to reach evan close to Wmax before a new loss hits
>> the sender again. That is, lower Wmax and lower additive increase factor
>> do not compensate the use of larger MD factor when sudden congestion is
>> encountered.
>>
>> #93 & #94 & (#89) Sec 5.3 still does not address any difficult
>> environments, in particular buffer-bloated paths (nor does Sec 5.4).
>> We need evidence (results) that show CUBIC is fair towards other CCs
>> (Reno) also in such environments. Note that CUBIC's decision to leave
>> Reno-friendly region is based on the size of cwnd which tends to be
>> incorrect with buffer-bloated bottlenecks because with huge buffers the
>> cwnd can be many times larger than what is actually needed to fully
>> utilize the available network bit-rate. Therefore, Reno CC has no problem
>> in fully utilizing such bottleneck links and CUBIC must stay in
>> Reno-friendly region longer but it leaves it too early because the same C
>> is used as with non-bloated environments. We lack experiments showing
>> CUBIC follows the congestion control principles and is fair to current
>> standard TCP CC; to my understanding no experiments with buffer-bloated
>> bottlenecks are cited to back up the claims even though buffer bloat is
>> very well known to be a common (difficult) environment in today's
>> Internet.
>>
>> #90 The current text on applying undo (a response to detected false
>> fastrexmit) does not provide correct result if someone implements it.
>> I have explained the problems there in github but seem to have not
>> replied to latest comments by Neal. I'll reply and try to explain more.
>> Again one major problem here is that the draft suggest a new response
>> algo for false fast rexmits but does not provide any experimental data to
>> support it. Long deployment experience has been suggested as
>> justification but again without any carefully evaluated experimental
>> data and evidence there is no meat. The issue is important to solve but
>> is
>> not specific to CUBIC. Instead, it is general problem for all TCP CC
>> variants. IMHO, this is not ready for standards track but deserves a
>> draft
>> of its own so that it can be carefully evaluated and discussed ion the
>> list. AFAIK there has been no discussion on this on the tcpm list, so
>> those probably interested and having experience are likely to be unaware
>> that this is part of CUBIC draft.
>>
>> #88 The problem with correctness of the AIMD model and setting alpha for
>> CUBIC requires further consideration. Bob provided an analysis that
>> leaves things still open. It seems that I never had time to review and
>> comment the analysis and clarify why the model does not work. I'll do
>> that
>> separately as it is important to ensure CUBIC behaves fairly as intended
>> for the Reno-friendly region.
>>
>> Best regards,
>>
>> /Markku
>>
>> On Mon, 31 Jan 2022, Yoshifumi Nishida wrote:
>>
>> > Hello,
>> >
>> > After some discussions among chairs, we decided to run the 2nd WGLC on
>> draft-ietf-tcpm-rfc8312bis in
>> > consideration of the importance of the draft.
>> > We'll be grateful if you could send your feedback to the ML. The WGLC
>> runs until *Feb 11*.
>> >
>> > If interested, you can check in-depth past discussions in the following
>> URL.
>> > https://github.com/NTAP/rfc8312bis/
>> >
>> > Thank you so much!
>> > --
>> > tcpm co-chairs
>> >
>> >
>> > On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <lars@eggert.org> wrote:
>> >       Hi,
>> >
>> >       this -06 version rolls in all the changes requested during (and
>> after) WGLC ended.
>> >
>> >       I'll leave it up to the chairs to decide if another WGLC is
>> warranted or the document can
>> >       progress as-is.
>> >
>> >       Thanks,
>> >       Lars
>> >
>> >
>> >       > On 2022-1-26, at 11:12, 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 TCP Maintenance and Minor
>> Extensions WG of the IETF.
>> >       >
>> >       >        Title           : CUBIC for Fast and Long-Distance
>> Networks
>> >       >        Authors         : Lisong Xu
>> >       >                          Sangtae Ha
>> >       >                          Injong Rhee
>> >       >                          Vidhi Goel
>> >       >                          Lars Eggert
>> >       >       Filename        : draft-ietf-tcpm-rfc8312bis-06.txt
>> >       >       Pages           : 35
>> >       >       Date            : 2022-01-26
>> >       >
>> >       > Abstract:
>> >       >   CUBIC is a standard TCP congestion control algorithm that
>> uses a
>> >       >   cubic function instead of a linear congestion window increase
>> >       >   function to improve scalability and stability over fast and
>> long-
>> >       >   distance networks.  CUBIC has been adopted as the default TCP
>> >       >   congestion control algorithm by the Linux, Windows, and Apple
>> stacks.
>> >       >
>> >       >   This document updates the specification of CUBIC to include
>> >       >   algorithmic improvements based on these implementations and
>> recent
>> >       >   academic work.  Based on the extensive deployment experience
>> with
>> >       >   CUBIC, it also moves the specification to the Standards Track,
>> >       >   obsoleting RFC 8312.  This also requires updating RFC 5681,
>> to allow
>> >       >   for CUBIC's occasionally more aggressive sending behavior.
>> >       >
>> >       >
>> >       > The IETF datatracker status page for this draft is:
>> >       > https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/
>> >       >
>> >       > There is also an HTML version available at:
>> >       >
>> https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html
>> >       >
>> >       > A diff from the previous version is available at:
>> >       > https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rfc8312bis-06
>> >       >
>> >       >
>> >       > Internet-Drafts are also available by rsync at rsync.ietf.org:
>> :internet-drafts
>> >       >
>> >       >
>> >       > _______________________________________________
>> >       > tcpm mailing list
>> >       > tcpm@ietf.org
>> >       > https://www.ietf.org/mailman/listinfo/tcpm
>> >
>> >       _______________________________________________
>> >       tcpm mailing list
>> >       tcpm@ietf.org
>> >       https://www.ietf.org/mailman/listinfo/tcpm
>> >
>> >
>> >
>
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm
>

--0000000000001be2ff05d82b2d7a
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">+1 to what Yoshifumi said for a minor update.<br></div><br=
><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Tue, F=
eb 15, 2022 at 11:43 PM Yoshifumi Nishida &lt;<a href=3D"mailto:nsd.ietf@gm=
ail.com">nsd.ietf@gmail.com</a>&gt; wrote:<br></div><blockquote class=3D"gm=
ail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,=
204,204);padding-left:1ex"><div dir=3D"ltr"><div>Hi=C2=A0Markku,</div><div>=
<br></div><div>Thanks for the comments. I think these are very valid points=
.=C2=A0</div><div>However, I would like to check several things as a co-cha=
ir and a doc shepherd before we discuss=C2=A0the points you&#39;ve=C2=A0rai=
sed.</div><div><br></div><div>In my understanding (please correct me if I&#=
39;m wrong), when this draft was adopted as an WG item, I think the goal of=
 the doc was some minor updates from RFC8312 which include more clarificati=
ons, minor=C2=A0changes and bug fixes.=C2=A0</div><div>However, if we try t=
o address your=C2=A0concerns, I think we&#39;ll need to invent a new versio=
n of CUBIC something like CUBIC++ or NewCUBIC in the end.=C2=A0</div><div>I=
 won&#39;t deny the value of such doc, but, this seems not to be what we ag=
reed on beforehand.=C2=A0=C2=A0</div><div>if we proceed in this direction, =
I think we will need to check the WG consensus whether this should be a new=
 goal for the doc.</div><div><br></div><div>So, I would like to check if th=
is is what you intend for the doc or you think we can address your points w=
hile aligning with the original goal.</div><div>Also, if someone has opinio=
ns on this, please share.</div><div><br></div><div>Thanks,</div><div>--</di=
v><div>Yoshi</div><div><br></div><br><div class=3D"gmail_quote"><div dir=3D=
"ltr" class=3D"gmail_attr">On Fri, Feb 11, 2022 at 9:34 PM Markku Kojo &lt;=
<a href=3D"mailto:kojo@cs.helsinki.fi" target=3D"_blank">kojo@cs.helsinki.f=
i</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin=
:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"=
>Hi Yoshi, all,<br>
<br>
It seems to me that many issues that I raised have been solved, thanks. <br=
>
However, there are still a number of important issues that have not been <b=
r>
addressed adequately. At least the following:<br>
<br>
#135 on W_max. Yoshi&#39;s observation was correct that this is not resolve=
d: <br>
the co-authors and original developpers of CUBIC (@lisongxu and <br>
@sangtaeha) agreed in their last message that Wmax needs different <br>
treatment for slow start and congestion avoidance and plan comprehensive <b=
r>
(new) evaluation of it. This is obviously an open issue but the issue <br>
was closed?<br>
<br>
#85 (&amp; #86 with basically same issue and these two were combined) This =
<br>
(#85) is about ECN but the major issue is on using the same MD <br>
factor in slow start and in congestion avoidance when using <br>
loss-based CC. This (#85) remained closed even though I provided a <br>
thorough explanation why it is wrong and against the original theory and <b=
r>
design by Van Jacobson, against the congestion control principles (RFC <br>
2914) and two co-authors agreed on this in their same last message to <br>
#135 when they agreed on Wmax needing rework. This is an important issue <b=
r>
that the wg should consider very carefully because it is not only <br>
updating RFC 5681 but also in conflict with RFC 2914. How can <br>
tcpm (and IETF) suggest and allow one CC algo to not follow congestion <br>
control principles as set in RFC 2914 while requiring all other CCs to <br>
follow RFC 2914 guidelines?<br>
The current draft does not provide any justification for using the same <br=
>
MD factor in slow start as in congestion avoidance. Nor am I <br>
awere of any experimental data that would support this change. <br>
The fact thet CUBIC has been long deployed does not alone provide any <br>
supporting evidence because CUBIC is likely to give good performance as <br=
>
it is overagressive and thereby unfair to competing traffic and users <br>
tend to be happy when measuring the performance of the sending CUBIC <br>
only, not the competing traffic that is badly impacted. HyStart++ is <br>
suggested as mitigation to the problem but it cannot; HyStart++ is only <br=
>
applicable during initial slow start, not during slow start after RTO! <br>
That is, the &quot;SHOULD use HYStart++&quot; text in Sec 4.10 is impossibl=
e <br>
to implement as I have pointed out in my comments earlier. Using a proper <=
br>
MD factor in slow start is even more important if loss is detectected <br>
during a RTO recovery because the sender is likely to face heavy <br>
congestion in such a case and it is very bad if the sender continues <br>
sending with overaggressive rate, stealing the capacity from and causing <b=
r>
harm to coexisting flows. In addition, as I have explained, HyStart++ <br>
does not remove the problem even for the initial slow start as it is not <b=
r>
shown to work always. Instead, the results with the HyStart++ draft show <b=
r>
that it reduces 50% of rexmits and only 36% RTOs, meaning that there is <br=
>
likely to be a notable percentage of cases when a sender is still <br>
in slow start when first loss is detected (i.e., HyStart++ had no effect) <=
br>
and a significant number of cases where a CUBIC sender is <br>
overaggressive continuing with a 1-40% larger cwnd than what is the <br>
available capacity. Note also that any delay-based heuristics like <br>
HyStart++ are known to work poorly in various wireless environmens where <b=
r>
link delay tends to vary a lot. We may come up with some other MD factor <b=
r>
that 0.5 when in slow start and HyStart++ is in use, but that is <br>
experimental, if not research, and definitely not ready for stds track.<br>
<br>
#114, #132, and #143 w.r.t flightsize vs. cwnd. The current text<br>
=C2=A0 =C2=A0 =C2=A0 does not quite correcly reflect what stacks that use c=
wnd do.<br>
=C2=A0 =C2=A0 =C2=A0 I&#39;ll comment in #143 separately.<br>
<br>
#96 &amp; #98: The text added does not address the problems raised which ar=
e <br>
also evidenced in the paper pointed by Bob in #96. Even though CUBIC has <b=
r>
been modified a bit after the paper was published, it does not <br>
automatically mean that the problem has been shown resolved: experimental <=
br>
evidence is required but not provided. The fact that CUBIC does not <br>
change MD factor for fast convergence is the root of the problem <br>
evidenced in the paper and remains so in the algo specified in this <br>
draft. This is also a significant problem when competing with Reno CC <br>
because CUBIC behaves much more aggressive than Reno CC when there is <br>
sudden congestion and all competing flows must converge fast down to a <br>
small fraction of the current cwnd to be fair to each other. This again <br=
>
cannot be evidenced not to be a problem by long deployment experience <br>
unless experimental data that measures the impact on competing traffic is <=
br>
presented to back the claims. Adjusting just Wmax for fast convergence is <=
br>
not enough and is even likely to be ineffective because there tend to be <b=
r>
several losses when sudden congestion is hit, and particularly if NewReno <=
br>
is in use the sender stays several RTTs in Fast Recovery being <br>
overaggressive and then possibly continues at the same rate in CA which <br=
>
is unlikely to reach evan close to Wmax before a new loss hits <br>
the sender again. That is, lower Wmax and lower additive increase factor <b=
r>
do not compensate the use of larger MD factor when sudden congestion is <br=
>
encountered.<br>
<br>
#93 &amp; #94 &amp; (#89) Sec 5.3 still does not address any difficult <br>
environments, in particular buffer-bloated paths (nor does Sec 5.4).<br>
We need evidence (results) that show CUBIC is fair towards other CCs <br>
(Reno) also in such environments. Note that CUBIC&#39;s decision to leave <=
br>
Reno-friendly region is based on the size of cwnd which tends to be <br>
incorrect with buffer-bloated bottlenecks because with huge buffers the <br=
>
cwnd can be many times larger than what is actually needed to fully <br>
utilize the available network bit-rate. Therefore, Reno CC has no problem <=
br>
in fully utilizing such bottleneck links and CUBIC must stay in <br>
Reno-friendly region longer but it leaves it too early because the same C <=
br>
is used as with non-bloated environments. We lack experiments showing <br>
CUBIC follows the congestion control principles and is fair to current <br>
standard TCP CC; to my understanding no experiments with buffer-bloated<br>
bottlenecks are cited to back up the claims even though buffer bloat is <br=
>
very well known to be a common (difficult) environment in today&#39;s <br>
Internet.<br>
<br>
#90 The current text on applying undo (a response to detected false <br>
fastrexmit) does not provide correct result if someone implements it.<br>
I have explained the problems there in github but seem to have not <br>
replied to latest comments by Neal. I&#39;ll reply and try to explain more.=
<br>
Again one major problem here is that the draft suggest a new response <br>
algo for false fast rexmits but does not provide any experimental data to <=
br>
support it. Long deployment experience has been suggested as <br>
justification but again without any carefully evaluated experimental <br>
data and evidence there is no meat. The issue is important to solve but is =
<br>
not specific to CUBIC. Instead, it is general problem for all TCP CC <br>
variants. IMHO, this is not ready for standards track but deserves a draft =
<br>
of its own so that it can be carefully evaluated and discussed ion the <br>
list. AFAIK there has been no discussion on this on the tcpm list, so <br>
those probably interested and having experience are likely to be unaware <b=
r>
that this is part of CUBIC draft.<br>
<br>
#88 The problem with correctness of the AIMD model and setting alpha for <b=
r>
CUBIC requires further consideration. Bob provided an analysis that <br>
leaves things still open. It seems that I never had time to review and <br>
comment the analysis and clarify why the model does not work. I&#39;ll do t=
hat <br>
separately as it is important to ensure CUBIC behaves fairly as intended <b=
r>
for the Reno-friendly region.<br>
<br>
Best regards,<br>
<br>
/Markku<br>
<br>
On Mon, 31 Jan 2022, Yoshifumi Nishida wrote:<br>
<br>
&gt; Hello,<br>
&gt; <br>
&gt; After some discussions among chairs, we decided to run the 2nd WGLC on=
 draft-ietf-tcpm-rfc8312bis in<br>
&gt; consideration of the importance of the draft.=C2=A0<br>
&gt; We&#39;ll be grateful if you could send your feedback to the ML. The W=
GLC runs until *Feb 11*.<br>
&gt; <br>
&gt; If interested, you can check in-depth past discussions in the followin=
g URL.<br>
&gt; <a href=3D"https://github.com/NTAP/rfc8312bis/" rel=3D"noreferrer" tar=
get=3D"_blank">https://github.com/NTAP/rfc8312bis/</a><br>
&gt; <br>
&gt; Thank you so much!<br>
&gt; --<br>
&gt; tcpm co-chairs<br>
&gt; <br>
&gt; <br>
&gt; On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert &lt;<a href=3D"mailto:lars=
@eggert.org" target=3D"_blank">lars@eggert.org</a>&gt; wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Hi,<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0this -06 version rolls in all the changes re=
quested during (and after) WGLC ended.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0I&#39;ll leave it up to the chairs to decide=
 if another WGLC is warranted or the document can<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0progress as-is.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Thanks,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Lars<br>
&gt; <br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; On 2022-1-26, at 11:12, <a href=3D"mail=
to:internet-drafts@ietf.org" target=3D"_blank">internet-drafts@ietf.org</a>=
 wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; A New Internet-Draft is available from =
the on-line Internet-Drafts directories.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; This draft is a work item of the TCP Ma=
intenance and Minor Extensions WG of the IETF.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Title=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: CUBIC for Fast and Long-Distance Networ=
ks<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Authors=C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: Lisong Xu<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Sangtae Ha<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Injong Rhee<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Vidhi Goel<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Lars Eggert<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Filename=C2=
=A0 =C2=A0 =C2=A0 =C2=A0 : draft-ietf-tcpm-rfc8312bis-06.txt<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Pages=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: 35<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Date=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 : 2022-01-26<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Abstract:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0CUBIC is a standard TCP con=
gestion control algorithm that uses a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0cubic function instead of a=
 linear congestion window increase<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0function to improve scalabi=
lity and stability over fast and long-<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0distance networks.=C2=A0 CU=
BIC has been adopted as the default TCP<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0congestion control algorith=
m by the Linux, Windows, and Apple stacks.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0This document updates the s=
pecification of CUBIC to include<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0algorithmic improvements ba=
sed on these implementations and recent<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0academic work.=C2=A0 Based =
on the extensive deployment experience with<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0CUBIC, it also moves the sp=
ecification to the Standards Track,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0obsoleting RFC 8312.=C2=A0 =
This also requires updating RFC 5681, to allow<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0for CUBIC&#39;s occasionall=
y more aggressive sending behavior.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; The IETF datatracker status page for th=
is draft is:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://datatracker.ietf.org=
/doc/draft-ietf-tcpm-rfc8312bis/" rel=3D"noreferrer" target=3D"_blank">http=
s://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; There is also an HTML version available=
 at:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://www.ietf.org/archive=
/id/draft-ietf-tcpm-rfc8312bis-06.html" rel=3D"noreferrer" target=3D"_blank=
">https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html</a><br=
>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; A diff from the previous version is ava=
ilable at:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://www.ietf.org/rfcdiff=
?url2=3Ddraft-ietf-tcpm-rfc8312bis-06" rel=3D"noreferrer" target=3D"_blank"=
>https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Internet-Drafts are also available by r=
sync at rsync.ietf.org::internet-drafts<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; _______________________________________=
________<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; tcpm mailing list<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"mailto:tcpm@ietf.org" target=
=3D"_blank">tcpm@ietf.org</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://www.ietf.org/mailman=
/listinfo/tcpm" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/m=
ailman/listinfo/tcpm</a><br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0____________________________________________=
___<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0tcpm mailing list<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"mailto:tcpm@ietf.org" target=3D"_=
blank">tcpm@ietf.org</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"https://www.ietf.org/mailman/list=
info/tcpm" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/mailma=
n/listinfo/tcpm</a><br>
&gt; <br>
&gt; <br>
&gt;</blockquote></div></div>
_______________________________________________<br>
tcpm mailing list<br>
<a href=3D"mailto:tcpm@ietf.org" target=3D"_blank">tcpm@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/tcpm" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/listinfo/tcpm</a><br>
</blockquote></div>

--0000000000001be2ff05d82b2d7a--


From nobody Thu Feb 17 06:38:09 2022
Return-Path: <noreply@ietf.org>
X-Original-To: tcpm@ietf.org
Delivered-To: tcpm@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id A768F3A08AE; Thu, 17 Feb 2022 06:37:56 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: Zaheduzzaman Sarker via Datatracker <noreply@ietf.org>
To: "The IESG" <iesg@ietf.org>
Cc: draft-ietf-tcpm-rfc793bis@ietf.org, tcpm-chairs@ietf.org, tcpm@ietf.org, Michael Scharf <michael.scharf@hs-esslingen.de>, michael.scharf@hs-esslingen.de
X-Test-IDTracker: no
X-IETF-IDTracker: 7.45.0
Auto-Submitted: auto-generated
Precedence: bulk
Reply-To: Zaheduzzaman Sarker <Zaheduzzaman.Sarker@ericsson.com>
Message-ID: <164510867666.30151.4433869292176440236@ietfa.amsl.com>
Date: Thu, 17 Feb 2022 06:37:56 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/T3Toi7TUbuIEoKV67BT0H34jHBs>
Subject: [tcpm] Zaheduzzaman Sarker's No Objection on draft-ietf-tcpm-rfc793bis-26: (with COMMENT)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 17 Feb 2022 14:37:57 -0000

Zaheduzzaman Sarker has entered the following ballot position for
draft-ietf-tcpm-rfc793bis-26: No Objection

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


Please refer to https://www.ietf.org/blog/handling-iesg-ballot-positions/
for more information about how to handle DISCUSS and COMMENT positions.


The document, along with other ballot positions, can be found here:
https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc793bis/



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

Thanks for addressing my Discuss and comments. I am happy with the current
version. This is really a good work...




From nobody Thu Feb 17 06:41:58 2022
Return-Path: <kojo@cs.helsinki.fi>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AC51C3A08F2; Thu, 17 Feb 2022 06:41:56 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cs.helsinki.fi
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6gSQthIyWfu3; Thu, 17 Feb 2022 06:41:50 -0800 (PST)
Received: from script.cs.helsinki.fi (script.cs.helsinki.fi [128.214.11.1]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E13523A08CC; Thu, 17 Feb 2022 06:41:42 -0800 (PST)
X-DKIM: Courier DKIM Filter v0.50+pk-2017-10-25 mail.cs.helsinki.fi Thu, 17 Feb 2022 16:41:30 +0200
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cs.helsinki.fi; h=date:from:to:cc:subject:in-reply-to:message-id:references :mime-version:content-type:content-id; s=dkim20130528; bh=Lnp0qk cBw5TDBtt7KPj6Mtk43b0pcAbMMUR5xZKaJwQ=; b=k8KJWxjaBj04SyKDGXTwLi VMy8NpJFPzQYFFNjBKsUXxidJ+v54UJCTOjzqRpEiJmxOITt3JbYB+lC41zoCqQ0 Ptpu2e5BEDnTs4HKBtE/9m9xkvh3Xr+ZfULe4z1OyZKLLrBgTxghaKecPwpEhB00 AlRZA8KueJNbH0CQhNtUs=
Received: from hp8x-60 (88-113-49-197.elisa-laajakaista.fi [88.113.49.197]) (AUTH: PLAIN kojo, TLS: TLSv1/SSLv3,256bits,AES256-GCM-SHA384) by mail.cs.helsinki.fi with ESMTPSA; Thu, 17 Feb 2022 16:41:30 +0200 id 00000000005A014E.00000000620E5E9A.0000504D
Date: Thu, 17 Feb 2022 16:41:29 +0200 (EET)
From: Markku Kojo <kojo@cs.helsinki.fi>
To: Yoshifumi Nishida <nsd.ietf@gmail.com>
cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
In-Reply-To: <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com>
Message-ID: <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com>
User-Agent: Alpine 2.21 (DEB 202 2017-01-01)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="=_script-20581-1645108890-0001-2"
Content-ID: <alpine.DEB.2.21.2202171541080.4019@hp8x-60.cs.helsinki.fi>
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/TkLSTJGq6W8yo3jfQIpB8MXI7QI>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 17 Feb 2022 14:41:57 -0000

This is a MIME-formatted message.  If you see this text it means that your
E-mail software does not support MIME-formatted messages.

--=_script-20581-1645108890-0001-2
Content-Type: text/plain; charset="iso-8859-15"; format=flowed
Content-Transfer-Encoding: quoted-printable
Content-ID: <alpine.DEB.2.21.2202171541081.4019@hp8x-60.cs.helsinki.fi>

Hi Yoshi,

On Tue, 15 Feb 2022, Yoshifumi Nishida wrote:

> Hi=A0Markku,
>=20
> Thanks for the comments. I think these are very valid points.=A0
> However, I would like to check several things as a co-chair and a doc s=
hepherd before we
> discuss=A0the points you've=A0raised.
>=20
> In my understanding (please correct me if I'm wrong), when this draft w=
as adopted as an WG
> item, I think the goal of the doc was some minor updates from RFC8312 w=
hich include more
> clarifications, minor=A0changes and bug fixes.=A0
> However, if we try to address your=A0concerns, I think we'll need to in=
vent a new version of
> CUBIC something like CUBIC++ or NewCUBIC in the end.=A0
> I won't deny the value of such doc, but, this seems not to be what we a=
greed on
> beforehand.=A0=A0
> if we proceed in this direction, I think we will need to check the WG c=
onsensus whether
> this should be a new goal for the doc.
>=20
> So, I would like to check if this is what you intend for the doc or you =
think we can
> address your points while aligning with the original goal.
> Also, if someone has opinions on this, please share.

I think it is important that we remember the status of RFC 8312 and the=20
decades long process that has been followed in tsv area for new=20
TCP congestion control algorithms that have been proposed and submitted=20
to IETF. In order to ensure that new cc algos are safe and fair, the=20
process that has been followed for all current stds track TCP cc algos=20
has required that the cc algo is first accepted and published as=20
experimental RFC and only once enough supportive experimental evidence=20
has been gathered the doc has become a candidate to be forwaded to stds=20
track. We have even agreed on a relatively strict evaluation process to=20
follow when cc algos are brought to the IETF to be published as=20
experimental:

https://www.ietf.org/about/groups/iesg/statements/experimental-congestion=
-control/

RFC 8312 was published as "Informational" and if I recall correctly the=20
idea was "just to publish what's out there" for the benefit of the=20
community. RFC 8312 was never really evaluated, particularly not in the=20
way new cc algos are supposed to be as per the agreed process.

I do not recall what/how exactly was agreed when rfc8312bis was launched=20
but I would be very interested to hear the justification why this doc=20
does not need to follow the process mentioned above but we would like to=20
propose IETF to publish a non-evaluated Informational doc to be published=20
"with minor updates", i.e., without actual evaluation, as a stds track=20
RFC? If the target really remains as PS then the bar should be even=20
higher than what is described for experimental docs in the above process=20
document, i.e, what we have followod for experimental to be moved to stds=20
track.

The only justification that I have heard has beed "because CUBIC has long=20
and wide deployment experience" and "the Internet has not smelted or that=20
"we should have noticed if there were problems". We must, however,=20
understand that in order to have noticeable bad impact CUBIC should cause=20
some sort of congestion collapse. Congestion collapse, however, is not an=20
issue with CUBIC nor with any other CC algo that applies an RTO mechanism=
s=20
together with correctly implemented Karn's algo that retains the=20
backed-off RTO until an Ack is received for a new (not rexmitted) data=20
packet. The issue is fairness to competing traffic. This cannot be=20
observed by deploying and measuring the performance and behaviour of CUBI=
C=20
alone. CUBIC being more aggressive than current stds track TCP CC would=20
just gives good performance results that one running CUBIC would be happy=20
with. One must evaluate CUBIC's impact on the competing (Reno CC) traffic=20
in range of environments which requires carefully designed active=20
measurements with thoroughly-analyzed results (as required by the above=20
process document, RFC 5033 and RFC 2914). What we seem to be missing is=20
this evidence on CUBIC's impact and that is something the IETF must focus=20
on, not just that whether CUBIC can achieve better performance than other=20
existing CCs. The latter has been shown in many publications and is the=20
majos focus in  many scientific papers proposing new algos.
I appreciate a lot that CUBIC has been implemented/developped and=20
deployed for long and I wonder whether those deploying CUBIC have=20
unpublished results the wg could review before taking the decicion?

I suggest everyone to read carefully RFC 2914 Sec 3.2 and particularly=20
what it says about more aggressive (than RFC 5681) congestion control=20
algorithms:

  Some of these may fail to implement
  the TCP congestion avoidance mechanisms correctly because of poor
  implementation [RFC2525].  Others may deliberately be implemented
  with congestion avoidance algorithms that are more aggressive in
  their use of bandwidth than other TCP implementations; this would
  allow a vendor to claim to have a "faster TCP".  The logical
  consequence of such implementations would be a spiral of increasingly
  aggressive TCP implementations, or increasingly aggressive transport
  protocols, leading back to the point where there is effectively no
  congestion avoidance and the Internet is chronically congested.

And:

  It is convenient to divide flows into three classes: (1) TCP-
  compatible flows, (2) unresponsive flows, i.e., flows that do not
  slow down when congestion occurs, and (3) flows that are responsive
  but are not TCP-compatible.  The last two classes contain more
  aggressive flows that pose significant threats to Internet
  performance,

As I have tried to point out there are several features with CUBIC where=20
it is likely to be (or to me it seems it obviously is) more aggressive=20
than what is reguired to be TCP-compatible. I'm not aware of evidince=20
presented to tcpm (or IETF/IRTF) which shows opposite (and I happy to be=20
educated what I have missed).

You may take my comments to be a part of the expert review phase=20
performed by the IRTF/ICCRG for CUBIC. I'm not requesting to modify this=20
doc to CUBIC++ (or something) but it seems to be that this would be=20
necessary if this doc intends to become published as PS. For experimental=
,=20
I think it would need some addtioinal updates and record the areas=20
uncertainty and where more experimentation (clearly) is required.

Thanks,

/Markku

> Thanks,
> --
> Yoshi
>=20
>=20
> On Fri, Feb 11, 2022 at 9:34 PM Markku Kojo <kojo@cs.helsinki.fi> wrote=
:
>       Hi Yoshi, all,
>
>       It seems to me that many issues that I raised have been solved, t=
hanks.
>       However, there are still a number of important issues that have n=
ot been
>       addressed adequately. At least the following:
>
>       #135 on W_max. Yoshi's observation was correct that this is not r=
esolved:
>       the co-authors and original developpers of CUBIC (@lisongxu and
>       @sangtaeha) agreed in their last message that Wmax needs differen=
t
>       treatment for slow start and congestion avoidance and plan compre=
hensive
>       (new) evaluation of it. This is obviously an open issue but the i=
ssue
>       was closed?
>
>       #85 (& #86 with basically same issue and these two were combined) =
This
>       (#85) is about ECN but the major issue is on using the same MD
>       factor in slow start and in congestion avoidance when using
>       loss-based CC. This (#85) remained closed even though I provided =
a
>       thorough explanation why it is wrong and against the original the=
ory and
>       design by Van Jacobson, against the congestion control principles =
(RFC
>       2914) and two co-authors agreed on this in their same last messag=
e to
>       #135 when they agreed on Wmax needing rework. This is an importan=
t issue
>       that the wg should consider very carefully because it is not only
>       updating RFC 5681 but also in conflict with RFC 2914. How can
>       tcpm (and IETF) suggest and allow one CC algo to not follow conge=
stion
>       control principles as set in RFC 2914 while requiring all other C=
Cs to
>       follow RFC 2914 guidelines?
>       The current draft does not provide any justification for using th=
e same
>       MD factor in slow start as in congestion avoidance. Nor am I
>       awere of any experimental data that would support this change.
>       The fact thet CUBIC has been long deployed does not alone provide =
any
>       supporting evidence because CUBIC is likely to give good performa=
nce as
>       it is overagressive and thereby unfair to competing traffic and u=
sers
>       tend to be happy when measuring the performance of the sending CU=
BIC
>       only, not the competing traffic that is badly impacted. HyStart++ =
is
>       suggested as mitigation to the problem but it cannot; HyStart++ i=
s only
>       applicable during initial slow start, not during slow start after =
RTO!
>       That is, the "SHOULD use HYStart++" text in Sec 4.10 is impossibl=
e
>       to implement as I have pointed out in my comments earlier. Using =
a proper
>       MD factor in slow start is even more important if loss is detecte=
cted
>       during a RTO recovery because the sender is likely to face heavy
>       congestion in such a case and it is very bad if the sender contin=
ues
>       sending with overaggressive rate, stealing the capacity from and =
causing
>       harm to coexisting flows. In addition, as I have explained, HySta=
rt++
>       does not remove the problem even for the initial slow start as it =
is not
>       shown to work always. Instead, the results with the HyStart++ dra=
ft show
>       that it reduces 50% of rexmits and only 36% RTOs, meaning that th=
ere is
>       likely to be a notable percentage of cases when a sender is still
>       in slow start when first loss is detected (i.e., HyStart++ had no =
effect)
>       and a significant number of cases where a CUBIC sender is
>       overaggressive continuing with a 1-40% larger cwnd than what is t=
he
>       available capacity. Note also that any delay-based heuristics lik=
e
>       HyStart++ are known to work poorly in various wireless environmen=
s where
>       link delay tends to vary a lot. We may come up with some other MD =
factor
>       that 0.5 when in slow start and HyStart++ is in use, but that is
>       experimental, if not research, and definitely not ready for stds =
track.
>
>       #114, #132, and #143 w.r.t flightsize vs. cwnd. The current text
>       =A0 =A0 =A0 does not quite correcly reflect what stacks that use =
cwnd do.
>       =A0 =A0 =A0 I'll comment in #143 separately.
>
>       #96 & #98: The text added does not address the problems raised wh=
ich are
>       also evidenced in the paper pointed by Bob in #96. Even though CU=
BIC has
>       been modified a bit after the paper was published, it does not
>       automatically mean that the problem has been shown resolved: expe=
rimental
>       evidence is required but not provided. The fact that CUBIC does n=
ot
>       change MD factor for fast convergence is the root of the problem
>       evidenced in the paper and remains so in the algo specified in th=
is
>       draft. This is also a significant problem when competing with Ren=
o CC
>       because CUBIC behaves much more aggressive than Reno CC when ther=
e is
>       sudden congestion and all competing flows must converge fast down =
to a
>       small fraction of the current cwnd to be fair to each other. This =
again
>       cannot be evidenced not to be a problem by long deployment experi=
ence
>       unless experimental data that measures the impact on competing tr=
affic is
>       presented to back the claims. Adjusting just Wmax for fast conver=
gence is
>       not enough and is even likely to be ineffective because there ten=
d to be
>       several losses when sudden congestion is hit, and particularly if =
NewReno
>       is in use the sender stays several RTTs in Fast Recovery being
>       overaggressive and then possibly continues at the same rate in CA =
which
>       is unlikely to reach evan close to Wmax before a new loss hits
>       the sender again. That is, lower Wmax and lower additive increase =
factor
>       do not compensate the use of larger MD factor when sudden congest=
ion is
>       encountered.
>
>       #93 & #94 & (#89) Sec 5.3 still does not address any difficult
>       environments, in particular buffer-bloated paths (nor does Sec 5.=
4).
>       We need evidence (results) that show CUBIC is fair towards other =
CCs
>       (Reno) also in such environments. Note that CUBIC's decision to l=
eave
>       Reno-friendly region is based on the size of cwnd which tends to =
be
>       incorrect with buffer-bloated bottlenecks because with huge buffe=
rs the
>       cwnd can be many times larger than what is actually needed to ful=
ly
>       utilize the available network bit-rate. Therefore, Reno CC has no =
problem
>       in fully utilizing such bottleneck links and CUBIC must stay in
>       Reno-friendly region longer but it leaves it too early because th=
e same C
>       is used as with non-bloated environments. We lack experiments sho=
wing
>       CUBIC follows the congestion control principles and is fair to cu=
rrent
>       standard TCP CC; to my understanding no experiments with buffer-b=
loated
>       bottlenecks are cited to back up the claims even though buffer bl=
oat is
>       very well known to be a common (difficult) environment in today's
>       Internet.
>
>       #90 The current text on applying undo (a response to detected fal=
se
>       fastrexmit) does not provide correct result if someone implements =
it.
>       I have explained the problems there in github but seem to have no=
t
>       replied to latest comments by Neal. I'll reply and try to explain =
more.
>       Again one major problem here is that the draft suggest a new resp=
onse
>       algo for false fast rexmits but does not provide any experimental =
data to
>       support it. Long deployment experience has been suggested as
>       justification but again without any carefully evaluated experimen=
tal
>       data and evidence there is no meat. The issue is important to sol=
ve but is
>       not specific to CUBIC. Instead, it is general problem for all TCP =
CC
>       variants. IMHO, this is not ready for standards track but deserve=
s a draft
>       of its own so that it can be carefully evaluated and discussed io=
n the
>       list. AFAIK there has been no discussion on this on the tcpm list=
, so
>       those probably interested and having experience are likely to be =
unaware
>       that this is part of CUBIC draft.
>
>       #88 The problem with correctness of the AIMD model and setting al=
pha for
>       CUBIC requires further consideration. Bob provided an analysis th=
at
>       leaves things still open. It seems that I never had time to revie=
w and
>       comment the analysis and clarify why the model does not work. I'l=
l do that
>       separately as it is important to ensure CUBIC behaves fairly as i=
ntended
>       for the Reno-friendly region.
>
>       Best regards,
>
>       /Markku
>
>       On Mon, 31 Jan 2022, Yoshifumi Nishida wrote:
>
>       > Hello,
>       >
>       > After some discussions among chairs, we decided to run the 2nd =
WGLC on
>       draft-ietf-tcpm-rfc8312bis in
>       > consideration of the importance of the draft.=A0
>       > We'll be grateful if you could send your feedback to the ML. Th=
e WGLC runs
>       until *Feb 11*.
>       >
>       > If interested, you can check in-depth past discussions in the f=
ollowing URL.
>       > https://github.com/NTAP/rfc8312bis/
>       >
>       > Thank you so much!
>       > --
>       > tcpm co-chairs
>       >
>       >
>       > On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <lars@eggert.org> w=
rote:
>       >=A0 =A0 =A0 =A0Hi,
>       >
>       >=A0 =A0 =A0 =A0this -06 version rolls in all the changes request=
ed during (and after)
>       WGLC ended.
>       >
>       >=A0 =A0 =A0 =A0I'll leave it up to the chairs to decide if anoth=
er WGLC is warranted
>       or the document can
>       >=A0 =A0 =A0 =A0progress as-is.
>       >
>       >=A0 =A0 =A0 =A0Thanks,
>       >=A0 =A0 =A0 =A0Lars
>       >
>       >
>       >=A0 =A0 =A0 =A0> On 2022-1-26, at 11:12, internet-drafts@ietf.or=
g wrote:
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> A New Internet-Draft is available from the on-l=
ine Internet-Drafts
>       directories.
>       >=A0 =A0 =A0 =A0> This draft is a work item of the TCP Maintenanc=
e and Minor
>       Extensions WG of the IETF.
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 Title=A0 =A0 =A0 =A0 =A0 =A0: CU=
BIC for Fast and Long-Distance Networks
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 Authors=A0 =A0 =A0 =A0 =A0: Liso=
ng Xu
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 Sangtae Ha
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 Injong Rhee
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 Vidhi Goel
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 Lars Eggert
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0Filename=A0 =A0 =A0 =A0 : draft-i=
etf-tcpm-rfc8312bis-06.txt
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0Pages=A0 =A0 =A0 =A0 =A0 =A0: 35
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0Date=A0 =A0 =A0 =A0 =A0 =A0 : 202=
2-01-26
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> Abstract:
>       >=A0 =A0 =A0 =A0>=A0 =A0CUBIC is a standard TCP congestion contro=
l algorithm that uses a
>       >=A0 =A0 =A0 =A0>=A0 =A0cubic function instead of a linear conges=
tion window increase
>       >=A0 =A0 =A0 =A0>=A0 =A0function to improve scalability and stabi=
lity over fast and long-
>       >=A0 =A0 =A0 =A0>=A0 =A0distance networks.=A0 CUBIC has been adop=
ted as the default TCP
>       >=A0 =A0 =A0 =A0>=A0 =A0congestion control algorithm by the Linux=
, Windows, and Apple
>       stacks.
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0This document updates the specification o=
f CUBIC to include
>       >=A0 =A0 =A0 =A0>=A0 =A0algorithmic improvements based on these i=
mplementations and recent
>       >=A0 =A0 =A0 =A0>=A0 =A0academic work.=A0 Based on the extensive =
deployment experience with
>       >=A0 =A0 =A0 =A0>=A0 =A0CUBIC, it also moves the specification to =
the Standards Track,
>       >=A0 =A0 =A0 =A0>=A0 =A0obsoleting RFC 8312.=A0 This also require=
s updating RFC 5681, to
>       allow
>       >=A0 =A0 =A0 =A0>=A0 =A0for CUBIC's occasionally more aggressive =
sending behavior.
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> The IETF datatracker status page for this draft =
is:
>       >=A0 =A0 =A0 =A0> https://datatracker.ietf.org/doc/draft-ietf-tcp=
m-rfc8312bis/
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> There is also an HTML version available at:
>       >=A0 =A0 =A0 =A0> https://www.ietf.org/archive/id/draft-ietf-tcpm=
-rfc8312bis-06.html
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> A diff from the previous version is available a=
t:
>       >=A0 =A0 =A0 =A0> https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-=
tcpm-rfc8312bis-06
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> Internet-Drafts are also available by rsync at
>       rsync.ietf.org::internet-drafts
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> _______________________________________________
>       >=A0 =A0 =A0 =A0> tcpm mailing list
>       >=A0 =A0 =A0 =A0> tcpm@ietf.org
>       >=A0 =A0 =A0 =A0> https://www.ietf.org/mailman/listinfo/tcpm
>       >
>       >=A0 =A0 =A0 =A0_______________________________________________
>       >=A0 =A0 =A0 =A0tcpm mailing list
>       >=A0 =A0 =A0 =A0tcpm@ietf.org
>       >=A0 =A0 =A0 =A0https://www.ietf.org/mailman/listinfo/tcpm
>       >
>       >
>       >
>=20
>=20
>
--=_script-20581-1645108890-0001-2--


From nobody Thu Feb 17 08:14:12 2022
Return-Path: <ncardwell@google.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 838B73A0A8E for <tcpm@ietfa.amsl.com>; Thu, 17 Feb 2022 08:14:10 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -17.598
X-Spam-Level: 
X-Spam-Status: No, score=-17.598 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_MED=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, ENV_AND_HDR_SPF_MATCH=-0.5, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001, USER_IN_DEF_DKIM_WL=-7.5, USER_IN_DEF_SPF_WL=-7.5] autolearn=unavailable autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=google.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gvuV1YSX1m1w for <tcpm@ietfa.amsl.com>; Thu, 17 Feb 2022 08:14:05 -0800 (PST)
Received: from mail-qk1-x730.google.com (mail-qk1-x730.google.com [IPv6:2607:f8b0:4864:20::730]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id EB22C3A0A8C for <tcpm@ietf.org>; Thu, 17 Feb 2022 08:14:04 -0800 (PST)
Received: by mail-qk1-x730.google.com with SMTP id bs32so5195825qkb.1 for <tcpm@ietf.org>; Thu, 17 Feb 2022 08:14:04 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=dTXU6Xy6qwaHemzXrNKnuE2UYYLxrVDGfD0syj/BEzA=; b=JY5DChdI5pugp2CwC82oNu9UY7/q+VhcQBMq7O3GBjeFOCugpnyUeSmN7CAg6Y/99f b4wjJp9/bFKIrbGYvBOpEoK/57dEj160g1FXYl4SGYa/x1zL8z8buce1ef3O6KSbRZxb rUY8TMfWUkVk8IE3rSL0SCKazoIG6rgBFL2cieI6e5sv8GFd5UEne2VeU2FB5tTetILi B4ecEEI/PV4IwcY4H4tNUO/sWDQ9KzFvAvwDMThKxhLVwGMI4ibonJV9dc/62CFQU4QZ yb6Sdj9HEl2+fHRakXPowrAjLaD5/RZrMb51Ac6Zg7bBrEYz9PA78HJiAPKDPdNBJUPQ dWdA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=dTXU6Xy6qwaHemzXrNKnuE2UYYLxrVDGfD0syj/BEzA=; b=Vn2K6MplOYpCr8XTNmdYzfRC1ihAaIxf4k92biyVsEO2Lc9IgmdEf4ACbv6OMVekXq EGt1EHgBHSr5K0zXMvqr90uDvMPWUcposl8UggH7v6EhRqD0TTUPtnRCSls1BPrXhYSc 5w8wmFtWEyfcMsZ1iJmE0Sxs6mRcwO3dPG2RG9QxUnedidcGpZ7FvyeM5Sgk0URCBF4o xa8iX95FSixYoLFf6P2VzFXukCVcJolEVlQshtkiv8A0BpJMsmzmlqlzQCnE8HZ3Yy42 ijCYqJAIrvbBNZzufsnSM5a1u849Q6sYYOJL3irxRTVK6MdbiR7OSgkgwoqAEkzKtOAO Y/gg==
X-Gm-Message-State: AOAM5322xlu+Z5rhADZR6eA0dFrIHWdCO28BPriYORm93jK/kauFiyh3 xmJNQHKJnZKAHaM/t+RNDdTJG03HPtaHEhU59qYtMQ==
X-Google-Smtp-Source: ABdhPJzYLcTWpp0PMgwh/JR2RVx8+CopPZx8xe4ipt1iA31g9VwgwDnW/gST1CN4wxcgvM65kXVZa4hTJnCRM2uytNY=
X-Received: by 2002:a37:c06:0:b0:49b:7a31:cd54 with SMTP id 6-20020a370c06000000b0049b7a31cd54mr2054383qkm.358.1645114443335; Thu, 17 Feb 2022 08:14:03 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi>
In-Reply-To: <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi>
From: Neal Cardwell <ncardwell@google.com>
Date: Thu, 17 Feb 2022 11:13:47 -0500
Message-ID: <CADVnQymwCdm0NoD95SX8JpRXFok=qMygPrwaeTJ9z29dqH=Avw@mail.gmail.com>
To: Markku Kojo <kojo=40cs.helsinki.fi@dmarc.ietf.org>
Cc: Yoshifumi Nishida <nsd.ietf@gmail.com>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>,  tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: multipart/alternative; boundary="000000000000033e8005d8390ebe"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/unZJAhRlHHMXivNo8_aJ2Wx6X8c>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 17 Feb 2022 16:14:11 -0000

--000000000000033e8005d8390ebe
Content-Type: text/plain; charset="UTF-8"

On Thu, Feb 17, 2022 at 9:42 AM Markku Kojo <kojo=
40cs.helsinki.fi@dmarc.ietf.org> wrote:

> Hi Yoshi,
>
> On Tue, 15 Feb 2022, Yoshifumi Nishida wrote:
>
> > Hi Markku,
> >
> > Thanks for the comments. I think these are very valid points.
> > However, I would like to check several things as a co-chair and a doc
> shepherd before we
> > discuss the points you've raised.
> >
> > In my understanding (please correct me if I'm wrong), when this draft
> was adopted as an WG
> > item, I think the goal of the doc was some minor updates from RFC8312
> which include more
> > clarifications, minor changes and bug fixes.
> > However, if we try to address your concerns, I think we'll need to
> invent a new version of
> > CUBIC something like CUBIC++ or NewCUBIC in the end.
> > I won't deny the value of such doc, but, this seems not to be what we
> agreed on
> > beforehand.
> > if we proceed in this direction, I think we will need to check the WG
> consensus whether
> > this should be a new goal for the doc.
> >
> > So, I would like to check if this is what you intend for the doc or you
> think we can
> > address your points while aligning with the original goal.
> > Also, if someone has opinions on this, please share.
>
> I think it is important that we remember the status of RFC 8312 and the
> decades long process that has been followed in tsv area for new
> TCP congestion control algorithms that have been proposed and submitted
> to IETF. In order to ensure that new cc algos are safe and fair, the
> process that has been followed for all current stds track TCP cc algos
> has required that the cc algo is first accepted and published as
> experimental RFC and only once enough supportive experimental evidence
> has been gathered the doc has become a candidate to be forwaded to stds
> track. We have even agreed on a relatively strict evaluation process to
> follow when cc algos are brought to the IETF to be published as
> experimental:
>
>
> https://www.ietf.org/about/groups/iesg/statements/experimental-congestion-control/
>
> RFC 8312 was published as "Informational" and if I recall correctly the
> idea was "just to publish what's out there" for the benefit of the
> community. RFC 8312 was never really evaluated, particularly not in the
> way new cc algos are supposed to be as per the agreed process.
>
> I do not recall what/how exactly was agreed when rfc8312bis was launched
> but I would be very interested to hear the justification why this doc
> does not need to follow the process mentioned above but we would like to
> propose IETF to publish a non-evaluated Informational doc to be published
> "with minor updates", i.e., without actual evaluation, as a stds track
> RFC? If the target really remains as PS then the bar should be even
> higher than what is described for experimental docs in the above process
> document, i.e, what we have followod for experimental to be moved to stds
> track.
>
> The only justification that I have heard has beed "because CUBIC has long
> and wide deployment experience" and "the Internet has not smelted or that
> "we should have noticed if there were problems". We must, however,
> understand that in order to have noticeable bad impact CUBIC should cause
> some sort of congestion collapse. Congestion collapse, however, is not an
> issue with CUBIC nor with any other CC algo that applies an RTO mechanisms
> together with correctly implemented Karn's algo that retains the
> backed-off RTO until an Ack is received for a new (not rexmitted) data
> packet. The issue is fairness to competing traffic. This cannot be
> observed by deploying and measuring the performance and behaviour of CUBIC
> alone. CUBIC being more aggressive than current stds track TCP CC would
> just gives good performance results that one running CUBIC would be happy
> with. One must evaluate CUBIC's impact on the competing (Reno CC) traffic
> in range of environments which requires carefully designed active
> measurements with thoroughly-analyzed results (as required by the above
> process document, RFC 5033 and RFC 2914). What we seem to be missing is
> this evidence on CUBIC's impact and that is something the IETF must focus
> on, not just that whether CUBIC can achieve better performance than other
> existing CCs. The latter has been shown in many publications and is the
> majos focus in  many scientific papers proposing new algos.
> I appreciate a lot that CUBIC has been implemented/developped and
> deployed for long and I wonder whether those deploying CUBIC have
> unpublished results the wg could review before taking the decicion?
>
> I suggest everyone to read carefully RFC 2914 Sec 3.2 and particularly
> what it says about more aggressive (than RFC 5681) congestion control
> algorithms:
>
>   Some of these may fail to implement
>   the TCP congestion avoidance mechanisms correctly because of poor
>   implementation [RFC2525].  Others may deliberately be implemented
>   with congestion avoidance algorithms that are more aggressive in
>   their use of bandwidth than other TCP implementations; this would
>   allow a vendor to claim to have a "faster TCP".  The logical
>   consequence of such implementations would be a spiral of increasingly
>   aggressive TCP implementations, or increasingly aggressive transport
>   protocols, leading back to the point where there is effectively no
>   congestion avoidance and the Internet is chronically congested.
>
> And:
>
>   It is convenient to divide flows into three classes: (1) TCP-
>   compatible flows, (2) unresponsive flows, i.e., flows that do not
>   slow down when congestion occurs, and (3) flows that are responsive
>   but are not TCP-compatible.  The last two classes contain more
>   aggressive flows that pose significant threats to Internet
>   performance,
>
> As I have tried to point out there are several features with CUBIC where
> it is likely to be (or to me it seems it obviously is) more aggressive
> than what is reguired to be TCP-compatible. I'm not aware of evidince
> presented to tcpm (or IETF/IRTF) which shows opposite (and I happy to be
> educated what I have missed).
>
> You may take my comments to be a part of the expert review phase
> performed by the IRTF/ICCRG for CUBIC. I'm not requesting to modify this
> doc to CUBIC++ (or something) but it seems to be that this would be
> necessary if this doc intends to become published as PS. For experimental,
> I think it would need some addtioinal updates and record the areas
> uncertainty and where more experimentation (clearly) is required.
>

Hi Markku,

Thanks for your careful argument above. Given your argument, in order to
continue to make constructive progress on rfc8312bis without blocking for
new research and experimentation, I would propose that we change rfc8312bis
so that it remains "Informational" (like RFC8312) rather than the
"Standards Track" status currently listed at:

  https://ntap.github.io/rfc8312bis/draft-ietf-tcpm-rfc8312bis.html

As far as I can tell, that would allow all of the valuable contributions in
rfc8312bis to carry forward and be published, while still respecting all of
the considerations you enumerated above, which center around concerns with
calling CUBIC "standards track".

How does that sound?

best,
neal

--000000000000033e8005d8390ebe
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote">=
<div dir=3D"ltr" class=3D"gmail_attr">On Thu, Feb 17, 2022 at 9:42 AM Markk=
u Kojo &lt;kojo=3D<a href=3D"mailto:40cs.helsinki.fi@dmarc.ietf.org">40cs.h=
elsinki.fi@dmarc.ietf.org</a>&gt; wrote:<br></div><blockquote class=3D"gmai=
l_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,20=
4,204);padding-left:1ex">Hi Yoshi,<br>
<br>
On Tue, 15 Feb 2022, Yoshifumi Nishida wrote:<br>
<br>
&gt; Hi=C2=A0Markku,<br>
&gt; <br>
&gt; Thanks for the comments. I think these are very valid points.=C2=A0<br=
>
&gt; However, I would like to check several things as a co-chair and a doc =
shepherd before we<br>
&gt; discuss=C2=A0the points you&#39;ve=C2=A0raised.<br>
&gt; <br>
&gt; In my understanding (please correct me if I&#39;m wrong), when this dr=
aft was adopted as an WG<br>
&gt; item, I think the goal of the doc was some minor updates from RFC8312 =
which include more<br>
&gt; clarifications, minor=C2=A0changes and bug fixes.=C2=A0<br>
&gt; However, if we try to address your=C2=A0concerns, I think we&#39;ll ne=
ed to invent a new version of<br>
&gt; CUBIC something like CUBIC++ or NewCUBIC in the end.=C2=A0<br>
&gt; I won&#39;t deny the value of such doc, but, this seems not to be what=
 we agreed on<br>
&gt; beforehand.=C2=A0=C2=A0<br>
&gt; if we proceed in this direction, I think we will need to check the WG =
consensus whether<br>
&gt; this should be a new goal for the doc.<br>
&gt; <br>
&gt; So, I would like to check if this is what you intend for the doc or yo=
u think we can<br>
&gt; address your points while aligning with the original goal.<br>
&gt; Also, if someone has opinions on this, please share.<br>
<br>
I think it is important that we remember the status of RFC 8312 and the <br=
>
decades long process that has been followed in tsv area for new <br>
TCP congestion control algorithms that have been proposed and submitted <br=
>
to IETF. In order to ensure that new cc algos are safe and fair, the <br>
process that has been followed for all current stds track TCP cc algos <br>
has required that the cc algo is first accepted and published as <br>
experimental RFC and only once enough supportive experimental evidence <br>
has been gathered the doc has become a candidate to be forwaded to stds <br=
>
track. We have even agreed on a relatively strict evaluation process to <br=
>
follow when cc algos are brought to the IETF to be published as <br>
experimental:<br>
<br>
<a href=3D"https://www.ietf.org/about/groups/iesg/statements/experimental-c=
ongestion-control/" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.o=
rg/about/groups/iesg/statements/experimental-congestion-control/</a><br>
<br>
RFC 8312 was published as &quot;Informational&quot; and if I recall correct=
ly the <br>
idea was &quot;just to publish what&#39;s out there&quot; for the benefit o=
f the <br>
community. RFC 8312 was never really evaluated, particularly not in the <br=
>
way new cc algos are supposed to be as per the agreed process.<br>
<br>
I do not recall what/how exactly was agreed when rfc8312bis was launched <b=
r>
but I would be very interested to hear the justification why this doc <br>
does not need to follow the process mentioned above but we would like to <b=
r>
propose IETF to publish a non-evaluated Informational doc to be published <=
br>
&quot;with minor updates&quot;, i.e., without actual evaluation, as a stds =
track <br>
RFC? If the target really remains as PS then the bar should be even <br>
higher than what is described for experimental docs in the above process <b=
r>
document, i.e, what we have followod for experimental to be moved to stds <=
br>
track.<br>
<br>
The only justification that I have heard has beed &quot;because CUBIC has l=
ong <br>
and wide deployment experience&quot; and &quot;the Internet has not smelted=
 or that <br>
&quot;we should have noticed if there were problems&quot;. We must, however=
, <br>
understand that in order to have noticeable bad impact CUBIC should cause <=
br>
some sort of congestion collapse. Congestion collapse, however, is not an <=
br>
issue with CUBIC nor with any other CC algo that applies an RTO mechanisms =
<br>
together with correctly implemented Karn&#39;s algo that retains the <br>
backed-off RTO until an Ack is received for a new (not rexmitted) data <br>
packet. The issue is fairness to competing traffic. This cannot be <br>
observed by deploying and measuring the performance and behaviour of CUBIC =
<br>
alone. CUBIC being more aggressive than current stds track TCP CC would <br=
>
just gives good performance results that one running CUBIC would be happy <=
br>
with. One must evaluate CUBIC&#39;s impact on the competing (Reno CC) traff=
ic <br>
in range of environments which requires carefully designed active <br>
measurements with thoroughly-analyzed results (as required by the above <br=
>
process document, RFC 5033 and RFC 2914). What we seem to be missing is <br=
>
this evidence on CUBIC&#39;s impact and that is something the IETF must foc=
us <br>
on, not just that whether CUBIC can achieve better performance than other <=
br>
existing CCs. The latter has been shown in many publications and is the <br=
>
majos focus in=C2=A0 many scientific papers proposing new algos.<br>
I appreciate a lot that CUBIC has been implemented/developped and <br>
deployed for long and I wonder whether those deploying CUBIC have <br>
unpublished results the wg could review before taking the decicion?<br>
<br>
I suggest everyone to read carefully RFC 2914 Sec 3.2 and particularly <br>
what it says about more aggressive (than RFC 5681) congestion control <br>
algorithms:<br>
<br>
=C2=A0 Some of these may fail to implement<br>
=C2=A0 the TCP congestion avoidance mechanisms correctly because of poor<br=
>
=C2=A0 implementation [RFC2525].=C2=A0 Others may deliberately be implement=
ed<br>
=C2=A0 with congestion avoidance algorithms that are more aggressive in<br>
=C2=A0 their use of bandwidth than other TCP implementations; this would<br=
>
=C2=A0 allow a vendor to claim to have a &quot;faster TCP&quot;.=C2=A0 The =
logical<br>
=C2=A0 consequence of such implementations would be a spiral of increasingl=
y<br>
=C2=A0 aggressive TCP implementations, or increasingly aggressive transport=
<br>
=C2=A0 protocols, leading back to the point where there is effectively no<b=
r>
=C2=A0 congestion avoidance and the Internet is chronically congested.<br>
<br>
And:<br>
<br>
=C2=A0 It is convenient to divide flows into three classes: (1) TCP-<br>
=C2=A0 compatible flows, (2) unresponsive flows, i.e., flows that do not<br=
>
=C2=A0 slow down when congestion occurs, and (3) flows that are responsive<=
br>
=C2=A0 but are not TCP-compatible.=C2=A0 The last two classes contain more<=
br>
=C2=A0 aggressive flows that pose significant threats to Internet<br>
=C2=A0 performance,<br>
<br>
As I have tried to point out there are several features with CUBIC where <b=
r>
it is likely to be (or to me it seems it obviously is) more aggressive <br>
than what is reguired to be TCP-compatible. I&#39;m not aware of evidince <=
br>
presented to tcpm (or IETF/IRTF) which shows opposite (and I happy to be <b=
r>
educated what I have missed).<br>
<br>
You may take my comments to be a part of the expert review phase <br>
performed by the IRTF/ICCRG for CUBIC. I&#39;m not requesting to modify thi=
s <br>
doc to CUBIC++ (or something) but it seems to be that this would be <br>
necessary if this doc intends to become published as PS. For experimental, =
<br>
I think it would need some addtioinal updates and record the areas <br>
uncertainty and where more experimentation (clearly) is required.<br></bloc=
kquote><div><br></div><div>Hi Markku,</div><div><br></div>Thanks for your c=
areful argument above. Given your argument, in order to continue to make co=
nstructive progress on rfc8312bis without blocking for new research and exp=
erimentation, I would propose that we change rfc8312bis so that it remains =
&quot;Informational&quot; (like RFC8312) rather than the &quot;Standards Tr=
ack&quot; status currently listed at:<br><br></div><div class=3D"gmail_quot=
e">=C2=A0 <a href=3D"https://ntap.github.io/rfc8312bis/draft-ietf-tcpm-rfc8=
312bis.html">https://ntap.github.io/rfc8312bis/draft-ietf-tcpm-rfc8312bis.h=
tml</a><br><br>As far as I can tell, that would allow all of the valuable c=
ontributions in rfc8312bis to carry forward and be published, while still r=
especting all of the considerations you enumerated above, which center arou=
nd concerns with calling CUBIC &quot;standards track&quot;.</div><div class=
=3D"gmail_quote"><br>How does that sound?<br><br>best,<br>neal<div><br></di=
v></div></div>

--000000000000033e8005d8390ebe--


From nobody Thu Feb 17 18:14:08 2022
Return-Path: <vidhi_goel@apple.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E82F63A0ED8; Thu, 17 Feb 2022 18:14:05 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.672
X-Spam-Level: 
X-Spam-Status: No, score=-2.672 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.576, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=apple.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 24mXDBw63gab; Thu, 17 Feb 2022 18:14:00 -0800 (PST)
Received: from rn-mailsvcp-ppex-lapp14.apple.com (rn-mailsvcp-ppex-lapp14.rno.apple.com [17.179.253.33]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8C7FE3A0EFF; Thu, 17 Feb 2022 18:13:56 -0800 (PST)
Received: from pps.filterd (rn-mailsvcp-ppex-lapp14.rno.apple.com [127.0.0.1]) by rn-mailsvcp-ppex-lapp14.rno.apple.com (8.16.1.2/8.16.1.2) with SMTP id 21I2AOlQ032712; Thu, 17 Feb 2022 18:13:55 -0800
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=apple.com; h=from : message-id : content-type : mime-version : subject : date : in-reply-to : cc : to : references; s=20180706; bh=Vbagg+SB4airFkxnUDid6Z3NdwBKG2UWYZkKZEbIc3E=; b=YhSlgb206OvS4xMgmJByHLwVGHI/lEPy6izUTOcwwKcimRZKZR5e055v307pVsALMDam OipnUJohIKJpUW6hiFIq7tHWZQes91PMgKEePcrO0faSu5hk3aXcWGjQeAq6QShdgdOa llwyDgYHyHN+8ODOlz1AWLos7UKo8cWnEM0v6J/nq6a3Qq56tTRx42r45Xej64jz6Q98 1KEh65mVF+uO+BWzdaeziYNi4c9Pk3FLF1/8Qka5Y3cUEDPrgZUpoBsovr1znd4o2BSb B3zUXI2HHJ62XVmvBiusA/QMOWohQGc0X6+pkXRnA3P8PNw/3ei/YecfnorRMmf5IpT0 XQ== 
Received: from rn-mailsvcp-mta-lapp02.rno.apple.com (rn-mailsvcp-mta-lapp02.rno.apple.com [10.225.203.150]) by rn-mailsvcp-ppex-lapp14.rno.apple.com with ESMTP id 3e8n9649s5-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NO); Thu, 17 Feb 2022 18:13:55 -0800
Received: from rn-mailsvcp-mmp-lapp03.rno.apple.com (rn-mailsvcp-mmp-lapp03.rno.apple.com [17.179.253.16]) by rn-mailsvcp-mta-lapp02.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPS id <0R7H00B5GA77D110@rn-mailsvcp-mta-lapp02.rno.apple.com>;  Thu, 17 Feb 2022 18:13:55 -0800 (PST)
Received: from process_milters-daemon.rn-mailsvcp-mmp-lapp03.rno.apple.com by rn-mailsvcp-mmp-lapp03.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) id <0R7H00P00A5KBV00@rn-mailsvcp-mmp-lapp03.rno.apple.com>; Thu, 17 Feb 2022 18:13:55 -0800 (PST)
X-Va-A: 
X-Va-T-CD: 4b052db715e8685a04b558cbaeaf3790
X-Va-E-CD: 749c6e7503a1c30456497a36735b331a
X-Va-R-CD: 59a455eaeb25ffef4add4d7c725dcf79
X-Va-CD: 0
X-Va-ID: 3473874d-885d-4026-ac20-d85b55f60608
X-V-A: 
X-V-T-CD: 4b052db715e8685a04b558cbaeaf3790
X-V-E-CD: 749c6e7503a1c30456497a36735b331a
X-V-R-CD: 59a455eaeb25ffef4add4d7c725dcf79
X-V-CD: 0
X-V-ID: 279c58e3-4c4e-461d-9697-2a4dea7c8df6
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-17_09:2022-02-17, 2022-02-17 signatures=0
Received: from smtpclient.apple (unknown [17.11.189.243]) by rn-mailsvcp-mmp-lapp03.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPSA id <0R7H00TJ6A767600@rn-mailsvcp-mmp-lapp03.rno.apple.com>; Thu, 17 Feb 2022 18:13:54 -0800 (PST)
From: Vidhi Goel <vidhi_goel@apple.com>
Message-id: <60794576-5ECC-42F0-8422-9556644E16B4@apple.com>
Content-type: multipart/signed; boundary="Apple-Mail=_CE9EFBDC-2E4C-43DC-A4CA-6676BD14A505"; protocol="application/pkcs7-signature"; micalg=sha-256
MIME-version: 1.0 (Mac OS X Mail 14.0 \(3654.100.0.2.11\))
Date: Thu, 17 Feb 2022 18:13:54 -0800
In-reply-to: <CADVnQymwCdm0NoD95SX8JpRXFok=qMygPrwaeTJ9z29dqH=Avw@mail.gmail.com>
Cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
To: Neal Cardwell <ncardwell=40google.com@dmarc.ietf.org>, Markku Kojo <kojo=40cs.helsinki.fi@dmarc.ietf.org>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CADVnQymwCdm0NoD95SX8JpRXFok=qMygPrwaeTJ9z29dqH=Avw@mail.gmail.com>
X-Mailer: Apple Mail (2.3654.100.0.2.11)
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-17_09:2022-02-17, 2022-02-17 signatures=0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/h3PK3j1PYCs68pO5drBCB0qZPho>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 18 Feb 2022 02:14:07 -0000

--Apple-Mail=_CE9EFBDC-2E4C-43DC-A4CA-6676BD14A505
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_3BE016BB-D9BB-426B-B120-37EE5ACE0369"


--Apple-Mail=_3BE016BB-D9BB-426B-B120-37EE5ACE0369
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

I have a general question for the WG, are the "current practices" =
defined in https://datatracker.ietf.org/doc/html/rfc2914 =
<https://datatracker.ietf.org/doc/html/rfc2914> still current? We are 22 =
years ahead of the published date of this draft and there have been =
magnitude of advancements in congestion control algorithms at the hosts =
and what is the right way to measure conformance for a congestion =
control. A very interesting paper about how to quantify a new congestion =
control, which most might have read, "Beyond Jain's Fairness Index: =
Setting the Bar For The Deployment of Congestion Control Algorithms"

I skimmed through RFC 2914 and it references obsoleted drafts for what =
it calls conformant TCP. Like below:

For TCP, this
   means congestion control algorithms conformant with the current TCP
   specification [RFC793, RFC1122 =
<https://datatracker.ietf.org/doc/html/rfc1122>, RFC2581 =
<https://datatracker.ietf.org/doc/html/rfc2581>].


I don=E2=80=99t think we can continue to use old RFCs as a gold standard =
although we can use some the text as is if it still holds true.

>   Some of these may fail to implement
>   the TCP congestion avoidance mechanisms correctly because of poor
>   implementation [RFC2525].  Others may deliberately be implemented
>   with congestion avoidance algorithms that are more aggressive in
>   their use of bandwidth than other TCP implementations; this would
>   allow a vendor to claim to have a "faster TCP".  The logical
>   consequence of such implementations would be a spiral of =
increasingly
>   aggressive TCP implementations, or increasingly aggressive transport
>   protocols, leading back to the point where there is effectively no
>   congestion avoidance and the Internet is chronically congested.

For example, in this text quoted by Markku, I agree that we shouldn=E2=80=99=
t end up with algorithms that eventually lead to a chronically congested =
networks but at the same time, if a more aggressive approach provides =
better efficiency by ramping up a bit faster without leading to more =
congestion, then that shouldn=E2=80=99t be discouraged. Faster TCP =
doesn=E2=80=99t necessarily mean that will cause congestion, it could =
just mean, instead of waiting 100s of RTTs to get to full link =
utilization, it does so in much lesser number of RTTs.

In using Reno as a gold standard, we tend to forget that Reno is =
probably the cause of buffer bloat in the network. It basically requires =
network queues to be configured double the capacity in order to continue =
to have full link utilization after 50% reduction.
And again, this might not have been evident at the time Reno was =
developed. We have realized these problems with time and we should move =
away from continuing to use Reno as the only acceptable Proposed Stds =
track congestion control.


>   It is convenient to divide flows into three classes: (1) TCP-
>   compatible flows, (2) unresponsive flows, i.e., flows that do not
>   slow down when congestion occurs, and (3) flows that are responsive
>   but are not TCP-compatible.  The last two classes contain more
>   aggressive flows that pose significant threats to Internet
>   performance,


This second paragraph is also outdated. IIUC, TCP-compatible flows is =
equivalent to the RFC 2581 which is obsolete. I think we need to replace =
TCP-compatible with something else for two reasons:

1. TCP is not the only transport protocol out there. What about QUIC, =
SCTP?
2. Congestion control algorithms should be evaluated in the same way for =
different transport protocols. So, a better word could be, Conformant =
congestion control, one which once accepted as Stds Track should just =
work for any transport protocol.

Perhaps, we need to show more experimental / deployment data for Cubic =
to become Stds Track but we shouldn=E2=80=99t rely on RFC2914 from the =
year 2000 to decide the acceptance criteria. If we continue to do so, =
then even 100 years from now, RFC 5681 will be the only TCP-conformant / =
TCP-compatible congestion control as defined by RFC 2914 and the one and =
only Stds Track congestion control to ever exist. (Please let me know if =
I missed any other Stds Track congestion control)


Thanks,
Vidhi


> On Feb 17, 2022, at 8:13 AM, Neal Cardwell =
<ncardwell=3D40google.com@dmarc.ietf.org> wrote:
>=20
>=20
>=20
> On Thu, Feb 17, 2022 at 9:42 AM Markku Kojo =
<kojo=3D40cs.helsinki.fi@dmarc.ietf.org =
<mailto:40cs.helsinki.fi@dmarc.ietf.org>> wrote:
> Hi Yoshi,
>=20
> On Tue, 15 Feb 2022, Yoshifumi Nishida wrote:
>=20
> > Hi Markku,
> >=20
> > Thanks for the comments. I think these are very valid points.=20
> > However, I would like to check several things as a co-chair and a =
doc shepherd before we
> > discuss the points you've raised.
> >=20
> > In my understanding (please correct me if I'm wrong), when this =
draft was adopted as an WG
> > item, I think the goal of the doc was some minor updates from =
RFC8312 which include more
> > clarifications, minor changes and bug fixes.=20
> > However, if we try to address your concerns, I think we'll need to =
invent a new version of
> > CUBIC something like CUBIC++ or NewCUBIC in the end.=20
> > I won't deny the value of such doc, but, this seems not to be what =
we agreed on
> > beforehand. =20
> > if we proceed in this direction, I think we will need to check the =
WG consensus whether
> > this should be a new goal for the doc.
> >=20
> > So, I would like to check if this is what you intend for the doc or =
you think we can
> > address your points while aligning with the original goal.
> > Also, if someone has opinions on this, please share.
>=20
> I think it is important that we remember the status of RFC 8312 and =
the=20
> decades long process that has been followed in tsv area for new=20
> TCP congestion control algorithms that have been proposed and =
submitted=20
> to IETF. In order to ensure that new cc algos are safe and fair, the=20=

> process that has been followed for all current stds track TCP cc algos=20=

> has required that the cc algo is first accepted and published as=20
> experimental RFC and only once enough supportive experimental evidence=20=

> has been gathered the doc has become a candidate to be forwaded to =
stds=20
> track. We have even agreed on a relatively strict evaluation process =
to=20
> follow when cc algos are brought to the IETF to be published as=20
> experimental:
>=20
> =
https://www.ietf.org/about/groups/iesg/statements/experimental-congestion-=
control/ =
<https://www.ietf.org/about/groups/iesg/statements/experimental-congestion=
-control/>
>=20
> RFC 8312 was published as "Informational" and if I recall correctly =
the=20
> idea was "just to publish what's out there" for the benefit of the=20
> community. RFC 8312 was never really evaluated, particularly not in =
the=20
> way new cc algos are supposed to be as per the agreed process.
>=20
> I do not recall what/how exactly was agreed when rfc8312bis was =
launched=20
> but I would be very interested to hear the justification why this doc=20=

> does not need to follow the process mentioned above but we would like =
to=20
> propose IETF to publish a non-evaluated Informational doc to be =
published=20
> "with minor updates", i.e., without actual evaluation, as a stds track=20=

> RFC? If the target really remains as PS then the bar should be even=20
> higher than what is described for experimental docs in the above =
process=20
> document, i.e, what we have followod for experimental to be moved to =
stds=20
> track.
>=20
> The only justification that I have heard has beed "because CUBIC has =
long=20
> and wide deployment experience" and "the Internet has not smelted or =
that=20
> "we should have noticed if there were problems". We must, however,=20
> understand that in order to have noticeable bad impact CUBIC should =
cause=20
> some sort of congestion collapse. Congestion collapse, however, is not =
an=20
> issue with CUBIC nor with any other CC algo that applies an RTO =
mechanisms=20
> together with correctly implemented Karn's algo that retains the=20
> backed-off RTO until an Ack is received for a new (not rexmitted) data=20=

> packet. The issue is fairness to competing traffic. This cannot be=20
> observed by deploying and measuring the performance and behaviour of =
CUBIC=20
> alone. CUBIC being more aggressive than current stds track TCP CC =
would=20
> just gives good performance results that one running CUBIC would be =
happy=20
> with. One must evaluate CUBIC's impact on the competing (Reno CC) =
traffic=20
> in range of environments which requires carefully designed active=20
> measurements with thoroughly-analyzed results (as required by the =
above=20
> process document, RFC 5033 and RFC 2914). What we seem to be missing =
is=20
> this evidence on CUBIC's impact and that is something the IETF must =
focus=20
> on, not just that whether CUBIC can achieve better performance than =
other=20
> existing CCs. The latter has been shown in many publications and is =
the=20
> majos focus in  many scientific papers proposing new algos.
> I appreciate a lot that CUBIC has been implemented/developped and=20
> deployed for long and I wonder whether those deploying CUBIC have=20
> unpublished results the wg could review before taking the decicion?
>=20
> I suggest everyone to read carefully RFC 2914 Sec 3.2 and particularly=20=

> what it says about more aggressive (than RFC 5681) congestion control=20=

> algorithms:
>=20
>   Some of these may fail to implement
>   the TCP congestion avoidance mechanisms correctly because of poor
>   implementation [RFC2525].  Others may deliberately be implemented
>   with congestion avoidance algorithms that are more aggressive in
>   their use of bandwidth than other TCP implementations; this would
>   allow a vendor to claim to have a "faster TCP".  The logical
>   consequence of such implementations would be a spiral of =
increasingly
>   aggressive TCP implementations, or increasingly aggressive transport
>   protocols, leading back to the point where there is effectively no
>   congestion avoidance and the Internet is chronically congested.
>=20
> And:
>=20
>   It is convenient to divide flows into three classes: (1) TCP-
>   compatible flows, (2) unresponsive flows, i.e., flows that do not
>   slow down when congestion occurs, and (3) flows that are responsive
>   but are not TCP-compatible.  The last two classes contain more
>   aggressive flows that pose significant threats to Internet
>   performance,
>=20
> As I have tried to point out there are several features with CUBIC =
where=20
> it is likely to be (or to me it seems it obviously is) more aggressive=20=

> than what is reguired to be TCP-compatible. I'm not aware of evidince=20=

> presented to tcpm (or IETF/IRTF) which shows opposite (and I happy to =
be=20
> educated what I have missed).
>=20
> You may take my comments to be a part of the expert review phase=20
> performed by the IRTF/ICCRG for CUBIC. I'm not requesting to modify =
this=20
> doc to CUBIC++ (or something) but it seems to be that this would be=20
> necessary if this doc intends to become published as PS. For =
experimental,=20
> I think it would need some addtioinal updates and record the areas=20
> uncertainty and where more experimentation (clearly) is required.
>=20
> Hi Markku,
>=20
> Thanks for your careful argument above. Given your argument, in order =
to continue to make constructive progress on rfc8312bis without blocking =
for new research and experimentation, I would propose that we change =
rfc8312bis so that it remains "Informational" (like RFC8312) rather than =
the "Standards Track" status currently listed at:
>=20
>   https://ntap.github.io/rfc8312bis/draft-ietf-tcpm-rfc8312bis.html =
<https://ntap.github.io/rfc8312bis/draft-ietf-tcpm-rfc8312bis.html>
>=20
> As far as I can tell, that would allow all of the valuable =
contributions in rfc8312bis to carry forward and be published, while =
still respecting all of the considerations you enumerated above, which =
center around concerns with calling CUBIC "standards track".
>=20
> How does that sound?
>=20
> best,
> neal
>=20
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org <mailto:tcpm@ietf.org>
> https://www.ietf.org/mailman/listinfo/tcpm =
<https://www.ietf.org/mailman/listinfo/tcpm>

--Apple-Mail=_3BE016BB-D9BB-426B-B120-37EE5ACE0369
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=utf-8

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3Dutf-8"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: after-white-space;" class=3D""><div =
class=3D"">I have a general question for the WG, are the "current =
practices" defined in&nbsp;<a =
href=3D"https://datatracker.ietf.org/doc/html/rfc2914" =
class=3D"">https://datatracker.ietf.org/doc/html/rfc2914</a>&nbsp;still =
current? We are 22 years ahead of the published date of this draft and =
there have been magnitude of advancements in congestion control =
algorithms at the hosts and what is the right way to measure conformance =
for a congestion control. A very interesting paper about how to quantify =
a new congestion control, which most might have read, <font =
face=3D"HelveticaNeue" class=3D"">"<b class=3D"">Beyond Jain's Fairness =
Index: Setting the Bar For The Deployment of Congestion Control =
Algorithms"</b></font></div><div class=3D""><br class=3D""></div><div =
class=3D"">I skimmed through RFC 2914 and it references obsoleted drafts =
for what it calls conformant TCP. Like below:</div><div class=3D""><br =
class=3D""></div><div class=3D""><pre class=3D"newpage" =
style=3D"font-size: 13.333333015441895px; margin-top: 0px; =
margin-bottom: 0px; break-before: page; caret-color: rgb(0, 0, 0); =
color: rgb(0, 0, 0);">For TCP, this
   means congestion control algorithms conformant with the current TCP
   specification [RFC793, <a =
href=3D"https://datatracker.ietf.org/doc/html/rfc1122" =
class=3D"">RFC1122</a>, <a =
href=3D"https://datatracker.ietf.org/doc/html/rfc2581" =
class=3D"">RFC2581</a>].</pre><div class=3D""><br =
class=3D""></div></div><div class=3D""><br class=3D""></div><div =
class=3D"">I don=E2=80=99t think we can continue to use old RFCs as a =
gold standard although we can use some the text as is if it still holds =
true.</div><div class=3D""><br class=3D""></div><div =
class=3D""><blockquote type=3D"cite" class=3D""><div dir=3D"ltr" =
class=3D""><div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" =
style=3D"margin: 0px 0px 0px 0.8ex; border-left-width: 1px; =
border-left-style: solid; border-left-color: rgb(204, 204, 204); =
padding-left: 1ex;">&nbsp; Some of these may fail to implement<br =
class=3D"">&nbsp;&nbsp;the TCP congestion avoidance mechanisms correctly =
because of poor<br class=3D"">&nbsp;&nbsp;implementation =
[RFC2525].&nbsp; Others may deliberately be implemented<br =
class=3D"">&nbsp;&nbsp;with congestion avoidance algorithms that are =
more aggressive in<br class=3D"">&nbsp;&nbsp;their use of bandwidth than =
other TCP implementations; this would<br class=3D"">&nbsp;&nbsp;allow a =
vendor to claim to have a "faster TCP".&nbsp; The logical<br =
class=3D"">&nbsp;&nbsp;consequence of such implementations would be a =
spiral of increasingly<br class=3D"">&nbsp;&nbsp;aggressive TCP =
implementations, or increasingly aggressive transport<br =
class=3D"">&nbsp;&nbsp;protocols, leading back to the point where there =
is effectively no<br class=3D"">&nbsp;&nbsp;congestion avoidance and the =
Internet is chronically =
congested.</blockquote></div></div></blockquote></div><div class=3D"">For =
example, in this text quoted by Markku, I agree that we shouldn=E2=80=99t =
end up with algorithms that eventually lead to a chronically congested =
networks but at the same time, if a more aggressive approach provides =
better efficiency by ramping up a bit faster without leading to more =
congestion, then that shouldn=E2=80=99t be discouraged. Faster TCP =
doesn=E2=80=99t necessarily mean that will cause congestion, it could =
just mean, instead of waiting 100s of RTTs to get to full link =
utilization, it does so in much lesser number of RTTs.</div><div =
class=3D""><br class=3D""></div><div class=3D"">In using Reno as a gold =
standard, we tend to forget that Reno is probably the cause of buffer =
bloat in the network. It basically requires network queues to be =
configured double the capacity in order to continue to have full link =
utilization after 50% reduction.</div><div class=3D"">And again, this =
might not have been evident at the time Reno was developed. We have =
realized these problems with time and we should move away from =
continuing to use Reno as the only acceptable Proposed Stds track =
congestion control.</div><div class=3D""><br class=3D""></div><div =
class=3D""><br class=3D""></div><div class=3D""><div =
class=3D""><blockquote type=3D"cite" class=3D""><div dir=3D"ltr" =
class=3D""><div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" =
style=3D"margin: 0px 0px 0px 0.8ex; border-left-width: 1px; =
border-left-style: solid; border-left-color: rgb(204, 204, 204); =
padding-left: 1ex;">&nbsp; It is convenient to divide flows into three =
classes: (1) TCP-<br class=3D"">&nbsp;&nbsp;compatible flows, (2) =
unresponsive flows, i.e., flows that do not<br class=3D"">&nbsp;&nbsp;slow=
 down when congestion occurs, and (3) flows that are responsive<br =
class=3D"">&nbsp;&nbsp;but are not TCP-compatible.&nbsp; The last two =
classes contain more<br class=3D"">&nbsp;&nbsp;aggressive flows that =
pose significant threats to Internet<br =
class=3D"">&nbsp;&nbsp;performance,</blockquote></div></div></blockquote><=
/div></div><div class=3D""><div dir=3D"ltr" class=3D""><div =
class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margin: =
0px 0px 0px 0.8ex; border-left-width: 1px; border-left-style: solid; =
border-left-color: rgb(204, 204, 204); padding-left: 1ex;"><br =
class=3D""></blockquote></div></div></div><div class=3D"">This second =
paragraph is also outdated. IIUC, TCP-compatible flows is equivalent to =
the RFC 2581 which is obsolete. I think we need to replace =
TCP-compatible with something else for two reasons:</div><div =
class=3D""><br class=3D""></div><div class=3D"">1. TCP is not the only =
transport protocol out there. What about QUIC, SCTP?</div><div =
class=3D"">2. Congestion control algorithms should be evaluated in the =
same way for different transport protocols. So, a better word could be, =
Conformant congestion control, one which once accepted as Stds Track =
should just work for any transport protocol.</div><div class=3D""><br =
class=3D""></div><div class=3D"">Perhaps, we need to show more =
experimental / deployment data for Cubic to become Stds Track but we =
shouldn=E2=80=99t rely on RFC2914 from the year 2000 to decide the =
acceptance criteria. If we continue to do so, then even 100 years from =
now, RFC 5681 will be the only TCP-conformant / TCP-compatible =
congestion control as defined by RFC 2914 and the one and only Stds =
Track congestion control to ever exist. (Please let me know if I missed =
any other Stds Track congestion control)</div><div class=3D""><br =
class=3D""></div><div class=3D""><br class=3D""></div><div =
class=3D"">Thanks,</div><div class=3D"">Vidhi</div><div class=3D""><br =
class=3D""></div><div><br class=3D""><blockquote type=3D"cite" =
class=3D""><div class=3D"">On Feb 17, 2022, at 8:13 AM, Neal Cardwell =
&lt;<a href=3D"mailto:ncardwell=3D40google.com@dmarc.ietf.org" =
class=3D"">ncardwell=3D40google.com@dmarc.ietf.org</a>&gt; =
wrote:</div><br class=3D"Apple-interchange-newline"><div class=3D""><meta =
charset=3D"UTF-8" class=3D""><div dir=3D"ltr" style=3D"caret-color: =
rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: =
normal; font-variant-caps: normal; font-weight: normal; letter-spacing: =
normal; text-align: start; text-indent: 0px; text-transform: none; =
white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; =
text-decoration: none;" class=3D""><br =
class=3D"Apple-interchange-newline"><br class=3D""><div =
class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Thu, Feb =
17, 2022 at 9:42 AM Markku Kojo &lt;kojo=3D<a =
href=3D"mailto:40cs.helsinki.fi@dmarc.ietf.org" =
class=3D"">40cs.helsinki.fi@dmarc.ietf.org</a>&gt; wrote:<br =
class=3D""></div><blockquote class=3D"gmail_quote" style=3D"margin: 0px =
0px 0px 0.8ex; border-left-width: 1px; border-left-style: solid; =
border-left-color: rgb(204, 204, 204); padding-left: 1ex;">Hi Yoshi,<br =
class=3D""><br class=3D"">On Tue, 15 Feb 2022, Yoshifumi Nishida =
wrote:<br class=3D""><br class=3D"">&gt; Hi&nbsp;Markku,<br =
class=3D"">&gt;<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">&gt; Thanks for the comments. I think these are very valid =
points.&nbsp;<br class=3D"">&gt; However, I would like to check several =
things as a co-chair and a doc shepherd before we<br class=3D"">&gt; =
discuss&nbsp;the points you've&nbsp;raised.<br class=3D"">&gt;<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">&gt; In my =
understanding (please correct me if I'm wrong), when this draft was =
adopted as an WG<br class=3D"">&gt; item, I think the goal of the doc =
was some minor updates from RFC8312 which include more<br class=3D"">&gt; =
clarifications, minor&nbsp;changes and bug fixes.&nbsp;<br class=3D"">&gt;=
 However, if we try to address your&nbsp;concerns, I think we'll need to =
invent a new version of<br class=3D"">&gt; CUBIC something like CUBIC++ =
or NewCUBIC in the end.&nbsp;<br class=3D"">&gt; I won't deny the value =
of such doc, but, this seems not to be what we agreed on<br =
class=3D"">&gt; beforehand.&nbsp;&nbsp;<br class=3D"">&gt; if we proceed =
in this direction, I think we will need to check the WG consensus =
whether<br class=3D"">&gt; this should be a new goal for the doc.<br =
class=3D"">&gt;<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">&gt; So, I would like to check if this is what you intend for =
the doc or you think we can<br class=3D"">&gt; address your points while =
aligning with the original goal.<br class=3D"">&gt; Also, if someone has =
opinions on this, please share.<br class=3D""><br class=3D"">I think it =
is important that we remember the status of RFC 8312 and the<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">decades long =
process that has been followed in tsv area for new<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">TCP =
congestion control algorithms that have been proposed and submitted<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">to IETF. In =
order to ensure that new cc algos are safe and fair, the<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">process that =
has been followed for all current stds track TCP cc algos<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">has required =
that the cc algo is first accepted and published as<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">experimental =
RFC and only once enough supportive experimental evidence<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">has been =
gathered the doc has become a candidate to be forwaded to stds<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">track. We =
have even agreed on a relatively strict evaluation process to<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">follow when =
cc algos are brought to the IETF to be published as<span =
class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">experimental:<br class=3D""><br class=3D""><a =
href=3D"https://www.ietf.org/about/groups/iesg/statements/experimental-con=
gestion-control/" rel=3D"noreferrer" target=3D"_blank" =
class=3D"">https://www.ietf.org/about/groups/iesg/statements/experimental-=
congestion-control/</a><br class=3D""><br class=3D"">RFC 8312 was =
published as "Informational" and if I recall correctly the<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">idea was =
"just to publish what's out there" for the benefit of the<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">community. =
RFC 8312 was never really evaluated, particularly not in the<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">way new cc =
algos are supposed to be as per the agreed process.<br class=3D""><br =
class=3D"">I do not recall what/how exactly was agreed when rfc8312bis =
was launched<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">but I would be very interested to hear the justification why =
this doc<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">does not need to follow the process mentioned above but we =
would like to<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">propose IETF to publish a non-evaluated Informational doc to =
be published<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">"with minor updates", i.e., without actual evaluation, as a =
stds track<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">RFC? If the target really remains as PS then the bar should =
be even<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">higher than what is described for experimental docs in the =
above process<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">document, i.e, what we have followod for experimental to be =
moved to stds<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">track.<br class=3D""><br class=3D"">The only justification =
that I have heard has beed "because CUBIC has long<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">and wide =
deployment experience" and "the Internet has not smelted or that<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">"we should =
have noticed if there were problems". We must, however,<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">understand =
that in order to have noticeable bad impact CUBIC should cause<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">some sort of =
congestion collapse. Congestion collapse, however, is not an<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">issue with =
CUBIC nor with any other CC algo that applies an RTO mechanisms<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">together =
with correctly implemented Karn's algo that retains the<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">backed-off =
RTO until an Ack is received for a new (not rexmitted) data<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">packet. The =
issue is fairness to competing traffic. This cannot be<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">observed by =
deploying and measuring the performance and behaviour of CUBIC<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">alone. CUBIC =
being more aggressive than current stds track TCP CC would<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">just gives =
good performance results that one running CUBIC would be happy<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">with. One =
must evaluate CUBIC's impact on the competing (Reno CC) traffic<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">in range of =
environments which requires carefully designed active<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">measurements =
with thoroughly-analyzed results (as required by the above<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">process =
document, RFC 5033 and RFC 2914). What we seem to be missing is<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">this =
evidence on CUBIC's impact and that is something the IETF must =
focus<span class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">on,=
 not just that whether CUBIC can achieve better performance than =
other<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">existing CCs. The latter has been shown in many publications =
and is the<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">majos focus in&nbsp; many scientific papers proposing new =
algos.<br class=3D"">I appreciate a lot that CUBIC has been =
implemented/developped and<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">deployed for =
long and I wonder whether those deploying CUBIC have<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">unpublished =
results the wg could review before taking the decicion?<br class=3D""><br =
class=3D"">I suggest everyone to read carefully RFC 2914 Sec 3.2 and =
particularly<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">what it says about more aggressive (than RFC 5681) congestion =
control<span class=3D"Apple-converted-space">&nbsp;</span><br =
class=3D"">algorithms:<br class=3D""><br class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>Some of these may fail to =
implement<br class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>the TCP congestion =
avoidance mechanisms correctly because of poor<br class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>implementation =
[RFC2525].&nbsp; Others may deliberately be implemented<br =
class=3D"">&nbsp;<span class=3D"Apple-converted-space">&nbsp;</span>with =
congestion avoidance algorithms that are more aggressive in<br =
class=3D"">&nbsp;<span class=3D"Apple-converted-space">&nbsp;</span>their =
use of bandwidth than other TCP implementations; this would<br =
class=3D"">&nbsp;<span class=3D"Apple-converted-space">&nbsp;</span>allow =
a vendor to claim to have a "faster TCP".&nbsp; The logical<br =
class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>consequence of such =
implementations would be a spiral of increasingly<br =
class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>aggressive TCP =
implementations, or increasingly aggressive transport<br =
class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>protocols, leading back to =
the point where there is effectively no<br class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>congestion avoidance and =
the Internet is chronically congested.<br class=3D""><br =
class=3D"">And:<br class=3D""><br class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>It is convenient to divide =
flows into three classes: (1) TCP-<br class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>compatible flows, (2) =
unresponsive flows, i.e., flows that do not<br class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>slow down when congestion =
occurs, and (3) flows that are responsive<br class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>but are not =
TCP-compatible.&nbsp; The last two classes contain more<br =
class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>aggressive flows that pose =
significant threats to Internet<br class=3D"">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>performance,<br =
class=3D""><br class=3D"">As I have tried to point out there are several =
features with CUBIC where<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">it is likely =
to be (or to me it seems it obviously is) more aggressive<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">than what is =
reguired to be TCP-compatible. I'm not aware of evidince<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">presented to =
tcpm (or IETF/IRTF) which shows opposite (and I happy to be<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">educated =
what I have missed).<br class=3D""><br class=3D"">You may take my =
comments to be a part of the expert review phase<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">performed by =
the IRTF/ICCRG for CUBIC. I'm not requesting to modify this<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">doc to =
CUBIC++ (or something) but it seems to be that this would be<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">necessary if =
this doc intends to become published as PS. For experimental,<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">I think it =
would need some addtioinal updates and record the areas<span =
class=3D"Apple-converted-space">&nbsp;</span><br class=3D"">uncertainty =
and where more experimentation (clearly) is required.<br =
class=3D""></blockquote><div class=3D""><br class=3D""></div><div =
class=3D"">Hi Markku,</div><div class=3D""><br class=3D""></div>Thanks =
for your careful argument above. Given your argument, in order to =
continue to make constructive progress on rfc8312bis without blocking =
for new research and experimentation, I would propose that we change =
rfc8312bis so that it remains "Informational" (like RFC8312) rather than =
the "Standards Track" status currently listed at:<br class=3D""><br =
class=3D""></div><div class=3D"gmail_quote">&nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span><a =
href=3D"https://ntap.github.io/rfc8312bis/draft-ietf-tcpm-rfc8312bis.html"=
 =
class=3D"">https://ntap.github.io/rfc8312bis/draft-ietf-tcpm-rfc8312bis.ht=
ml</a><br class=3D""><br class=3D"">As far as I can tell, that would =
allow all of the valuable contributions in rfc8312bis to carry forward =
and be published, while still respecting all of the considerations you =
enumerated above, which center around concerns with calling CUBIC =
"standards track".</div><div class=3D"gmail_quote"><br class=3D"">How =
does that sound?<br class=3D""><br class=3D"">best,<br class=3D"">neal<div=
 class=3D""><br class=3D""></div></div></div><span style=3D"caret-color: =
rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: =
normal; font-variant-caps: normal; font-weight: normal; letter-spacing: =
normal; text-align: start; text-indent: 0px; text-transform: none; =
white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; =
text-decoration: none; float: none; display: inline !important;" =
class=3D"">_______________________________________________</span><br =
style=3D"caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: =
12px; font-style: normal; font-variant-caps: normal; font-weight: =
normal; letter-spacing: normal; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; word-spacing: 0px; =
-webkit-text-stroke-width: 0px; text-decoration: none;" class=3D""><span =
style=3D"caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: =
12px; font-style: normal; font-variant-caps: normal; font-weight: =
normal; letter-spacing: normal; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; word-spacing: 0px; =
-webkit-text-stroke-width: 0px; text-decoration: none; float: none; =
display: inline !important;" class=3D"">tcpm mailing list</span><br =
style=3D"caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: =
12px; font-style: normal; font-variant-caps: normal; font-weight: =
normal; letter-spacing: normal; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; word-spacing: 0px; =
-webkit-text-stroke-width: 0px; text-decoration: none;" class=3D""><a =
href=3D"mailto:tcpm@ietf.org" style=3D"font-family: Helvetica; =
font-size: 12px; font-style: normal; font-variant-caps: normal; =
font-weight: normal; letter-spacing: normal; orphans: auto; text-align: =
start; text-indent: 0px; text-transform: none; white-space: normal; =
widows: auto; word-spacing: 0px; -webkit-text-size-adjust: auto; =
-webkit-text-stroke-width: 0px;" class=3D"">tcpm@ietf.org</a><br =
style=3D"caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: =
12px; font-style: normal; font-variant-caps: normal; font-weight: =
normal; letter-spacing: normal; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; word-spacing: 0px; =
-webkit-text-stroke-width: 0px; text-decoration: none;" class=3D""><a =
href=3D"https://www.ietf.org/mailman/listinfo/tcpm" style=3D"font-family: =
Helvetica; font-size: 12px; font-style: normal; font-variant-caps: =
normal; font-weight: normal; letter-spacing: normal; orphans: auto; =
text-align: start; text-indent: 0px; text-transform: none; white-space: =
normal; widows: auto; word-spacing: 0px; -webkit-text-size-adjust: auto; =
-webkit-text-stroke-width: 0px;" =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm</a></div></blockquot=
e></div><br class=3D""></body></html>=

--Apple-Mail=_3BE016BB-D9BB-426B-B120-37EE5ACE0369--

--Apple-Mail=_CE9EFBDC-2E4C-43DC-A4CA-6676BD14A505
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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=
--Apple-Mail=_CE9EFBDC-2E4C-43DC-A4CA-6676BD14A505--


From nobody Fri Feb 18 00:40:04 2022
Return-Path: <nsd.ietf@gmail.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B40733A0AE9; Fri, 18 Feb 2022 00:40:01 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.096
X-Spam-Level: 
X-Spam-Status: No, score=-2.096 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Jv5nGxiUd9yu; Fri, 18 Feb 2022 00:39:57 -0800 (PST)
Received: from mail-qv1-xf35.google.com (mail-qv1-xf35.google.com [IPv6:2607:f8b0:4864:20::f35]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C0CD53A0AF2; Fri, 18 Feb 2022 00:39:56 -0800 (PST)
Received: by mail-qv1-xf35.google.com with SMTP id n6so13382442qvk.13; Fri, 18 Feb 2022 00:39:56 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112;  h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=s6rl2oQgIdfU3LfGfvfAg/EzgB9Q+zjFqScALqElklk=; b=gjU/kD/rKgoknJ4AsbeV7+Psu0O0h197TRBXoFSDjmVfh4RvrFHhc8SWcp57dsq3wJ g39JIYaWX6VIP4LKqYPGOSSlMbsP8COKIRsXpgZAdU/AjCaQ5yG0Rn+2xFAOIZR2tMF3 2pUKsdCBQ0+Coc+QYwFISueQHMjvO+0XTPmOFw4K4Klg0e4/k+RyXCp4Tsv2M7gjT2C6 RIh5cCABPpjxPaCZQRMbqQ2DMTXsHPtzkQhVbOK8iyA9zu303ucOutFNDDHEvJ/cJbRS 7gPPP+VldP7AYXLVQPWtvfN9oSVtPTRYYiM2iQWkrOmLU7jmmYZ21d8hgQx4f3CdYRuw 15KQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=s6rl2oQgIdfU3LfGfvfAg/EzgB9Q+zjFqScALqElklk=; b=RPBGPpmGICkcUm3P5VxT61LJu73OPpBBN0r9rCdQfSA7oxk1LtjEYgJgq/RqgofFj3 s03hcoK5BJCJnupbcVy8JGfw52DBVCVUSEg7nBE33VsniHOSLxoprULFqB4+GpVmBn7r frA/Z7YEqGJCliDeqEbztCtiAZ43yfPRBo9CW1ynGWbdxMhGPY1zk66khu2MU95xl45V MEKGrIvPuvubQHz13vNVT2WGIDiuoN7oh0tnqGfwfVDFTH4rFlBzCW8Kvzm7tCzmgCGI 1jXjf7NqF4H/qMoFOmX5mmnVowOOgUAq7MJeoJa5CDsBm+/XKKq9hKR54nn6hVwg1ea/ dw8g==
X-Gm-Message-State: AOAM532ryT2GDMSfXBXChR3ymO89pkrTdl04LwqdRnCed2m5jwVz9J3F Otmt2J7EBtvcE7CJ0uO2e5emfNaxZWy6R1R3bzcajMa4wzQ=
X-Google-Smtp-Source: ABdhPJxwBsVnI18JwX8gRyaq1QWP89rsxqhWf8X96utHzaIAh1OLIrlBnj0+/unWuMCy/2nV1ecf/iMvGL3VNqipOqI=
X-Received: by 2002:ac8:5b91:0:b0:2db:acb7:2412 with SMTP id a17-20020ac85b91000000b002dbacb72412mr5735361qta.475.1645173595435; Fri, 18 Feb 2022 00:39:55 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi>
In-Reply-To: <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi>
From: Yoshifumi Nishida <nsd.ietf@gmail.com>
Date: Fri, 18 Feb 2022 00:39:44 -0800
Message-ID: <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com>
To: Markku Kojo <kojo@cs.helsinki.fi>
Cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: multipart/alternative; boundary="000000000000bff3b305d846d3ac"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/36764w6ZEekT8KHljPetsLQq7T4>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 18 Feb 2022 08:40:02 -0000

--000000000000bff3b305d846d3ac
Content-Type: text/plain; charset="UTF-8"

Hi Markku,

Thanks for the comments. Yes, these are valid and important concerns.
However, I have mixed feelings about them.
I don't want to open a can of worms, but here's are some of my personally
thoughts...

One thing is that "long and wide deployment experience" might not be a very
strong claim, but then, I am wondering what would be the strong one.
One may bring some solid experiment results, but the others might argue it
uses too convenient configurations, or the scale is too small or too short.

The other thing is that if we set a high bar for a PS doc related to CC, we
should apply the same bar to all related docs.
For example, I am personally wondering how much we have analyzed and
performed experiments when we published RFC9002. (BTW, my intention is NOT
to accuse a specific doc, there should be some more examples.)
I am not sure how we can maintain this kind of fairness between the docs at
the moment.

Thanks,
--
Yoshi


On Thu, Feb 17, 2022 at 6:41 AM Markku Kojo <kojo@cs.helsinki.fi> wrote:

> Hi Yoshi,
>
> On Tue, 15 Feb 2022, Yoshifumi Nishida wrote:
>
> > Hi Markku,
> >
> > Thanks for the comments. I think these are very valid points.
> > However, I would like to check several things as a co-chair and a doc
> shepherd before we
> > discuss the points you've raised.
> >
> > In my understanding (please correct me if I'm wrong), when this draft
> was adopted as an WG
> > item, I think the goal of the doc was some minor updates from RFC8312
> which include more
> > clarifications, minor changes and bug fixes.
> > However, if we try to address your concerns, I think we'll need to
> invent a new version of
> > CUBIC something like CUBIC++ or NewCUBIC in the end.
> > I won't deny the value of such doc, but, this seems not to be what we
> agreed on
> > beforehand.
> > if we proceed in this direction, I think we will need to check the WG
> consensus whether
> > this should be a new goal for the doc.
> >
> > So, I would like to check if this is what you intend for the doc or you
> think we can
> > address your points while aligning with the original goal.
> > Also, if someone has opinions on this, please share.
>
> I think it is important that we remember the status of RFC 8312 and the
> decades long process that has been followed in tsv area for new
> TCP congestion control algorithms that have been proposed and submitted
> to IETF. In order to ensure that new cc algos are safe and fair, the
> process that has been followed for all current stds track TCP cc algos
> has required that the cc algo is first accepted and published as
> experimental RFC and only once enough supportive experimental evidence
> has been gathered the doc has become a candidate to be forwaded to stds
> track. We have even agreed on a relatively strict evaluation process to
> follow when cc algos are brought to the IETF to be published as
> experimental:
>
>
> https://www.ietf.org/about/groups/iesg/statements/experimental-congestion-control/
>
> RFC 8312 was published as "Informational" and if I recall correctly the
> idea was "just to publish what's out there" for the benefit of the
> community. RFC 8312 was never really evaluated, particularly not in the
> way new cc algos are supposed to be as per the agreed process.
>
> I do not recall what/how exactly was agreed when rfc8312bis was launched
> but I would be very interested to hear the justification why this doc
> does not need to follow the process mentioned above but we would like to
> propose IETF to publish a non-evaluated Informational doc to be published
> "with minor updates", i.e., without actual evaluation, as a stds track
> RFC? If the target really remains as PS then the bar should be even
> higher than what is described for experimental docs in the above process
> document, i.e, what we have followod for experimental to be moved to stds
> track.
>
> The only justification that I have heard has beed "because CUBIC has long
> and wide deployment experience" and "the Internet has not smelted or that
> "we should have noticed if there were problems". We must, however,
> understand that in order to have noticeable bad impact CUBIC should cause
> some sort of congestion collapse. Congestion collapse, however, is not an
> issue with CUBIC nor with any other CC algo that applies an RTO mechanisms
> together with correctly implemented Karn's algo that retains the
> backed-off RTO until an Ack is received for a new (not rexmitted) data
> packet. The issue is fairness to competing traffic. This cannot be
> observed by deploying and measuring the performance and behaviour of CUBIC
> alone. CUBIC being more aggressive than current stds track TCP CC would
> just gives good performance results that one running CUBIC would be happy
> with. One must evaluate CUBIC's impact on the competing (Reno CC) traffic
> in range of environments which requires carefully designed active
> measurements with thoroughly-analyzed results (as required by the above
> process document, RFC 5033 and RFC 2914). What we seem to be missing is
> this evidence on CUBIC's impact and that is something the IETF must focus
> on, not just that whether CUBIC can achieve better performance than other
> existing CCs. The latter has been shown in many publications and is the
> majos focus in  many scientific papers proposing new algos.
> I appreciate a lot that CUBIC has been implemented/developped and
> deployed for long and I wonder whether those deploying CUBIC have
> unpublished results the wg could review before taking the decicion?
>
> I suggest everyone to read carefully RFC 2914 Sec 3.2 and particularly
> what it says about more aggressive (than RFC 5681) congestion control
> algorithms:
>
>   Some of these may fail to implement
>   the TCP congestion avoidance mechanisms correctly because of poor
>   implementation [RFC2525].  Others may deliberately be implemented
>   with congestion avoidance algorithms that are more aggressive in
>   their use of bandwidth than other TCP implementations; this would
>   allow a vendor to claim to have a "faster TCP".  The logical
>   consequence of such implementations would be a spiral of increasingly
>   aggressive TCP implementations, or increasingly aggressive transport
>   protocols, leading back to the point where there is effectively no
>   congestion avoidance and the Internet is chronically congested.
>
> And:
>
>   It is convenient to divide flows into three classes: (1) TCP-
>   compatible flows, (2) unresponsive flows, i.e., flows that do not
>   slow down when congestion occurs, and (3) flows that are responsive
>   but are not TCP-compatible.  The last two classes contain more
>   aggressive flows that pose significant threats to Internet
>   performance,
>
> As I have tried to point out there are several features with CUBIC where
> it is likely to be (or to me it seems it obviously is) more aggressive
> than what is reguired to be TCP-compatible. I'm not aware of evidince
> presented to tcpm (or IETF/IRTF) which shows opposite (and I happy to be
> educated what I have missed).
>
> You may take my comments to be a part of the expert review phase
> performed by the IRTF/ICCRG for CUBIC. I'm not requesting to modify this
> doc to CUBIC++ (or something) but it seems to be that this would be
> necessary if this doc intends to become published as PS. For experimental,
> I think it would need some addtioinal updates and record the areas
> uncertainty and where more experimentation (clearly) is required.
>
> Thanks,
>
> /Markku
>
> > Thanks,
> > --
> > Yoshi
> >
> >
> > On Fri, Feb 11, 2022 at 9:34 PM Markku Kojo <kojo@cs.helsinki.fi> wrote:
> >       Hi Yoshi, all,
> >
> >       It seems to me that many issues that I raised have been solved,
> thanks.
> >       However, there are still a number of important issues that have
> not been
> >       addressed adequately. At least the following:
> >
> >       #135 on W_max. Yoshi's observation was correct that this is not
> resolved:
> >       the co-authors and original developpers of CUBIC (@lisongxu and
> >       @sangtaeha) agreed in their last message that Wmax needs different
> >       treatment for slow start and congestion avoidance and plan
> comprehensive
> >       (new) evaluation of it. This is obviously an open issue but the
> issue
> >       was closed?
> >
> >       #85 (& #86 with basically same issue and these two were combined)
> This
> >       (#85) is about ECN but the major issue is on using the same MD
> >       factor in slow start and in congestion avoidance when using
> >       loss-based CC. This (#85) remained closed even though I provided a
> >       thorough explanation why it is wrong and against the original
> theory and
> >       design by Van Jacobson, against the congestion control principles
> (RFC
> >       2914) and two co-authors agreed on this in their same last message
> to
> >       #135 when they agreed on Wmax needing rework. This is an important
> issue
> >       that the wg should consider very carefully because it is not only
> >       updating RFC 5681 but also in conflict with RFC 2914. How can
> >       tcpm (and IETF) suggest and allow one CC algo to not follow
> congestion
> >       control principles as set in RFC 2914 while requiring all other
> CCs to
> >       follow RFC 2914 guidelines?
> >       The current draft does not provide any justification for using the
> same
> >       MD factor in slow start as in congestion avoidance. Nor am I
> >       awere of any experimental data that would support this change.
> >       The fact thet CUBIC has been long deployed does not alone provide
> any
> >       supporting evidence because CUBIC is likely to give good
> performance as
> >       it is overagressive and thereby unfair to competing traffic and
> users
> >       tend to be happy when measuring the performance of the sending
> CUBIC
> >       only, not the competing traffic that is badly impacted. HyStart++
> is
> >       suggested as mitigation to the problem but it cannot; HyStart++ is
> only
> >       applicable during initial slow start, not during slow start after
> RTO!
> >       That is, the "SHOULD use HYStart++" text in Sec 4.10 is impossible
> >       to implement as I have pointed out in my comments earlier. Using a
> proper
> >       MD factor in slow start is even more important if loss is
> detectected
> >       during a RTO recovery because the sender is likely to face heavy
> >       congestion in such a case and it is very bad if the sender
> continues
> >       sending with overaggressive rate, stealing the capacity from and
> causing
> >       harm to coexisting flows. In addition, as I have explained,
> HyStart++
> >       does not remove the problem even for the initial slow start as it
> is not
> >       shown to work always. Instead, the results with the HyStart++
> draft show
> >       that it reduces 50% of rexmits and only 36% RTOs, meaning that
> there is
> >       likely to be a notable percentage of cases when a sender is still
> >       in slow start when first loss is detected (i.e., HyStart++ had no
> effect)
> >       and a significant number of cases where a CUBIC sender is
> >       overaggressive continuing with a 1-40% larger cwnd than what is the
> >       available capacity. Note also that any delay-based heuristics like
> >       HyStart++ are known to work poorly in various wireless environmens
> where
> >       link delay tends to vary a lot. We may come up with some other MD
> factor
> >       that 0.5 when in slow start and HyStart++ is in use, but that is
> >       experimental, if not research, and definitely not ready for stds
> track.
> >
> >       #114, #132, and #143 w.r.t flightsize vs. cwnd. The current text
> >             does not quite correcly reflect what stacks that use cwnd do.
> >             I'll comment in #143 separately.
> >
> >       #96 & #98: The text added does not address the problems raised
> which are
> >       also evidenced in the paper pointed by Bob in #96. Even though
> CUBIC has
> >       been modified a bit after the paper was published, it does not
> >       automatically mean that the problem has been shown resolved:
> experimental
> >       evidence is required but not provided. The fact that CUBIC does not
> >       change MD factor for fast convergence is the root of the problem
> >       evidenced in the paper and remains so in the algo specified in this
> >       draft. This is also a significant problem when competing with Reno
> CC
> >       because CUBIC behaves much more aggressive than Reno CC when there
> is
> >       sudden congestion and all competing flows must converge fast down
> to a
> >       small fraction of the current cwnd to be fair to each other. This
> again
> >       cannot be evidenced not to be a problem by long deployment
> experience
> >       unless experimental data that measures the impact on competing
> traffic is
> >       presented to back the claims. Adjusting just Wmax for fast
> convergence is
> >       not enough and is even likely to be ineffective because there tend
> to be
> >       several losses when sudden congestion is hit, and particularly if
> NewReno
> >       is in use the sender stays several RTTs in Fast Recovery being
> >       overaggressive and then possibly continues at the same rate in CA
> which
> >       is unlikely to reach evan close to Wmax before a new loss hits
> >       the sender again. That is, lower Wmax and lower additive increase
> factor
> >       do not compensate the use of larger MD factor when sudden
> congestion is
> >       encountered.
> >
> >       #93 & #94 & (#89) Sec 5.3 still does not address any difficult
> >       environments, in particular buffer-bloated paths (nor does Sec
> 5.4).
> >       We need evidence (results) that show CUBIC is fair towards other
> CCs
> >       (Reno) also in such environments. Note that CUBIC's decision to
> leave
> >       Reno-friendly region is based on the size of cwnd which tends to be
> >       incorrect with buffer-bloated bottlenecks because with huge
> buffers the
> >       cwnd can be many times larger than what is actually needed to fully
> >       utilize the available network bit-rate. Therefore, Reno CC has no
> problem
> >       in fully utilizing such bottleneck links and CUBIC must stay in
> >       Reno-friendly region longer but it leaves it too early because the
> same C
> >       is used as with non-bloated environments. We lack experiments
> showing
> >       CUBIC follows the congestion control principles and is fair to
> current
> >       standard TCP CC; to my understanding no experiments with
> buffer-bloated
> >       bottlenecks are cited to back up the claims even though buffer
> bloat is
> >       very well known to be a common (difficult) environment in today's
> >       Internet.
> >
> >       #90 The current text on applying undo (a response to detected false
> >       fastrexmit) does not provide correct result if someone implements
> it.
> >       I have explained the problems there in github but seem to have not
> >       replied to latest comments by Neal. I'll reply and try to explain
> more.
> >       Again one major problem here is that the draft suggest a new
> response
> >       algo for false fast rexmits but does not provide any experimental
> data to
> >       support it. Long deployment experience has been suggested as
> >       justification but again without any carefully evaluated
> experimental
> >       data and evidence there is no meat. The issue is important to
> solve but is
> >       not specific to CUBIC. Instead, it is general problem for all TCP
> CC
> >       variants. IMHO, this is not ready for standards track but deserves
> a draft
> >       of its own so that it can be carefully evaluated and discussed ion
> the
> >       list. AFAIK there has been no discussion on this on the tcpm list,
> so
> >       those probably interested and having experience are likely to be
> unaware
> >       that this is part of CUBIC draft.
> >
> >       #88 The problem with correctness of the AIMD model and setting
> alpha for
> >       CUBIC requires further consideration. Bob provided an analysis that
> >       leaves things still open. It seems that I never had time to review
> and
> >       comment the analysis and clarify why the model does not work. I'll
> do that
> >       separately as it is important to ensure CUBIC behaves fairly as
> intended
> >       for the Reno-friendly region.
> >
> >       Best regards,
> >
> >       /Markku
> >
> >       On Mon, 31 Jan 2022, Yoshifumi Nishida wrote:
> >
> >       > Hello,
> >       >
> >       > After some discussions among chairs, we decided to run the 2nd
> WGLC on
> >       draft-ietf-tcpm-rfc8312bis in
> >       > consideration of the importance of the draft.
> >       > We'll be grateful if you could send your feedback to the ML. The
> WGLC runs
> >       until *Feb 11*.
> >       >
> >       > If interested, you can check in-depth past discussions in the
> following URL.
> >       > https://github.com/NTAP/rfc8312bis/
> >       >
> >       > Thank you so much!
> >       > --
> >       > tcpm co-chairs
> >       >
> >       >
> >       > On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <lars@eggert.org>
> wrote:
> >       >       Hi,
> >       >
> >       >       this -06 version rolls in all the changes requested during
> (and after)
> >       WGLC ended.
> >       >
> >       >       I'll leave it up to the chairs to decide if another WGLC
> is warranted
> >       or the document can
> >       >       progress as-is.
> >       >
> >       >       Thanks,
> >       >       Lars
> >       >
> >       >
> >       >       > On 2022-1-26, at 11:12, 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 TCP Maintenance and
> Minor
> >       Extensions WG of the IETF.
> >       >       >
> >       >       >        Title           : CUBIC for Fast and
> Long-Distance Networks
> >       >       >        Authors         : Lisong Xu
> >       >       >                          Sangtae Ha
> >       >       >                          Injong Rhee
> >       >       >                          Vidhi Goel
> >       >       >                          Lars Eggert
> >       >       >       Filename        : draft-ietf-tcpm-rfc8312bis-06.txt
> >       >       >       Pages           : 35
> >       >       >       Date            : 2022-01-26
> >       >       >
> >       >       > Abstract:
> >       >       >   CUBIC is a standard TCP congestion control algorithm
> that uses a
> >       >       >   cubic function instead of a linear congestion window
> increase
> >       >       >   function to improve scalability and stability over
> fast and long-
> >       >       >   distance networks.  CUBIC has been adopted as the
> default TCP
> >       >       >   congestion control algorithm by the Linux, Windows,
> and Apple
> >       stacks.
> >       >       >
> >       >       >   This document updates the specification of CUBIC to
> include
> >       >       >   algorithmic improvements based on these
> implementations and recent
> >       >       >   academic work.  Based on the extensive deployment
> experience with
> >       >       >   CUBIC, it also moves the specification to the
> Standards Track,
> >       >       >   obsoleting RFC 8312.  This also requires updating RFC
> 5681, to
> >       allow
> >       >       >   for CUBIC's occasionally more aggressive sending
> behavior.
> >       >       >
> >       >       >
> >       >       > The IETF datatracker status page for this draft is:
> >       >       >
> https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/
> >       >       >
> >       >       > There is also an HTML version available at:
> >       >       >
> https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html
> >       >       >
> >       >       > A diff from the previous version is available at:
> >       >       >
> https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rfc8312bis-06
> >       >       >
> >       >       >
> >       >       > Internet-Drafts are also available by rsync at
> >       rsync.ietf.org::internet-drafts
> >       >       >
> >       >       >
> >       >       > _______________________________________________
> >       >       > tcpm mailing list
> >       >       > tcpm@ietf.org
> >       >       > https://www.ietf.org/mailman/listinfo/tcpm
> >       >
> >       >       _______________________________________________
> >       >       tcpm mailing list
> >       >       tcpm@ietf.org
> >       >       https://www.ietf.org/mailman/listinfo/tcpm
> >       >
> >       >
> >       >
> >
> >
> >

--000000000000bff3b305d846d3ac
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi Markku,</div><div><br></div><div>Thanks for the co=
mments. Yes, these are valid and important concerns. However, I have mixed =
feelings about them.</div><div>I don&#39;t want to open a can=C2=A0of worms=
, but here&#39;s are some of my personally thoughts...</div><div><br></div>=
<div>One thing is that &quot;long and wide deployment experience&quot; migh=
t not be a very strong claim, but then, I am wondering what would be the st=
rong one.</div><div>One may bring some solid experiment results, but the ot=
hers might argue it uses too convenient configurations, or the scale is too=
 small or too short.</div><div><br></div>The other thing is that if we set =
a high bar for a PS doc related to CC, we should apply the same bar to all =
related docs.<div>For example, I am personally wondering how much we have a=
nalyzed and performed experiments when we published RFC9002. (BTW, my inten=
tion is NOT to accuse a specific doc, there should be some more examples.)=
=C2=A0</div><div>I am not sure how we can maintain this kind of fairness be=
tween the docs at the moment.</div><div><br></div><div>Thanks,</div><div>--=
</div><div>Yoshi=C2=A0</div><div><div><br></div><div><br><div class=3D"gmai=
l_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Thu, Feb 17, 2022 at 6:41=
 AM Markku Kojo &lt;<a href=3D"mailto:kojo@cs.helsinki.fi">kojo@cs.helsinki=
.fi</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"marg=
in:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1e=
x">Hi Yoshi,<br>
<br>
On Tue, 15 Feb 2022, Yoshifumi Nishida wrote:<br>
<br>
&gt; Hi=C2=A0Markku,<br>
&gt; <br>
&gt; Thanks for the comments. I think these are very valid points.=C2=A0<br=
>
&gt; However, I would like to check several things as a co-chair and a doc =
shepherd before we<br>
&gt; discuss=C2=A0the points you&#39;ve=C2=A0raised.<br>
&gt; <br>
&gt; In my understanding (please correct me if I&#39;m wrong), when this dr=
aft was adopted as an WG<br>
&gt; item, I think the goal of the doc was some minor updates from RFC8312 =
which include more<br>
&gt; clarifications, minor=C2=A0changes and bug fixes.=C2=A0<br>
&gt; However, if we try to address your=C2=A0concerns, I think we&#39;ll ne=
ed to invent a new version of<br>
&gt; CUBIC something like CUBIC++ or NewCUBIC in the end.=C2=A0<br>
&gt; I won&#39;t deny the value of such doc, but, this seems not to be what=
 we agreed on<br>
&gt; beforehand.=C2=A0=C2=A0<br>
&gt; if we proceed in this direction, I think we will need to check the WG =
consensus whether<br>
&gt; this should be a new goal for the doc.<br>
&gt; <br>
&gt; So, I would like to check if this is what you intend for the doc or yo=
u think we can<br>
&gt; address your points while aligning with the original goal.<br>
&gt; Also, if someone has opinions on this, please share.<br>
<br>
I think it is important that we remember the status of RFC 8312 and the <br=
>
decades long process that has been followed in tsv area for new <br>
TCP congestion control algorithms that have been proposed and submitted <br=
>
to IETF. In order to ensure that new cc algos are safe and fair, the <br>
process that has been followed for all current stds track TCP cc algos <br>
has required that the cc algo is first accepted and published as <br>
experimental RFC and only once enough supportive experimental evidence <br>
has been gathered the doc has become a candidate to be forwaded to stds <br=
>
track. We have even agreed on a relatively strict evaluation process to <br=
>
follow when cc algos are brought to the IETF to be published as <br>
experimental:<br>
<br>
<a href=3D"https://www.ietf.org/about/groups/iesg/statements/experimental-c=
ongestion-control/" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.o=
rg/about/groups/iesg/statements/experimental-congestion-control/</a><br>
<br>
RFC 8312 was published as &quot;Informational&quot; and if I recall correct=
ly the <br>
idea was &quot;just to publish what&#39;s out there&quot; for the benefit o=
f the <br>
community. RFC 8312 was never really evaluated, particularly not in the <br=
>
way new cc algos are supposed to be as per the agreed process.<br>
<br>
I do not recall what/how exactly was agreed when rfc8312bis was launched <b=
r>
but I would be very interested to hear the justification why this doc <br>
does not need to follow the process mentioned above but we would like to <b=
r>
propose IETF to publish a non-evaluated Informational doc to be published <=
br>
&quot;with minor updates&quot;, i.e., without actual evaluation, as a stds =
track <br>
RFC? If the target really remains as PS then the bar should be even <br>
higher than what is described for experimental docs in the above process <b=
r>
document, i.e, what we have followod for experimental to be moved to stds <=
br>
track.<br>
<br>
The only justification that I have heard has beed &quot;because CUBIC has l=
ong <br>
and wide deployment experience&quot; and &quot;the Internet has not smelted=
 or that <br>
&quot;we should have noticed if there were problems&quot;. We must, however=
, <br>
understand that in order to have noticeable bad impact CUBIC should cause <=
br>
some sort of congestion collapse. Congestion collapse, however, is not an <=
br>
issue with CUBIC nor with any other CC algo that applies an RTO mechanisms =
<br>
together with correctly implemented Karn&#39;s algo that retains the <br>
backed-off RTO until an Ack is received for a new (not rexmitted) data <br>
packet. The issue is fairness to competing traffic. This cannot be <br>
observed by deploying and measuring the performance and behaviour of CUBIC =
<br>
alone. CUBIC being more aggressive than current stds track TCP CC would <br=
>
just gives good performance results that one running CUBIC would be happy <=
br>
with. One must evaluate CUBIC&#39;s impact on the competing (Reno CC) traff=
ic <br>
in range of environments which requires carefully designed active <br>
measurements with thoroughly-analyzed results (as required by the above <br=
>
process document, RFC 5033 and RFC 2914). What we seem to be missing is <br=
>
this evidence on CUBIC&#39;s impact and that is something the IETF must foc=
us <br>
on, not just that whether CUBIC can achieve better performance than other <=
br>
existing CCs. The latter has been shown in many publications and is the <br=
>
majos focus in=C2=A0 many scientific papers proposing new algos.<br>
I appreciate a lot that CUBIC has been implemented/developped and <br>
deployed for long and I wonder whether those deploying CUBIC have <br>
unpublished results the wg could review before taking the decicion?<br>
<br>
I suggest everyone to read carefully RFC 2914 Sec 3.2 and particularly <br>
what it says about more aggressive (than RFC 5681) congestion control <br>
algorithms:<br>
<br>
=C2=A0 Some of these may fail to implement<br>
=C2=A0 the TCP congestion avoidance mechanisms correctly because of poor<br=
>
=C2=A0 implementation [RFC2525].=C2=A0 Others may deliberately be implement=
ed<br>
=C2=A0 with congestion avoidance algorithms that are more aggressive in<br>
=C2=A0 their use of bandwidth than other TCP implementations; this would<br=
>
=C2=A0 allow a vendor to claim to have a &quot;faster TCP&quot;.=C2=A0 The =
logical<br>
=C2=A0 consequence of such implementations would be a spiral of increasingl=
y<br>
=C2=A0 aggressive TCP implementations, or increasingly aggressive transport=
<br>
=C2=A0 protocols, leading back to the point where there is effectively no<b=
r>
=C2=A0 congestion avoidance and the Internet is chronically congested.<br>
<br>
And:<br>
<br>
=C2=A0 It is convenient to divide flows into three classes: (1) TCP-<br>
=C2=A0 compatible flows, (2) unresponsive flows, i.e., flows that do not<br=
>
=C2=A0 slow down when congestion occurs, and (3) flows that are responsive<=
br>
=C2=A0 but are not TCP-compatible.=C2=A0 The last two classes contain more<=
br>
=C2=A0 aggressive flows that pose significant threats to Internet<br>
=C2=A0 performance,<br>
<br>
As I have tried to point out there are several features with CUBIC where <b=
r>
it is likely to be (or to me it seems it obviously is) more aggressive <br>
than what is reguired to be TCP-compatible. I&#39;m not aware of evidince <=
br>
presented to tcpm (or IETF/IRTF) which shows opposite (and I happy to be <b=
r>
educated what I have missed).<br>
<br>
You may take my comments to be a part of the expert review phase <br>
performed by the IRTF/ICCRG for CUBIC. I&#39;m not requesting to modify thi=
s <br>
doc to CUBIC++ (or something) but it seems to be that this would be <br>
necessary if this doc intends to become published as PS. For experimental, =
<br>
I think it would need some addtioinal updates and record the areas <br>
uncertainty and where more experimentation (clearly) is required.<br>
<br>
Thanks,<br>
<br>
/Markku<br>
<br>
&gt; Thanks,<br>
&gt; --<br>
&gt; Yoshi<br>
&gt; <br>
&gt; <br>
&gt; On Fri, Feb 11, 2022 at 9:34 PM Markku Kojo &lt;<a href=3D"mailto:kojo=
@cs.helsinki.fi" target=3D"_blank">kojo@cs.helsinki.fi</a>&gt; wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Hi Yoshi, all,<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0It seems to me that many issues that I raise=
d have been solved, thanks.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0However, there are still a number of importa=
nt issues that have not been<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0addressed adequately. At least the following=
:<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#135 on W_max. Yoshi&#39;s observation was c=
orrect that this is not resolved:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0the co-authors and original developpers of C=
UBIC (@lisongxu and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0@sangtaeha) agreed in their last message tha=
t Wmax needs different<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0treatment for slow start and congestion avoi=
dance and plan comprehensive<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0(new) evaluation of it. This is obviously an=
 open issue but the issue<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0was closed?<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#85 (&amp; #86 with basically same issue and=
 these two were combined) This<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0(#85) is about ECN but the major issue is on=
 using the same MD<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0factor in slow start and in congestion avoid=
ance when using<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0loss-based CC. This (#85) remained closed ev=
en though I provided a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0thorough explanation why it is wrong and aga=
inst the original theory and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0design by Van Jacobson, against the congesti=
on control principles (RFC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A02914) and two co-authors agreed on this in t=
heir same last message to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#135 when they agreed on Wmax needing rework=
. This is an important issue<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0that the wg should consider very carefully b=
ecause it is not only<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0updating RFC 5681 but also in conflict with =
RFC 2914. How can<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0tcpm (and IETF) suggest and allow one CC alg=
o to not follow congestion<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0control principles as set in RFC 2914 while =
requiring all other CCs to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0follow RFC 2914 guidelines?<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0The current draft does not provide any justi=
fication for using the same<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0MD factor in slow start as in congestion avo=
idance. Nor am I<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0awere of any experimental data that would su=
pport this change.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0The fact thet CUBIC has been long deployed d=
oes not alone provide any<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0supporting evidence because CUBIC is likely =
to give good performance as<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0it is overagressive and thereby unfair to co=
mpeting traffic and users<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0tend to be happy when measuring the performa=
nce of the sending CUBIC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0only, not the competing traffic that is badl=
y impacted. HyStart++ is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0suggested as mitigation to the problem but i=
t cannot; HyStart++ is only<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0applicable during initial slow start, not du=
ring slow start after RTO!<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0That is, the &quot;SHOULD use HYStart++&quot=
; text in Sec 4.10 is impossible<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0to implement as I have pointed out in my com=
ments earlier. Using a proper<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0MD factor in slow start is even more importa=
nt if loss is detectected<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0during a RTO recovery because the sender is =
likely to face heavy<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0congestion in such a case and it is very bad=
 if the sender continues<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0sending with overaggressive rate, stealing t=
he capacity from and causing<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0harm to coexisting flows. In addition, as I =
have explained, HyStart++<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0does not remove the problem even for the ini=
tial slow start as it is not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0shown to work always. Instead, the results w=
ith the HyStart++ draft show<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0that it reduces 50% of rexmits and only 36% =
RTOs, meaning that there is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0likely to be a notable percentage of cases w=
hen a sender is still<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0in slow start when first loss is detected (i=
.e., HyStart++ had no effect)<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0and a significant number of cases where a CU=
BIC sender is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0overaggressive continuing with a 1-40% large=
r cwnd than what is the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0available capacity. Note also that any delay=
-based heuristics like<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0HyStart++ are known to work poorly in variou=
s wireless environmens where<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0link delay tends to vary a lot. We may come =
up with some other MD factor<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0that 0.5 when in slow start and HyStart++ is=
 in use, but that is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0experimental, if not research, and definitel=
y not ready for stds track.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#114, #132, and #143 w.r.t flightsize vs. cw=
nd. The current text<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 =C2=A0 =C2=A0 does not quite correcly=
 reflect what stacks that use cwnd do.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 =C2=A0 =C2=A0 I&#39;ll comment in #14=
3 separately.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#96 &amp; #98: The text added does not addre=
ss the problems raised which are<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0also evidenced in the paper pointed by Bob i=
n #96. Even though CUBIC has<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0been modified a bit after the paper was publ=
ished, it does not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0automatically mean that the problem has been=
 shown resolved: experimental<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0evidence is required but not provided. The f=
act that CUBIC does not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0change MD factor for fast convergence is the=
 root of the problem<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0evidenced in the paper and remains so in the=
 algo specified in this<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0draft. This is also a significant problem wh=
en competing with Reno CC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0because CUBIC behaves much more aggressive t=
han Reno CC when there is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0sudden congestion and all competing flows mu=
st converge fast down to a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0small fraction of the current cwnd to be fai=
r to each other. This again<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0cannot be evidenced not to be a problem by l=
ong deployment experience<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0unless experimental data that measures the i=
mpact on competing traffic is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0presented to back the claims. Adjusting just=
 Wmax for fast convergence is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0not enough and is even likely to be ineffect=
ive because there tend to be<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0several losses when sudden congestion is hit=
, and particularly if NewReno<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is in use the sender stays several RTTs in F=
ast Recovery being<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0overaggressive and then possibly continues a=
t the same rate in CA which<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is unlikely to reach evan close to Wmax befo=
re a new loss hits<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0the sender again. That is, lower Wmax and lo=
wer additive increase factor<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0do not compensate the use of larger MD facto=
r when sudden congestion is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0encountered.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#93 &amp; #94 &amp; (#89) Sec 5.3 still does=
 not address any difficult<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0environments, in particular buffer-bloated p=
aths (nor does Sec 5.4).<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0We need evidence (results) that show CUBIC i=
s fair towards other CCs<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0(Reno) also in such environments. Note that =
CUBIC&#39;s decision to leave<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Reno-friendly region is based on the size of=
 cwnd which tends to be<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0incorrect with buffer-bloated bottlenecks be=
cause with huge buffers the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0cwnd can be many times larger than what is a=
ctually needed to fully<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0utilize the available network bit-rate. Ther=
efore, Reno CC has no problem<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0in fully utilizing such bottleneck links and=
 CUBIC must stay in<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Reno-friendly region longer but it leaves it=
 too early because the same C<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is used as with non-bloated environments. We=
 lack experiments showing<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0CUBIC follows the congestion control princip=
les and is fair to current<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0standard TCP CC; to my understanding no expe=
riments with buffer-bloated<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0bottlenecks are cited to back up the claims =
even though buffer bloat is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0very well known to be a common (difficult) e=
nvironment in today&#39;s<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Internet.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#90 The current text on applying undo (a res=
ponse to detected false<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0fastrexmit) does not provide correct result =
if someone implements it.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0I have explained the problems there in githu=
b but seem to have not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0replied to latest comments by Neal. I&#39;ll=
 reply and try to explain more.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Again one major problem here is that the dra=
ft suggest a new response<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0algo for false fast rexmits but does not pro=
vide any experimental data to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0support it. Long deployment experience has b=
een suggested as<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0justification but again without any carefull=
y evaluated experimental<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0data and evidence there is no meat. The issu=
e is important to solve but is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0not specific to CUBIC. Instead, it is genera=
l problem for all TCP CC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0variants. IMHO, this is not ready for standa=
rds track but deserves a draft<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0of its own so that it can be carefully evalu=
ated and discussed ion the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0list. AFAIK there has been no discussion on =
this on the tcpm list, so<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0those probably interested and having experie=
nce are likely to be unaware<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0that this is part of CUBIC draft.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#88 The problem with correctness of the AIMD=
 model and setting alpha for<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0CUBIC requires further consideration. Bob pr=
ovided an analysis that<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0leaves things still open. It seems that I ne=
ver had time to review and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0comment the analysis and clarify why the mod=
el does not work. I&#39;ll do that<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0separately as it is important to ensure CUBI=
C behaves fairly as intended<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0for the Reno-friendly region.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Best regards,<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0/Markku<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0On Mon, 31 Jan 2022, Yoshifumi Nishida wrote=
:<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Hello,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; After some discussions among chairs, we=
 decided to run the 2nd WGLC on<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0draft-ietf-tcpm-rfc8312bis in<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; consideration of the importance of the =
draft.=C2=A0<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; We&#39;ll be grateful if you could send=
 your feedback to the ML. The WGLC runs<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0until *Feb 11*.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; If interested, you can check in-depth p=
ast discussions in the following URL.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://github.com/NTAP/rfc8=
312bis/" rel=3D"noreferrer" target=3D"_blank">https://github.com/NTAP/rfc83=
12bis/</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Thank you so much!<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; --<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; tcpm co-chairs<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; On Wed, Jan 26, 2022 at 2:50 AM Lars Eg=
gert &lt;<a href=3D"mailto:lars@eggert.org" target=3D"_blank">lars@eggert.o=
rg</a>&gt; wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Hi,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0this -06 vers=
ion rolls in all the changes requested during (and after)<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0WGLC ended.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0I&#39;ll leav=
e it up to the chairs to decide if another WGLC is warranted<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0or the document can<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0progress as-i=
s.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Thanks,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Lars<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; On 2022-=
1-26, at 11:12, <a href=3D"mailto:internet-drafts@ietf.org" target=3D"_blan=
k">internet-drafts@ietf.org</a> wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; A New In=
ternet-Draft is available from the on-line Internet-Drafts<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0directories.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; This dra=
ft is a work item of the TCP Maintenance and Minor<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Extensions WG of the IETF.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Title=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: CUBIC =
for Fast and Long-Distance Networks<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Authors=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: Lisong Xu<b=
r>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 Sangtae Ha<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 Injong Rhee<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 Vidhi Goel<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 Lars Eggert<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0Filename=C2=A0 =C2=A0 =C2=A0 =C2=A0 : draft-ietf-tcpm-r=
fc8312bis-06.txt<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0Pages=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: 35<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0Date=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 : 2022-01=
-26<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Abstract=
:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0CUBIC is a standard TCP congestion control algorithm that uses a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0cubic function instead of a linear congestion window increase<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0function to improve scalability and stability over fast and long-<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0distance networks.=C2=A0 CUBIC has been adopted as the default TCP<br=
>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0congestion control algorithm by the Linux, Windows, and Apple<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0stacks.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0This document updates the specification of CUBIC to include<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0algorithmic improvements based on these implementations and recent<br=
>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0academic work.=C2=A0 Based on the extensive deployment experience wit=
h<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0CUBIC, it also moves the specification to the Standards Track,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0obsoleting RFC 8312.=C2=A0 This also requires updating RFC 5681, to<b=
r>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0allow<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0for CUBIC&#39;s occasionally more aggressive sending behavior.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; The IETF=
 datatracker status page for this draft is:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=
=3D"https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/" rel=3D"no=
referrer" target=3D"_blank">https://datatracker.ietf.org/doc/draft-ietf-tcp=
m-rfc8312bis/</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; There is=
 also an HTML version available at:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=
=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html" rel=
=3D"noreferrer" target=3D"_blank">https://www.ietf.org/archive/id/draft-iet=
f-tcpm-rfc8312bis-06.html</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; A diff f=
rom the previous version is available at:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=
=3D"https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06" rel=
=3D"noreferrer" target=3D"_blank">https://www.ietf.org/rfcdiff?url2=3Ddraft=
-ietf-tcpm-rfc8312bis-06</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Internet=
-Drafts are also available by rsync at<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0rsync.ietf.org::internet-drafts<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; ________=
_______________________________________<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; tcpm mai=
ling list<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=
=3D"mailto:tcpm@ietf.org" target=3D"_blank">tcpm@ietf.org</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=
=3D"https://www.ietf.org/mailman/listinfo/tcpm" rel=3D"noreferrer" target=
=3D"_blank">https://www.ietf.org/mailman/listinfo/tcpm</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0_____________=
__________________________________<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0tcpm mailing =
list<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"ma=
ilto:tcpm@ietf.org" target=3D"_blank">tcpm@ietf.org</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"ht=
tps://www.ietf.org/mailman/listinfo/tcpm" rel=3D"noreferrer" target=3D"_bla=
nk">https://www.ietf.org/mailman/listinfo/tcpm</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt; <br>
&gt; <br>
&gt;</blockquote></div></div></div></div>

--000000000000bff3b305d846d3ac--


From nobody Fri Feb 18 02:24:01 2022
Return-Path: <kojo@cs.helsinki.fi>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C9D863A0926; Fri, 18 Feb 2022 02:23:59 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cs.helsinki.fi
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EVNsyzqPhCrq; Fri, 18 Feb 2022 02:23:54 -0800 (PST)
Received: from script.cs.helsinki.fi (script.cs.helsinki.fi [128.214.11.1]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 505723A0917; Fri, 18 Feb 2022 02:23:53 -0800 (PST)
X-DKIM: Courier DKIM Filter v0.50+pk-2017-10-25 mail.cs.helsinki.fi Fri, 18 Feb 2022 12:23:45 +0200
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cs.helsinki.fi; h=date:from:to:cc:subject:in-reply-to:message-id:references :mime-version:content-type:content-id; s=dkim20130528; bh=zndmfY OfjdJsECE9gFccMYRYAkQi+SbZFwlN+7JkQ4M=; b=DVycszzAysWgmWIIV0XtIi jY8xtB3zHH3b4l4ZzLxYc8SmuHnOXqbk5C3tHsg0uxFnZ9N3VHZFycrNrYZ7+7dX P+dmk+X0AfTX4DsnK8fHg5gvl5mrOiP1TVFMwdnh4q4t5+OqYoLS1cEkKvFaIp8U SNnImxwwGdY4TRfU1d+UY=
Received: from hp8x-60 (85-76-116-210-nat.elisa-mobile.fi [85.76.116.210]) (AUTH: PLAIN kojo, TLS: TLSv1/SSLv3,256bits,AES256-GCM-SHA384) by mail.cs.helsinki.fi with ESMTPSA; Fri, 18 Feb 2022 12:23:44 +0200 id 00000000005A0069.00000000620F73B0.00004C52
Date: Fri, 18 Feb 2022 12:23:42 +0200 (EET)
From: Markku Kojo <kojo@cs.helsinki.fi>
To: Yoshifumi Nishida <nsd.ietf@gmail.com>
cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
In-Reply-To: <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com>
Message-ID: <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com>
User-Agent: Alpine 2.21 (DEB 202 2017-01-01)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="=_script-19562-1645179825-0001-2"
Content-ID: <alpine.DEB.2.21.2202181203290.4019@hp8x-60.cs.helsinki.fi>
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/J9ITTE3Sm2aLgDyqB-0r8sjp9Qw>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 18 Feb 2022 10:24:00 -0000

This is a MIME-formatted message.  If you see this text it means that your
E-mail software does not support MIME-formatted messages.

--=_script-19562-1645179825-0001-2
Content-Type: text/plain; charset="iso-8859-15"; format=flowed
Content-Transfer-Encoding: quoted-printable
Content-ID: <alpine.DEB.2.21.2202181203291.4019@hp8x-60.cs.helsinki.fi>

Hi Yoshi,

On Fri, 18 Feb 2022, Yoshifumi Nishida wrote:

> Hi Markku,
>=20
> Thanks for the comments. Yes, these are valid and important concerns. H=
owever, I have
> mixed feelings about them.
> I don't want to open a can=A0of worms, but here's are some of my person=
ally thoughts...
>=20
> One thing is that "long and wide deployment experience" might not be a =
very strong claim,
> but then, I am wondering what would be the strong one.
> One may bring some solid experiment results, but the others might argue =
it uses too
> convenient configurations, or the scale is too small or too short.

Well, that has always been and IMO still is for the wg to decide when the=20
evidence through experiments is good enough to proceed with publishing th=
e=20
doc. E.g., when RFC 5562 was considered for publication by the wg, the wg=20
requested for more experiments before making the decision and Sally=20
provided what was requested. And it was just about adding ECN capability=20
to SYNACK but here we are discussing much more significant changes.

And more importantly, here we are mainly discussing whether there is any=20
experimental evidence at all; for example, changing MD (from 0.5 to 0.7 o=
r=20
to whatever) must be considered separately when in slow start and when in=20
congestion avoidance as I have pointed out. This requires tests that=20
specifically target at studying the slow start overshoot phase and the=20
following recovery phase. AFAIK, there are no such experimental results,=20
no analysis nor evaluation presented to support the change when=20
the sender is in slow start. So, we are not discussing the level of=20
evidence but whether it exists at all, which is very different.

> The other thing is that if we set a high bar for a PS doc related to CC=
, we should apply
> the same bar to all related docs.For example, I am personally wondering =
how much we have
> analyzed and performed experiments when we published RFC9002. (BTW, my =
intention is NOT to
> accuse a specific doc, there should be some more examples.)=A0

Yes, IMO the bar should be the same. RFC 9002 was first targetted as=20
experimental but was later changed to PS. I did not have time to follow=20
the progress much at all but in my opinion it would probably have been=20
better to publish it also as experimental first. On the other hand, it=20
intends to implement NewReno as close as possible, and therefore not to=20
change the congestion control principles that we have. I cannot tell how=20
much experimental effort was put to verify the behaviour of those feature=
s=20
that are new in QUIC and/or differ from stds track TCP but hopefully=20
good enough. (Though there seems to be one issue related to resetting the=20
exponential backoff of PTO, which seems to require a more accurate=20
specification to ensure unambiguous interpretation and correct outcome,=20
but that's an off topic here).

> I am not sure how we can maintain this kind of fairness between the doc=
s at the moment.

I think it is the responsibility for the entire tsv area to do it, and=20
eventually the TSV ADs and IESG. The process to follow is clear and=20
robust, and when followed it has shown to be effectual in producing good=20
quality documents. AFAIK, we have not decided to change it, maybe just=20
many of us are unaware of it or have forgotten its existence?
Traditionally tsv area has been very careful in considering new CC=20
proposals because congestion control is so essential for the stability=20
and fairness of the Internet and I hope we are not sliping too much from=20
there. AFAIK, that's one reason why the process was documented.

Thanks,

/Markku

> Thanks,
> --
> Yoshi=A0
>=20
>=20
> On Thu, Feb 17, 2022 at 6:41 AM Markku Kojo <kojo@cs.helsinki.fi> wrote=
:
>       Hi Yoshi,
>
>       On Tue, 15 Feb 2022, Yoshifumi Nishida wrote:
>
>       > Hi=A0Markku,
>       >
>       > Thanks for the comments. I think these are very valid points.=A0
>       > However, I would like to check several things as a co-chair and =
a doc
>       shepherd before we
>       > discuss=A0the points you've=A0raised.
>       >
>       > In my understanding (please correct me if I'm wrong), when this =
draft was
>       adopted as an WG
>       > item, I think the goal of the doc was some minor updates from R=
FC8312 which
>       include more
>       > clarifications, minor=A0changes and bug fixes.=A0
>       > However, if we try to address your=A0concerns, I think we'll ne=
ed to invent a
>       new version of
>       > CUBIC something like CUBIC++ or NewCUBIC in the end.=A0
>       > I won't deny the value of such doc, but, this seems not to be w=
hat we agreed
>       on
>       > beforehand.=A0=A0
>       > if we proceed in this direction, I think we will need to check =
the WG
>       consensus whether
>       > this should be a new goal for the doc.
>       >
>       > So, I would like to check if this is what you intend for the do=
c or you
>       think we can
>       > address your points while aligning with the original goal.
>       > Also, if someone has opinions on this, please share.
>
>       I think it is important that we remember the status of RFC 8312 a=
nd the
>       decades long process that has been followed in tsv area for new
>       TCP congestion control algorithms that have been proposed and sub=
mitted
>       to IETF. In order to ensure that new cc algos are safe and fair, =
the
>       process that has been followed for all current stds track TCP cc =
algos
>       has required that the cc algo is first accepted and published as
>       experimental RFC and only once enough supportive experimental evi=
dence
>       has been gathered the doc has become a candidate to be forwaded t=
o stds
>       track. We have even agreed on a relatively strict evaluation proc=
ess to
>       follow when cc algos are brought to the IETF to be published as
>       experimental:
>
>       https://www.ietf.org/about/groups/iesg/statements/experimental-co=
ngestion-control/
>
>       RFC 8312 was published as "Informational" and if I recall correct=
ly the
>       idea was "just to publish what's out there" for the benefit of th=
e
>       community. RFC 8312 was never really evaluated, particularly not =
in the
>       way new cc algos are supposed to be as per the agreed process.
>
>       I do not recall what/how exactly was agreed when rfc8312bis was l=
aunched
>       but I would be very interested to hear the justification why this =
doc
>       does not need to follow the process mentioned above but we would =
like to
>       propose IETF to publish a non-evaluated Informational doc to be p=
ublished
>       "with minor updates", i.e., without actual evaluation, as a stds =
track
>       RFC? If the target really remains as PS then the bar should be ev=
en
>       higher than what is described for experimental docs in the above =
process
>       document, i.e, what we have followod for experimental to be moved =
to stds
>       track.
>
>       The only justification that I have heard has beed "because CUBIC =
has long
>       and wide deployment experience" and "the Internet has not smelted =
or that
>       "we should have noticed if there were problems". We must, however=
,
>       understand that in order to have noticeable bad impact CUBIC shou=
ld cause
>       some sort of congestion collapse. Congestion collapse, however, i=
s not an
>       issue with CUBIC nor with any other CC algo that applies an RTO m=
echanisms
>       together with correctly implemented Karn's algo that retains the
>       backed-off RTO until an Ack is received for a new (not rexmitted) =
data
>       packet. The issue is fairness to competing traffic. This cannot b=
e
>       observed by deploying and measuring the performance and behaviour =
of CUBIC
>       alone. CUBIC being more aggressive than current stds track TCP CC =
would
>       just gives good performance results that one running CUBIC would =
be happy
>       with. One must evaluate CUBIC's impact on the competing (Reno CC) =
traffic
>       in range of environments which requires carefully designed active
>       measurements with thoroughly-analyzed results (as required by the =
above
>       process document, RFC 5033 and RFC 2914). What we seem to be miss=
ing is
>       this evidence on CUBIC's impact and that is something the IETF mu=
st focus
>       on, not just that whether CUBIC can achieve better performance th=
an other
>       existing CCs. The latter has been shown in many publications and =
is the
>       majos focus in=A0 many scientific papers proposing new algos.
>       I appreciate a lot that CUBIC has been implemented/developped and
>       deployed for long and I wonder whether those deploying CUBIC have
>       unpublished results the wg could review before taking the decicio=
n?
>
>       I suggest everyone to read carefully RFC 2914 Sec 3.2 and particu=
larly
>       what it says about more aggressive (than RFC 5681) congestion con=
trol
>       algorithms:
>
>       =A0 Some of these may fail to implement
>       =A0 the TCP congestion avoidance mechanisms correctly because of =
poor
>       =A0 implementation [RFC2525].=A0 Others may deliberately be imple=
mented
>       =A0 with congestion avoidance algorithms that are more aggressive =
in
>       =A0 their use of bandwidth than other TCP implementations; this w=
ould
>       =A0 allow a vendor to claim to have a "faster TCP".=A0 The logica=
l
>       =A0 consequence of such implementations would be a spiral of incr=
easingly
>       =A0 aggressive TCP implementations, or increasingly aggressive tr=
ansport
>       =A0 protocols, leading back to the point where there is effective=
ly no
>       =A0 congestion avoidance and the Internet is chronically congeste=
d.
>
>       And:
>
>       =A0 It is convenient to divide flows into three classes: (1) TCP-
>       =A0 compatible flows, (2) unresponsive flows, i.e., flows that do =
not
>       =A0 slow down when congestion occurs, and (3) flows that are resp=
onsive
>       =A0 but are not TCP-compatible.=A0 The last two classes contain m=
ore
>       =A0 aggressive flows that pose significant threats to Internet
>       =A0 performance,
>
>       As I have tried to point out there are several features with CUBI=
C where
>       it is likely to be (or to me it seems it obviously is) more aggre=
ssive
>       than what is reguired to be TCP-compatible. I'm not aware of evid=
ince
>       presented to tcpm (or IETF/IRTF) which shows opposite (and I happ=
y to be
>       educated what I have missed).
>
>       You may take my comments to be a part of the expert review phase
>       performed by the IRTF/ICCRG for CUBIC. I'm not requesting to modi=
fy this
>       doc to CUBIC++ (or something) but it seems to be that this would =
be
>       necessary if this doc intends to become published as PS. For expe=
rimental,
>       I think it would need some addtioinal updates and record the area=
s
>       uncertainty and where more experimentation (clearly) is required.
>
>       Thanks,
>
>       /Markku
>
>       > Thanks,
>       > --
>       > Yoshi
>       >
>       >
>       > On Fri, Feb 11, 2022 at 9:34 PM Markku Kojo <kojo@cs.helsinki.f=
i> wrote:
>       >=A0 =A0 =A0 =A0Hi Yoshi, all,
>       >
>       >=A0 =A0 =A0 =A0It seems to me that many issues that I raised hav=
e been solved,
>       thanks.
>       >=A0 =A0 =A0 =A0However, there are still a number of important is=
sues that have not
>       been
>       >=A0 =A0 =A0 =A0addressed adequately. At least the following:
>       >
>       >=A0 =A0 =A0 =A0#135 on W_max. Yoshi's observation was correct th=
at this is not
>       resolved:
>       >=A0 =A0 =A0 =A0the co-authors and original developpers of CUBIC =
(@lisongxu and
>       >=A0 =A0 =A0 =A0@sangtaeha) agreed in their last message that Wma=
x needs different
>       >=A0 =A0 =A0 =A0treatment for slow start and congestion avoidance =
and plan
>       comprehensive
>       >=A0 =A0 =A0 =A0(new) evaluation of it. This is obviously an open =
issue but the issue
>       >=A0 =A0 =A0 =A0was closed?
>       >
>       >=A0 =A0 =A0 =A0#85 (& #86 with basically same issue and these tw=
o were combined) This
>       >=A0 =A0 =A0 =A0(#85) is about ECN but the major issue is on usin=
g the same MD
>       >=A0 =A0 =A0 =A0factor in slow start and in congestion avoidance =
when using
>       >=A0 =A0 =A0 =A0loss-based CC. This (#85) remained closed even th=
ough I provided a
>       >=A0 =A0 =A0 =A0thorough explanation why it is wrong and against =
the original theory
>       and
>       >=A0 =A0 =A0 =A0design by Van Jacobson, against the congestion co=
ntrol principles (RFC
>       >=A0 =A0 =A0 =A02914) and two co-authors agreed on this in their =
same last message to
>       >=A0 =A0 =A0 =A0#135 when they agreed on Wmax needing rework. Thi=
s is an important
>       issue
>       >=A0 =A0 =A0 =A0that the wg should consider very carefully becaus=
e it is not only
>       >=A0 =A0 =A0 =A0updating RFC 5681 but also in conflict with RFC 2=
914. How can
>       >=A0 =A0 =A0 =A0tcpm (and IETF) suggest and allow one CC algo to =
not follow congestion
>       >=A0 =A0 =A0 =A0control principles as set in RFC 2914 while requi=
ring all other CCs to
>       >=A0 =A0 =A0 =A0follow RFC 2914 guidelines?
>       >=A0 =A0 =A0 =A0The current draft does not provide any justificat=
ion for using the
>       same
>       >=A0 =A0 =A0 =A0MD factor in slow start as in congestion avoidanc=
e. Nor am I
>       >=A0 =A0 =A0 =A0awere of any experimental data that would support =
this change.
>       >=A0 =A0 =A0 =A0The fact thet CUBIC has been long deployed does n=
ot alone provide any
>       >=A0 =A0 =A0 =A0supporting evidence because CUBIC is likely to gi=
ve good performance
>       as
>       >=A0 =A0 =A0 =A0it is overagressive and thereby unfair to competi=
ng traffic and users
>       >=A0 =A0 =A0 =A0tend to be happy when measuring the performance o=
f the sending CUBIC
>       >=A0 =A0 =A0 =A0only, not the competing traffic that is badly imp=
acted. HyStart++ is
>       >=A0 =A0 =A0 =A0suggested as mitigation to the problem but it can=
not; HyStart++ is
>       only
>       >=A0 =A0 =A0 =A0applicable during initial slow start, not during =
slow start after RTO!
>       >=A0 =A0 =A0 =A0That is, the "SHOULD use HYStart++" text in Sec 4=
.10 is impossible
>       >=A0 =A0 =A0 =A0to implement as I have pointed out in my comments =
earlier. Using a
>       proper
>       >=A0 =A0 =A0 =A0MD factor in slow start is even more important if =
loss is detectected
>       >=A0 =A0 =A0 =A0during a RTO recovery because the sender is likel=
y to face heavy
>       >=A0 =A0 =A0 =A0congestion in such a case and it is very bad if t=
he sender continues
>       >=A0 =A0 =A0 =A0sending with overaggressive rate, stealing the ca=
pacity from and
>       causing
>       >=A0 =A0 =A0 =A0harm to coexisting flows. In addition, as I have =
explained, HyStart++
>       >=A0 =A0 =A0 =A0does not remove the problem even for the initial =
slow start as it is
>       not
>       >=A0 =A0 =A0 =A0shown to work always. Instead, the results with t=
he HyStart++ draft
>       show
>       >=A0 =A0 =A0 =A0that it reduces 50% of rexmits and only 36% RTOs, =
meaning that there
>       is
>       >=A0 =A0 =A0 =A0likely to be a notable percentage of cases when a =
sender is still
>       >=A0 =A0 =A0 =A0in slow start when first loss is detected (i.e., =
HyStart++ had no
>       effect)
>       >=A0 =A0 =A0 =A0and a significant number of cases where a CUBIC s=
ender is
>       >=A0 =A0 =A0 =A0overaggressive continuing with a 1-40% larger cwn=
d than what is the
>       >=A0 =A0 =A0 =A0available capacity. Note also that any delay-base=
d heuristics like
>       >=A0 =A0 =A0 =A0HyStart++ are known to work poorly in various wir=
eless environmens
>       where
>       >=A0 =A0 =A0 =A0link delay tends to vary a lot. We may come up wi=
th some other MD
>       factor
>       >=A0 =A0 =A0 =A0that 0.5 when in slow start and HyStart++ is in u=
se, but that is
>       >=A0 =A0 =A0 =A0experimental, if not research, and definitely not =
ready for stds
>       track.
>       >
>       >=A0 =A0 =A0 =A0#114, #132, and #143 w.r.t flightsize vs. cwnd. T=
he current text
>       >=A0 =A0 =A0 =A0=A0 =A0 =A0 does not quite correcly reflect what =
stacks that use cwnd do.
>       >=A0 =A0 =A0 =A0=A0 =A0 =A0 I'll comment in #143 separately.
>       >
>       >=A0 =A0 =A0 =A0#96 & #98: The text added does not address the pr=
oblems raised which
>       are
>       >=A0 =A0 =A0 =A0also evidenced in the paper pointed by Bob in #96=
. Even though CUBIC
>       has
>       >=A0 =A0 =A0 =A0been modified a bit after the paper was published=
, it does not
>       >=A0 =A0 =A0 =A0automatically mean that the problem has been show=
n resolved:
>       experimental
>       >=A0 =A0 =A0 =A0evidence is required but not provided. The fact t=
hat CUBIC does not
>       >=A0 =A0 =A0 =A0change MD factor for fast convergence is the root =
of the problem
>       >=A0 =A0 =A0 =A0evidenced in the paper and remains so in the algo =
specified in this
>       >=A0 =A0 =A0 =A0draft. This is also a significant problem when co=
mpeting with Reno CC
>       >=A0 =A0 =A0 =A0because CUBIC behaves much more aggressive than R=
eno CC when there is
>       >=A0 =A0 =A0 =A0sudden congestion and all competing flows must co=
nverge fast down to a
>       >=A0 =A0 =A0 =A0small fraction of the current cwnd to be fair to =
each other. This
>       again
>       >=A0 =A0 =A0 =A0cannot be evidenced not to be a problem by long d=
eployment experience
>       >=A0 =A0 =A0 =A0unless experimental data that measures the impact =
on competing traffic
>       is
>       >=A0 =A0 =A0 =A0presented to back the claims. Adjusting just Wmax =
for fast convergence
>       is
>       >=A0 =A0 =A0 =A0not enough and is even likely to be ineffective b=
ecause there tend to
>       be
>       >=A0 =A0 =A0 =A0several losses when sudden congestion is hit, and =
particularly if
>       NewReno
>       >=A0 =A0 =A0 =A0is in use the sender stays several RTTs in Fast R=
ecovery being
>       >=A0 =A0 =A0 =A0overaggressive and then possibly continues at the =
same rate in CA
>       which
>       >=A0 =A0 =A0 =A0is unlikely to reach evan close to Wmax before a =
new loss hits
>       >=A0 =A0 =A0 =A0the sender again. That is, lower Wmax and lower a=
dditive increase
>       factor
>       >=A0 =A0 =A0 =A0do not compensate the use of larger MD factor whe=
n sudden congestion
>       is
>       >=A0 =A0 =A0 =A0encountered.
>       >
>       >=A0 =A0 =A0 =A0#93 & #94 & (#89) Sec 5.3 still does not address =
any difficult
>       >=A0 =A0 =A0 =A0environments, in particular buffer-bloated paths =
(nor does Sec 5.4).
>       >=A0 =A0 =A0 =A0We need evidence (results) that show CUBIC is fai=
r towards other CCs
>       >=A0 =A0 =A0 =A0(Reno) also in such environments. Note that CUBIC=
's decision to leave
>       >=A0 =A0 =A0 =A0Reno-friendly region is based on the size of cwnd =
which tends to be
>       >=A0 =A0 =A0 =A0incorrect with buffer-bloated bottlenecks because =
with huge buffers
>       the
>       >=A0 =A0 =A0 =A0cwnd can be many times larger than what is actual=
ly needed to fully
>       >=A0 =A0 =A0 =A0utilize the available network bit-rate. Therefore=
, Reno CC has no
>       problem
>       >=A0 =A0 =A0 =A0in fully utilizing such bottleneck links and CUBI=
C must stay in
>       >=A0 =A0 =A0 =A0Reno-friendly region longer but it leaves it too =
early because the
>       same C
>       >=A0 =A0 =A0 =A0is used as with non-bloated environments. We lack =
experiments showing
>       >=A0 =A0 =A0 =A0CUBIC follows the congestion control principles a=
nd is fair to current
>       >=A0 =A0 =A0 =A0standard TCP CC; to my understanding no experimen=
ts with
>       buffer-bloated
>       >=A0 =A0 =A0 =A0bottlenecks are cited to back up the claims even =
though buffer bloat
>       is
>       >=A0 =A0 =A0 =A0very well known to be a common (difficult) enviro=
nment in today's
>       >=A0 =A0 =A0 =A0Internet.
>       >
>       >=A0 =A0 =A0 =A0#90 The current text on applying undo (a response =
to detected false
>       >=A0 =A0 =A0 =A0fastrexmit) does not provide correct result if so=
meone implements it.
>       >=A0 =A0 =A0 =A0I have explained the problems there in github but =
seem to have not
>       >=A0 =A0 =A0 =A0replied to latest comments by Neal. I'll reply an=
d try to explain
>       more.
>       >=A0 =A0 =A0 =A0Again one major problem here is that the draft su=
ggest a new response
>       >=A0 =A0 =A0 =A0algo for false fast rexmits but does not provide =
any experimental data
>       to
>       >=A0 =A0 =A0 =A0support it. Long deployment experience has been s=
uggested as
>       >=A0 =A0 =A0 =A0justification but again without any carefully eva=
luated experimental
>       >=A0 =A0 =A0 =A0data and evidence there is no meat. The issue is =
important to solve
>       but is
>       >=A0 =A0 =A0 =A0not specific to CUBIC. Instead, it is general pro=
blem for all TCP CC
>       >=A0 =A0 =A0 =A0variants. IMHO, this is not ready for standards t=
rack but deserves a
>       draft
>       >=A0 =A0 =A0 =A0of its own so that it can be carefully evaluated =
and discussed ion the
>       >=A0 =A0 =A0 =A0list. AFAIK there has been no discussion on this =
on the tcpm list, so
>       >=A0 =A0 =A0 =A0those probably interested and having experience a=
re likely to be
>       unaware
>       >=A0 =A0 =A0 =A0that this is part of CUBIC draft.
>       >
>       >=A0 =A0 =A0 =A0#88 The problem with correctness of the AIMD mode=
l and setting alpha
>       for
>       >=A0 =A0 =A0 =A0CUBIC requires further consideration. Bob provide=
d an analysis that
>       >=A0 =A0 =A0 =A0leaves things still open. It seems that I never h=
ad time to review and
>       >=A0 =A0 =A0 =A0comment the analysis and clarify why the model do=
es not work. I'll do
>       that
>       >=A0 =A0 =A0 =A0separately as it is important to ensure CUBIC beh=
aves fairly as
>       intended
>       >=A0 =A0 =A0 =A0for the Reno-friendly region.
>       >
>       >=A0 =A0 =A0 =A0Best regards,
>       >
>       >=A0 =A0 =A0 =A0/Markku
>       >
>       >=A0 =A0 =A0 =A0On Mon, 31 Jan 2022, Yoshifumi Nishida wrote:
>       >
>       >=A0 =A0 =A0 =A0> Hello,
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> After some discussions among chairs, we decided =
to run the 2nd WGLC
>       on
>       >=A0 =A0 =A0 =A0draft-ietf-tcpm-rfc8312bis in
>       >=A0 =A0 =A0 =A0> consideration of the importance of the draft.=A0
>       >=A0 =A0 =A0 =A0> We'll be grateful if you could send your feedba=
ck to the ML. The
>       WGLC runs
>       >=A0 =A0 =A0 =A0until *Feb 11*.
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> If interested, you can check in-depth past disc=
ussions in the
>       following URL.
>       >=A0 =A0 =A0 =A0> https://github.com/NTAP/rfc8312bis/
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> Thank you so much!
>       >=A0 =A0 =A0 =A0> --
>       >=A0 =A0 =A0 =A0> tcpm co-chairs
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0> On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <la=
rs@eggert.org> wrote:
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0Hi,
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0this -06 version rolls in all the =
changes requested during
>       (and after)
>       >=A0 =A0 =A0 =A0WGLC ended.
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0I'll leave it up to the chairs to =
decide if another WGLC is
>       warranted
>       >=A0 =A0 =A0 =A0or the document can
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0progress as-is.
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0Thanks,
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0Lars
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> On 2022-1-26, at 11:12, interne=
t-drafts@ietf.org wrote:
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> A New Internet-Draft is availab=
le from the on-line
>       Internet-Drafts
>       >=A0 =A0 =A0 =A0directories.
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> This draft is a work item of th=
e TCP Maintenance and Minor
>       >=A0 =A0 =A0 =A0Extensions WG of the IETF.
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 Title=A0 =A0 =A0 =
=A0 =A0 =A0: CUBIC for Fast and Long-Distance
>       Networks
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 Authors=A0 =A0 =A0 =
=A0 =A0: Lisong Xu
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 =A0 =A0 =A0 =A0 Sangtae Ha
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 =A0 =A0 =A0 =A0 Injong Rhee
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 =A0 =A0 =A0 =A0 Vidhi Goel
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 =A0 =A0 =A0 =A0 Lars Eggert
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0Filename=A0 =A0 =A0 =
=A0 : draft-ietf-tcpm-rfc8312bis-06.txt
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0Pages=A0 =A0 =A0 =
=A0 =A0 =A0: 35
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0Date=A0 =A0 =A0 =A0 =
=A0 =A0 : 2022-01-26
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> Abstract:
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0CUBIC is a standard TCP c=
ongestion control algorithm that
>       uses a
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0cubic function instead of =
a linear congestion window
>       increase
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0function to improve scala=
bility and stability over fast
>       and long-
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0distance networks.=A0 CUB=
IC has been adopted as the default
>       TCP
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0congestion control algori=
thm by the Linux, Windows, and
>       Apple
>       >=A0 =A0 =A0 =A0stacks.
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0This document updates the =
specification of CUBIC to
>       include
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0algorithmic improvements =
based on these implementations
>       and recent
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0academic work.=A0 Based o=
n the extensive deployment
>       experience with
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0CUBIC, it also moves the =
specification to the Standards
>       Track,
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0obsoleting RFC 8312.=A0 T=
his also requires updating RFC
>       5681, to
>       >=A0 =A0 =A0 =A0allow
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>=A0 =A0for CUBIC's occasionally =
more aggressive sending behavior.
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> The IETF datatracker status pag=
e for this draft is:
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> https://datatracker.ietf.org/do=
c/draft-ietf-tcpm-rfc8312bis/
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> There is also an HTML version a=
vailable at:
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.htm=
l
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> A diff from the previous versio=
n is available at:
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> Internet-Drafts are also availa=
ble by rsync at
>       >=A0 =A0 =A0 =A0rsync.ietf.org::internet-drafts
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> _______________________________=
________________
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> tcpm mailing list
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> tcpm@ietf.org
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0> https://www.ietf.org/mailman/li=
stinfo/tcpm
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0_________________________________=
______________
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0tcpm mailing list
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0tcpm@ietf.org
>       >=A0 =A0 =A0 =A0>=A0 =A0 =A0 =A0https://www.ietf.org/mailman/list=
info/tcpm
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>
>       >=A0 =A0 =A0 =A0>
>       >
>       >
>       >
>=20
>=20
>
--=_script-19562-1645179825-0001-2--


From nobody Fri Feb 18 02:26:51 2022
Return-Path: <kojo@cs.helsinki.fi>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 737963A0E8E; Fri, 18 Feb 2022 02:26:49 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cs.helsinki.fi
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fXBnGgq6McIH; Fri, 18 Feb 2022 02:26:44 -0800 (PST)
Received: from script.cs.helsinki.fi (script.cs.helsinki.fi [128.214.11.1]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E1F8E3A0E92; Fri, 18 Feb 2022 02:26:42 -0800 (PST)
X-DKIM: Courier DKIM Filter v0.50+pk-2017-10-25 mail.cs.helsinki.fi Fri, 18 Feb 2022 12:26:35 +0200
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cs.helsinki.fi; h=date:from:to:cc:subject:in-reply-to:message-id:references :mime-version:content-type:content-id; s=dkim20130528; bh=5VLNNN uwvLupq67KrRonhCZ+HU3knXf89pC7f8vwhxE=; b=JHMti3pw6BiVxWpelFV5Pm f8DWcOYb0hugjW9qAusinMHEeDDWbilCBZDCr+zGQ8hONk8T9uvxvfa+TrKfL7dm G6mAlXElgqe3Ievf/6VM3hkpVr5Bw0PT+i/Ohawsgr6G9skHQMLdyPjX0Y/FZVPh TKuZ/zMCx4CH3GMuE1JCM=
Received: from hp8x-60 (85-76-116-210-nat.elisa-mobile.fi [85.76.116.210]) (AUTH: PLAIN kojo, TLS: TLSv1/SSLv3,256bits,AES256-GCM-SHA384) by mail.cs.helsinki.fi with ESMTPSA; Fri, 18 Feb 2022 12:26:35 +0200 id 00000000005A0069.00000000620F745B.00005AEA
Date: Fri, 18 Feb 2022 12:26:34 +0200 (EET)
From: Markku Kojo <kojo@cs.helsinki.fi>
To: Neal Cardwell <ncardwell=40google.com@dmarc.ietf.org>
cc: Yoshifumi Nishida <nsd.ietf@gmail.com>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
In-Reply-To: <CADVnQymwCdm0NoD95SX8JpRXFok=qMygPrwaeTJ9z29dqH=Avw@mail.gmail.com>
Message-ID: <alpine.DEB.2.21.2202181031280.4019@hp8x-60.cs.helsinki.fi>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CADVnQymwCdm0NoD95SX8JpRXFok=qMygPrwaeTJ9z29dqH=Avw@mail.gmail.com>
User-Agent: Alpine 2.21 (DEB 202 2017-01-01)
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="=_script-23298-1645179995-0001-2"
Content-ID: <alpine.DEB.2.21.2202181224000.4019@hp8x-60.cs.helsinki.fi>
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/8ENVvngZxhNyKcmnYDFpVPnUyxM>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 18 Feb 2022 10:26:50 -0000

This is a MIME-formatted message.  If you see this text it means that your
E-mail software does not support MIME-formatted messages.

--=_script-23298-1645179995-0001-2
Content-Type: text/plain; charset="iso-8859-15"; format=flowed
Content-Transfer-Encoding: quoted-printable
Content-ID: <alpine.DEB.2.21.2202181154281.4019@hp8x-60.cs.helsinki.fi>

Hi Neal,

On Thu, 17 Feb 2022, Neal Cardwell wrote:

>=20
>=20
> On Thu, Feb 17, 2022 at 9:42 AM Markku Kojo <kojo=3D40cs.helsinki.fi@dm=
arc.ietf.org> wrote:
>       Hi Yoshi,
>
>       On Tue, 15 Feb 2022, Yoshifumi Nishida wrote:
>
>       > Hi=A0Markku,
>       >
>       > Thanks for the comments. I think these are very valid points.=A0
>       > However, I would like to check several things as a co-chair and =
a doc
>       shepherd before we
>       > discuss=A0the points you've=A0raised.
>       >
>       > In my understanding (please correct me if I'm wrong), when this =
draft was
>       adopted as an WG
>       > item, I think the goal of the doc was some minor updates from R=
FC8312 which
>       include more
>       > clarifications, minor=A0changes and bug fixes.=A0
>       > However, if we try to address your=A0concerns, I think we'll ne=
ed to invent a
>       new version of
>       > CUBIC something like CUBIC++ or NewCUBIC in the end.=A0
>       > I won't deny the value of such doc, but, this seems not to be w=
hat we agreed
>       on
>       > beforehand.=A0=A0
>       > if we proceed in this direction, I think we will need to check =
the WG
>       consensus whether
>       > this should be a new goal for the doc.
>       >
>       > So, I would like to check if this is what you intend for the do=
c or you
>       think we can
>       > address your points while aligning with the original goal.
>       > Also, if someone has opinions on this, please share.
>
>       I think it is important that we remember the status of RFC 8312 a=
nd the
>       decades long process that has been followed in tsv area for new
>       TCP congestion control algorithms that have been proposed and sub=
mitted
>       to IETF. In order to ensure that new cc algos are safe and fair, =
the
>       process that has been followed for all current stds track TCP cc =
algos
>       has required that the cc algo is first accepted and published as
>       experimental RFC and only once enough supportive experimental evi=
dence
>       has been gathered the doc has become a candidate to be forwaded t=
o stds
>       track. We have even agreed on a relatively strict evaluation proc=
ess to
>       follow when cc algos are brought to the IETF to be published as
>       experimental:
>
>       https://www.ietf.org/about/groups/iesg/statements/experimental-co=
ngestion-control/
>
>       RFC 8312 was published as "Informational" and if I recall correct=
ly the
>       idea was "just to publish what's out there" for the benefit of th=
e
>       community. RFC 8312 was never really evaluated, particularly not =
in the
>       way new cc algos are supposed to be as per the agreed process.
>
>       I do not recall what/how exactly was agreed when rfc8312bis was l=
aunched
>       but I would be very interested to hear the justification why this =
doc
>       does not need to follow the process mentioned above but we would =
like to
>       propose IETF to publish a non-evaluated Informational doc to be p=
ublished
>       "with minor updates", i.e., without actual evaluation, as a stds =
track
>       RFC? If the target really remains as PS then the bar should be ev=
en
>       higher than what is described for experimental docs in the above =
process
>       document, i.e, what we have followod for experimental to be moved =
to stds
>       track.
>
>       The only justification that I have heard has beed "because CUBIC =
has long
>       and wide deployment experience" and "the Internet has not smelted =
or that
>       "we should have noticed if there were problems". We must, however=
,
>       understand that in order to have noticeable bad impact CUBIC shou=
ld cause
>       some sort of congestion collapse. Congestion collapse, however, i=
s not an
>       issue with CUBIC nor with any other CC algo that applies an RTO m=
echanisms
>       together with correctly implemented Karn's algo that retains the
>       backed-off RTO until an Ack is received for a new (not rexmitted) =
data
>       packet. The issue is fairness to competing traffic. This cannot b=
e
>       observed by deploying and measuring the performance and behaviour =
of CUBIC
>       alone. CUBIC being more aggressive than current stds track TCP CC =
would
>       just gives good performance results that one running CUBIC would =
be happy
>       with. One must evaluate CUBIC's impact on the competing (Reno CC) =
traffic
>       in range of environments which requires carefully designed active
>       measurements with thoroughly-analyzed results (as required by the =
above
>       process document, RFC 5033 and RFC 2914). What we seem to be miss=
ing is
>       this evidence on CUBIC's impact and that is something the IETF mu=
st focus
>       on, not just that whether CUBIC can achieve better performance th=
an other
>       existing CCs. The latter has been shown in many publications and =
is the
>       majos focus in=A0 many scientific papers proposing new algos.
>       I appreciate a lot that CUBIC has been implemented/developped and
>       deployed for long and I wonder whether those deploying CUBIC have
>       unpublished results the wg could review before taking the decicio=
n?
>
>       I suggest everyone to read carefully RFC 2914 Sec 3.2 and particu=
larly
>       what it says about more aggressive (than RFC 5681) congestion con=
trol
>       algorithms:
>
>       =A0 Some of these may fail to implement
>       =A0 the TCP congestion avoidance mechanisms correctly because of =
poor
>       =A0 implementation [RFC2525].=A0 Others may deliberately be imple=
mented
>       =A0 with congestion avoidance algorithms that are more aggressive =
in
>       =A0 their use of bandwidth than other TCP implementations; this w=
ould
>       =A0 allow a vendor to claim to have a "faster TCP".=A0 The logica=
l
>       =A0 consequence of such implementations would be a spiral of incr=
easingly
>       =A0 aggressive TCP implementations, or increasingly aggressive tr=
ansport
>       =A0 protocols, leading back to the point where there is effective=
ly no
>       =A0 congestion avoidance and the Internet is chronically congeste=
d.
>
>       And:
>
>       =A0 It is convenient to divide flows into three classes: (1) TCP-
>       =A0 compatible flows, (2) unresponsive flows, i.e., flows that do =
not
>       =A0 slow down when congestion occurs, and (3) flows that are resp=
onsive
>       =A0 but are not TCP-compatible.=A0 The last two classes contain m=
ore
>       =A0 aggressive flows that pose significant threats to Internet
>       =A0 performance,
>
>       As I have tried to point out there are several features with CUBI=
C where
>       it is likely to be (or to me it seems it obviously is) more aggre=
ssive
>       than what is reguired to be TCP-compatible. I'm not aware of evid=
ince
>       presented to tcpm (or IETF/IRTF) which shows opposite (and I happ=
y to be
>       educated what I have missed).
>
>       You may take my comments to be a part of the expert review phase
>       performed by the IRTF/ICCRG for CUBIC. I'm not requesting to modi=
fy this
>       doc to CUBIC++ (or something) but it seems to be that this would =
be
>       necessary if this doc intends to become published as PS. For expe=
rimental,
>       I think it would need some addtioinal updates and record the area=
s
>       uncertainty and where more experimentation (clearly) is required.
>=20
>=20
> Hi Markku,
>=20
> Thanks for your careful argument above. Given your argument, in order t=
o continue to make
> constructive progress on rfc8312bis without blocking for new research a=
nd experimentation,
> I would propose that we change rfc8312bis so that it remains "Informati=
onal" (like
> RFC8312) rather than the "Standards Track" status currently listed at:
>=20
> =A0 https://ntap.github.io/rfc8312bis/draft-ietf-tcpm-rfc8312bis.html
>=20
> As far as I can tell, that would allow all of the valuable contribution=
s in rfc8312bis to
> carry forward and be published, while still respecting all of the consi=
derations you
> enumerated above, which center around concerns with calling CUBIC "stan=
dards track".
>=20
> How does that sound?

Sure, that could be one possible path forward. In my opinion, publishing=20
it as experimental would be more logical and would also provide an=20
inherent path forward to PS once enough evidence has been gathered and wg=20
considers it is mature enough to proceed.  I don't have a strong opinion=20
on this. Either way, IMO, it would be important to clearly point out and=20
discuss the areas that require more evaluation so that it encourages=20
interested parties to focus their experimentation to those important=20
areas.

Thanks,

/Markku

  > best,
> neal
>=20
>
--=_script-23298-1645179995-0001-2--


From nobody Fri Feb 18 02:39:35 2022
Return-Path: <vidhi_goel@apple.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DE5E13A0F74; Fri, 18 Feb 2022 02:39:28 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.673
X-Spam-Level: 
X-Spam-Status: No, score=-2.673 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.576, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=unavailable autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=apple.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GusyXovSWMYK; Fri, 18 Feb 2022 02:39:23 -0800 (PST)
Received: from ma1-aaemail-dr-lapp02.apple.com (ma1-aaemail-dr-lapp02.apple.com [17.171.2.68]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E04843A0BF6; Fri, 18 Feb 2022 02:39:18 -0800 (PST)
Received: from pps.filterd (ma1-aaemail-dr-lapp02.apple.com [127.0.0.1]) by ma1-aaemail-dr-lapp02.apple.com (8.16.0.42/8.16.0.42) with SMTP id 21IASCm7013105; Fri, 18 Feb 2022 02:39:16 -0800
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=apple.com; h=from : message-id : content-type : mime-version : subject : date : in-reply-to : cc : to : references; s=20180706; bh=Hov9C+fp8P0OmntE3vrLVNR0rYgiMxKu0MD/2KKTS0k=; b=dVxKcPoHd+z4In3Q844zL4fWd2bajAx983vqxcFW4dRz9vwvJe5kLjE/OULVimb4S89t 9IQ+YyVQOOFYoagNvPdnysiWb4qdrpf+JOYo9vKr4DUTkeOUwpV028o3+jliYd0HiJIN +0ts7+vohImkdgE9jWCEIBqNNMNZn138du3YMCGa8LXs+BZboxWmUw+YDh9g+WnKKMvR 9o59u7NfpHVk4NaUBdQyBSyHVxyv8YOYBN+CRHC5RH5nJrgQtixXM4irtWGam3W8aaIo W6IDkwVUTJUtFQPmrs78JlrVT6rtlwEvK/39TUuWp6Hp+hgXRgk9VuNLSf5Kf5dlfBc1 pQ== 
Received: from rn-mailsvcp-mta-lapp01.rno.apple.com (rn-mailsvcp-mta-lapp01.rno.apple.com [10.225.203.149]) by ma1-aaemail-dr-lapp02.apple.com with ESMTP id 3e8n5mbjdp-2 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NO); Fri, 18 Feb 2022 02:39:16 -0800
Received: from rn-mailsvcp-mmp-lapp03.rno.apple.com (rn-mailsvcp-mmp-lapp03.rno.apple.com [17.179.253.16]) by rn-mailsvcp-mta-lapp01.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPS id <0R7H00OT9XLGT1F0@rn-mailsvcp-mta-lapp01.rno.apple.com>;  Fri, 18 Feb 2022 02:39:16 -0800 (PST)
Received: from process_milters-daemon.rn-mailsvcp-mmp-lapp03.rno.apple.com by rn-mailsvcp-mmp-lapp03.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) id <0R7H00X00XBLVH00@rn-mailsvcp-mmp-lapp03.rno.apple.com>; Fri, 18 Feb 2022 02:39:16 -0800 (PST)
X-Va-A: 
X-Va-T-CD: 3995b166a75559493a3d304aae4b10d4
X-Va-E-CD: 749c6e7503a1c30456497a36735b331a
X-Va-R-CD: 59a455eaeb25ffef4add4d7c725dcf79
X-Va-CD: 0
X-Va-ID: 4dcaaf97-8726-4675-92ef-dfa36a34772c
X-V-A: 
X-V-T-CD: 3995b166a75559493a3d304aae4b10d4
X-V-E-CD: 749c6e7503a1c30456497a36735b331a
X-V-R-CD: 59a455eaeb25ffef4add4d7c725dcf79
X-V-CD: 0
X-V-ID: 6dfc95af-c45f-4fea-8f1b-17c5b668864f
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-18_04:2022-02-17, 2022-02-18 signatures=0
Received: from smtpclient.apple (unknown [17.11.114.97]) by rn-mailsvcp-mmp-lapp03.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPSA id <0R7H00OCHXLFB400@rn-mailsvcp-mmp-lapp03.rno.apple.com>; Fri, 18 Feb 2022 02:39:16 -0800 (PST)
From: Vidhi Goel <vidhi_goel@apple.com>
Message-id: <6709DA75-CDC2-49CE-BF5F-AEB94F67FDDE@apple.com>
Content-type: multipart/signed; boundary="Apple-Mail=_5612D5A8-FDCE-4B25-BE35-46A10DD7B51A"; protocol="application/pkcs7-signature"; micalg=sha-256
MIME-version: 1.0 (Mac OS X Mail 16.0 \(3696.80.82.1.1\))
Date: Fri, 18 Feb 2022 02:39:14 -0800
In-reply-to: <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi>
Cc: Yoshifumi Nishida <nsd.ietf@gmail.com>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
To: Markku Kojo <kojo=40cs.helsinki.fi@dmarc.ietf.org>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi>
X-Mailer: Apple Mail (2.3696.80.82.1.1)
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-18_04:2022-02-17, 2022-02-18 signatures=0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/4WMB7hdu4_HPAfscZaBkym71Bso>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 18 Feb 2022 10:39:33 -0000

--Apple-Mail=_5612D5A8-FDCE-4B25-BE35-46A10DD7B51A
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_DAA60ABD-F9B5-4889-96E7-C9BC6BD2C5AA"


--Apple-Mail=_DAA60ABD-F9B5-4889-96E7-C9BC6BD2C5AA
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii


>> The other thing is that if we set a high bar for a PS doc related to =
CC, we should apply
>> the same bar to all related docs.For example, I am personally =
wondering how much we have
>> analyzed and performed experiments when we published RFC9002. (BTW, =
my intention is NOT to
>> accuse a specific doc, there should be some more examples.)=20
>=20
> Yes, IMO the bar should be the same. RFC 9002 was first targetted as =
experimental but was later changed to PS. I did not have time to follow =
the progress much at all but in my opinion it would probably have been =
better to publish it also as experimental first. On the other hand, it =
intends to implement NewReno as close as possible, and therefore not to =
change the congestion control principles that we have. I cannot tell how =
much experimental effort was put to verify the behaviour of those =
features that are new in QUIC and/or differ from stds track TCP but =
hopefully good enough. (Though there seems to be one issue related to =
resetting the exponential backoff of PTO, which seems to require a more =
accurate specification to ensure unambiguous interpretation and correct =
outcome, but that's an off topic here).

RFC 9002 deviates from Reno as it uses `cwnd` during multiplicative =
decrease on a congestion event - =
https://www.rfc-editor.org/rfc/rfc9002.html#name-on-new-congestion-event =
<https://www.rfc-editor.org/rfc/rfc9002.html#name-on-new-congestion-event>=
. This is definitely a strong deviation from RFC 5681 which clearly =
mentions not to use cwnd.
Its good to know that you are hopeful about the experimental effort put =
in RFC 9002 without the evidence being explicitly presented at IETF.

Thanks,
Vidhi

> On Feb 18, 2022, at 2:23 AM, Markku Kojo =
<kojo=3D40cs.helsinki.fi@dmarc.ietf.org> wrote:
>=20
> Hi Yoshi,
>=20
> On Fri, 18 Feb 2022, Yoshifumi Nishida wrote:
>=20
>> Hi Markku,
>> Thanks for the comments. Yes, these are valid and important concerns. =
However, I have
>> mixed feelings about them.
>> I don't want to open a can of worms, but here's are some of my =
personally thoughts...
>> One thing is that "long and wide deployment experience" might not be =
a very strong claim,
>> but then, I am wondering what would be the strong one.
>> One may bring some solid experiment results, but the others might =
argue it uses too
>> convenient configurations, or the scale is too small or too short.
>=20
> Well, that has always been and IMO still is for the wg to decide when =
the evidence through experiments is good enough to proceed with =
publishing the doc. E.g., when RFC 5562 was considered for publication =
by the wg, the wg requested for more experiments before making the =
decision and Sally provided what was requested. And it was just about =
adding ECN capability to SYNACK but here we are discussing much more =
significant changes.
>=20
> And more importantly, here we are mainly discussing whether there is =
any experimental evidence at all; for example, changing MD (from 0.5 to =
0.7 or to whatever) must be considered separately when in slow start and =
when in congestion avoidance as I have pointed out. This requires tests =
that specifically target at studying the slow start overshoot phase and =
the following recovery phase. AFAIK, there are no such experimental =
results, no analysis nor evaluation presented to support the change when =
the sender is in slow start. So, we are not discussing the level of =
evidence but whether it exists at all, which is very different.
>=20
>> The other thing is that if we set a high bar for a PS doc related to =
CC, we should apply
>> the same bar to all related docs.For example, I am personally =
wondering how much we have
>> analyzed and performed experiments when we published RFC9002. (BTW, =
my intention is NOT to
>> accuse a specific doc, there should be some more examples.)=20
>=20
> Yes, IMO the bar should be the same. RFC 9002 was first targetted as =
experimental but was later changed to PS. I did not have time to follow =
the progress much at all but in my opinion it would probably have been =
better to publish it also as experimental first. On the other hand, it =
intends to implement NewReno as close as possible, and therefore not to =
change the congestion control principles that we have. I cannot tell how =
much experimental effort was put to verify the behaviour of those =
features that are new in QUIC and/or differ from stds track TCP but =
hopefully good enough. (Though there seems to be one issue related to =
resetting the exponential backoff of PTO, which seems to require a more =
accurate specification to ensure unambiguous interpretation and correct =
outcome, but that's an off topic here).
>=20
>> I am not sure how we can maintain this kind of fairness between the =
docs at the moment.
>=20
> I think it is the responsibility for the entire tsv area to do it, and =
eventually the TSV ADs and IESG. The process to follow is clear and =
robust, and when followed it has shown to be effectual in producing good =
quality documents. AFAIK, we have not decided to change it, maybe just =
many of us are unaware of it or have forgotten its existence?
> Traditionally tsv area has been very careful in considering new CC =
proposals because congestion control is so essential for the stability =
and fairness of the Internet and I hope we are not sliping too much from =
there. AFAIK, that's one reason why the process was documented.
>=20
> Thanks,
>=20
> /Markku
>=20
>> Thanks,
>> --
>> Yoshi=20
>> On Thu, Feb 17, 2022 at 6:41 AM Markku Kojo <kojo@cs.helsinki.fi> =
wrote:
>>      Hi Yoshi,
>>=20
>>      On Tue, 15 Feb 2022, Yoshifumi Nishida wrote:
>>=20
>>      > Hi Markku,
>>      >
>>      > Thanks for the comments. I think these are very valid points.=20=

>>      > However, I would like to check several things as a co-chair =
and a doc
>>      shepherd before we
>>      > discuss the points you've raised.
>>      >
>>      > In my understanding (please correct me if I'm wrong), when =
this draft was
>>      adopted as an WG
>>      > item, I think the goal of the doc was some minor updates from =
RFC8312 which
>>      include more
>>      > clarifications, minor changes and bug fixes.=20
>>      > However, if we try to address your concerns, I think we'll =
need to invent a
>>      new version of
>>      > CUBIC something like CUBIC++ or NewCUBIC in the end.=20
>>      > I won't deny the value of such doc, but, this seems not to be =
what we agreed
>>      on
>>      > beforehand. =20
>>      > if we proceed in this direction, I think we will need to check =
the WG
>>      consensus whether
>>      > this should be a new goal for the doc.
>>      >
>>      > So, I would like to check if this is what you intend for the =
doc or you
>>      think we can
>>      > address your points while aligning with the original goal.
>>      > Also, if someone has opinions on this, please share.
>>=20
>>      I think it is important that we remember the status of RFC 8312 =
and the
>>      decades long process that has been followed in tsv area for new
>>      TCP congestion control algorithms that have been proposed and =
submitted
>>      to IETF. In order to ensure that new cc algos are safe and fair, =
the
>>      process that has been followed for all current stds track TCP cc =
algos
>>      has required that the cc algo is first accepted and published as
>>      experimental RFC and only once enough supportive experimental =
evidence
>>      has been gathered the doc has become a candidate to be forwaded =
to stds
>>      track. We have even agreed on a relatively strict evaluation =
process to
>>      follow when cc algos are brought to the IETF to be published as
>>      experimental:
>>=20
>>      =
https://www.ietf.org/about/groups/iesg/statements/experimental-congestion-=
control/
>>=20
>>      RFC 8312 was published as "Informational" and if I recall =
correctly the
>>      idea was "just to publish what's out there" for the benefit of =
the
>>      community. RFC 8312 was never really evaluated, particularly not =
in the
>>      way new cc algos are supposed to be as per the agreed process.
>>=20
>>      I do not recall what/how exactly was agreed when rfc8312bis was =
launched
>>      but I would be very interested to hear the justification why =
this doc
>>      does not need to follow the process mentioned above but we would =
like to
>>      propose IETF to publish a non-evaluated Informational doc to be =
published
>>      "with minor updates", i.e., without actual evaluation, as a stds =
track
>>      RFC? If the target really remains as PS then the bar should be =
even
>>      higher than what is described for experimental docs in the above =
process
>>      document, i.e, what we have followod for experimental to be =
moved to stds
>>      track.
>>=20
>>      The only justification that I have heard has beed "because CUBIC =
has long
>>      and wide deployment experience" and "the Internet has not =
smelted or that
>>      "we should have noticed if there were problems". We must, =
however,
>>      understand that in order to have noticeable bad impact CUBIC =
should cause
>>      some sort of congestion collapse. Congestion collapse, however, =
is not an
>>      issue with CUBIC nor with any other CC algo that applies an RTO =
mechanisms
>>      together with correctly implemented Karn's algo that retains the
>>      backed-off RTO until an Ack is received for a new (not =
rexmitted) data
>>      packet. The issue is fairness to competing traffic. This cannot =
be
>>      observed by deploying and measuring the performance and =
behaviour of CUBIC
>>      alone. CUBIC being more aggressive than current stds track TCP =
CC would
>>      just gives good performance results that one running CUBIC would =
be happy
>>      with. One must evaluate CUBIC's impact on the competing (Reno =
CC) traffic
>>      in range of environments which requires carefully designed =
active
>>      measurements with thoroughly-analyzed results (as required by =
the above
>>      process document, RFC 5033 and RFC 2914). What we seem to be =
missing is
>>      this evidence on CUBIC's impact and that is something the IETF =
must focus
>>      on, not just that whether CUBIC can achieve better performance =
than other
>>      existing CCs. The latter has been shown in many publications and =
is the
>>      majos focus in  many scientific papers proposing new algos.
>>      I appreciate a lot that CUBIC has been implemented/developped =
and
>>      deployed for long and I wonder whether those deploying CUBIC =
have
>>      unpublished results the wg could review before taking the =
decicion?
>>=20
>>      I suggest everyone to read carefully RFC 2914 Sec 3.2 and =
particularly
>>      what it says about more aggressive (than RFC 5681) congestion =
control
>>      algorithms:
>>=20
>>        Some of these may fail to implement
>>        the TCP congestion avoidance mechanisms correctly because of =
poor
>>        implementation [RFC2525].  Others may deliberately be =
implemented
>>        with congestion avoidance algorithms that are more aggressive =
in
>>        their use of bandwidth than other TCP implementations; this =
would
>>        allow a vendor to claim to have a "faster TCP".  The logical
>>        consequence of such implementations would be a spiral of =
increasingly
>>        aggressive TCP implementations, or increasingly aggressive =
transport
>>        protocols, leading back to the point where there is =
effectively no
>>        congestion avoidance and the Internet is chronically =
congested.
>>=20
>>      And:
>>=20
>>        It is convenient to divide flows into three classes: (1) TCP-
>>        compatible flows, (2) unresponsive flows, i.e., flows that do =
not
>>        slow down when congestion occurs, and (3) flows that are =
responsive
>>        but are not TCP-compatible.  The last two classes contain more
>>        aggressive flows that pose significant threats to Internet
>>        performance,
>>=20
>>      As I have tried to point out there are several features with =
CUBIC where
>>      it is likely to be (or to me it seems it obviously is) more =
aggressive
>>      than what is reguired to be TCP-compatible. I'm not aware of =
evidince
>>      presented to tcpm (or IETF/IRTF) which shows opposite (and I =
happy to be
>>      educated what I have missed).
>>=20
>>      You may take my comments to be a part of the expert review phase
>>      performed by the IRTF/ICCRG for CUBIC. I'm not requesting to =
modify this
>>      doc to CUBIC++ (or something) but it seems to be that this would =
be
>>      necessary if this doc intends to become published as PS. For =
experimental,
>>      I think it would need some addtioinal updates and record the =
areas
>>      uncertainty and where more experimentation (clearly) is =
required.
>>=20
>>      Thanks,
>>=20
>>      /Markku
>>=20
>>      > Thanks,
>>      > --
>>      > Yoshi
>>      >
>>      >
>>      > On Fri, Feb 11, 2022 at 9:34 PM Markku Kojo =
<kojo@cs.helsinki.fi> wrote:
>>      >       Hi Yoshi, all,
>>      >
>>      >       It seems to me that many issues that I raised have been =
solved,
>>      thanks.
>>      >       However, there are still a number of important issues =
that have not
>>      been
>>      >       addressed adequately. At least the following:
>>      >
>>      >       #135 on W_max. Yoshi's observation was correct that this =
is not
>>      resolved:
>>      >       the co-authors and original developpers of CUBIC =
(@lisongxu and
>>      >       @sangtaeha) agreed in their last message that Wmax needs =
different
>>      >       treatment for slow start and congestion avoidance and =
plan
>>      comprehensive
>>      >       (new) evaluation of it. This is obviously an open issue =
but the issue
>>      >       was closed?
>>      >
>>      >       #85 (& #86 with basically same issue and these two were =
combined) This
>>      >       (#85) is about ECN but the major issue is on using the =
same MD
>>      >       factor in slow start and in congestion avoidance when =
using
>>      >       loss-based CC. This (#85) remained closed even though I =
provided a
>>      >       thorough explanation why it is wrong and against the =
original theory
>>      and
>>      >       design by Van Jacobson, against the congestion control =
principles (RFC
>>      >       2914) and two co-authors agreed on this in their same =
last message to
>>      >       #135 when they agreed on Wmax needing rework. This is an =
important
>>      issue
>>      >       that the wg should consider very carefully because it is =
not only
>>      >       updating RFC 5681 but also in conflict with RFC 2914. =
How can
>>      >       tcpm (and IETF) suggest and allow one CC algo to not =
follow congestion
>>      >       control principles as set in RFC 2914 while requiring =
all other CCs to
>>      >       follow RFC 2914 guidelines?
>>      >       The current draft does not provide any justification for =
using the
>>      same
>>      >       MD factor in slow start as in congestion avoidance. Nor =
am I
>>      >       awere of any experimental data that would support this =
change.
>>      >       The fact thet CUBIC has been long deployed does not =
alone provide any
>>      >       supporting evidence because CUBIC is likely to give good =
performance
>>      as
>>      >       it is overagressive and thereby unfair to competing =
traffic and users
>>      >       tend to be happy when measuring the performance of the =
sending CUBIC
>>      >       only, not the competing traffic that is badly impacted. =
HyStart++ is
>>      >       suggested as mitigation to the problem but it cannot; =
HyStart++ is
>>      only
>>      >       applicable during initial slow start, not during slow =
start after RTO!
>>      >       That is, the "SHOULD use HYStart++" text in Sec 4.10 is =
impossible
>>      >       to implement as I have pointed out in my comments =
earlier. Using a
>>      proper
>>      >       MD factor in slow start is even more important if loss =
is detectected
>>      >       during a RTO recovery because the sender is likely to =
face heavy
>>      >       congestion in such a case and it is very bad if the =
sender continues
>>      >       sending with overaggressive rate, stealing the capacity =
from and
>>      causing
>>      >       harm to coexisting flows. In addition, as I have =
explained, HyStart++
>>      >       does not remove the problem even for the initial slow =
start as it is
>>      not
>>      >       shown to work always. Instead, the results with the =
HyStart++ draft
>>      show
>>      >       that it reduces 50% of rexmits and only 36% RTOs, =
meaning that there
>>      is
>>      >       likely to be a notable percentage of cases when a sender =
is still
>>      >       in slow start when first loss is detected (i.e., =
HyStart++ had no
>>      effect)
>>      >       and a significant number of cases where a CUBIC sender =
is
>>      >       overaggressive continuing with a 1-40% larger cwnd than =
what is the
>>      >       available capacity. Note also that any delay-based =
heuristics like
>>      >       HyStart++ are known to work poorly in various wireless =
environmens
>>      where
>>      >       link delay tends to vary a lot. We may come up with some =
other MD
>>      factor
>>      >       that 0.5 when in slow start and HyStart++ is in use, but =
that is
>>      >       experimental, if not research, and definitely not ready =
for stds
>>      track.
>>      >
>>      >       #114, #132, and #143 w.r.t flightsize vs. cwnd. The =
current text
>>      >             does not quite correcly reflect what stacks that =
use cwnd do.
>>      >             I'll comment in #143 separately.
>>      >
>>      >       #96 & #98: The text added does not address the problems =
raised which
>>      are
>>      >       also evidenced in the paper pointed by Bob in #96. Even =
though CUBIC
>>      has
>>      >       been modified a bit after the paper was published, it =
does not
>>      >       automatically mean that the problem has been shown =
resolved:
>>      experimental
>>      >       evidence is required but not provided. The fact that =
CUBIC does not
>>      >       change MD factor for fast convergence is the root of the =
problem
>>      >       evidenced in the paper and remains so in the algo =
specified in this
>>      >       draft. This is also a significant problem when competing =
with Reno CC
>>      >       because CUBIC behaves much more aggressive than Reno CC =
when there is
>>      >       sudden congestion and all competing flows must converge =
fast down to a
>>      >       small fraction of the current cwnd to be fair to each =
other. This
>>      again
>>      >       cannot be evidenced not to be a problem by long =
deployment experience
>>      >       unless experimental data that measures the impact on =
competing traffic
>>      is
>>      >       presented to back the claims. Adjusting just Wmax for =
fast convergence
>>      is
>>      >       not enough and is even likely to be ineffective because =
there tend to
>>      be
>>      >       several losses when sudden congestion is hit, and =
particularly if
>>      NewReno
>>      >       is in use the sender stays several RTTs in Fast Recovery =
being
>>      >       overaggressive and then possibly continues at the same =
rate in CA
>>      which
>>      >       is unlikely to reach evan close to Wmax before a new =
loss hits
>>      >       the sender again. That is, lower Wmax and lower additive =
increase
>>      factor
>>      >       do not compensate the use of larger MD factor when =
sudden congestion
>>      is
>>      >       encountered.
>>      >
>>      >       #93 & #94 & (#89) Sec 5.3 still does not address any =
difficult
>>      >       environments, in particular buffer-bloated paths (nor =
does Sec 5.4).
>>      >       We need evidence (results) that show CUBIC is fair =
towards other CCs
>>      >       (Reno) also in such environments. Note that CUBIC's =
decision to leave
>>      >       Reno-friendly region is based on the size of cwnd which =
tends to be
>>      >       incorrect with buffer-bloated bottlenecks because with =
huge buffers
>>      the
>>      >       cwnd can be many times larger than what is actually =
needed to fully
>>      >       utilize the available network bit-rate. Therefore, Reno =
CC has no
>>      problem
>>      >       in fully utilizing such bottleneck links and CUBIC must =
stay in
>>      >       Reno-friendly region longer but it leaves it too early =
because the
>>      same C
>>      >       is used as with non-bloated environments. We lack =
experiments showing
>>      >       CUBIC follows the congestion control principles and is =
fair to current
>>      >       standard TCP CC; to my understanding no experiments with
>>      buffer-bloated
>>      >       bottlenecks are cited to back up the claims even though =
buffer bloat
>>      is
>>      >       very well known to be a common (difficult) environment =
in today's
>>      >       Internet.
>>      >
>>      >       #90 The current text on applying undo (a response to =
detected false
>>      >       fastrexmit) does not provide correct result if someone =
implements it.
>>      >       I have explained the problems there in github but seem =
to have not
>>      >       replied to latest comments by Neal. I'll reply and try =
to explain
>>      more.
>>      >       Again one major problem here is that the draft suggest a =
new response
>>      >       algo for false fast rexmits but does not provide any =
experimental data
>>      to
>>      >       support it. Long deployment experience has been =
suggested as
>>      >       justification but again without any carefully evaluated =
experimental
>>      >       data and evidence there is no meat. The issue is =
important to solve
>>      but is
>>      >       not specific to CUBIC. Instead, it is general problem =
for all TCP CC
>>      >       variants. IMHO, this is not ready for standards track =
but deserves a
>>      draft
>>      >       of its own so that it can be carefully evaluated and =
discussed ion the
>>      >       list. AFAIK there has been no discussion on this on the =
tcpm list, so
>>      >       those probably interested and having experience are =
likely to be
>>      unaware
>>      >       that this is part of CUBIC draft.
>>      >
>>      >       #88 The problem with correctness of the AIMD model and =
setting alpha
>>      for
>>      >       CUBIC requires further consideration. Bob provided an =
analysis that
>>      >       leaves things still open. It seems that I never had time =
to review and
>>      >       comment the analysis and clarify why the model does not =
work. I'll do
>>      that
>>      >       separately as it is important to ensure CUBIC behaves =
fairly as
>>      intended
>>      >       for the Reno-friendly region.
>>      >
>>      >       Best regards,
>>      >
>>      >       /Markku
>>      >
>>      >       On Mon, 31 Jan 2022, Yoshifumi Nishida wrote:
>>      >
>>      >       > Hello,
>>      >       >
>>      >       > After some discussions among chairs, we decided to run =
the 2nd WGLC
>>      on
>>      >       draft-ietf-tcpm-rfc8312bis in
>>      >       > consideration of the importance of the draft.=20
>>      >       > We'll be grateful if you could send your feedback to =
the ML. The
>>      WGLC runs
>>      >       until *Feb 11*.
>>      >       >
>>      >       > If interested, you can check in-depth past discussions =
in the
>>      following URL.
>>      >       > https://github.com/NTAP/rfc8312bis/
>>      >       >
>>      >       > Thank you so much!
>>      >       > --
>>      >       > tcpm co-chairs
>>      >       >
>>      >       >
>>      >       > On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert =
<lars@eggert.org> wrote:
>>      >       >       Hi,
>>      >       >
>>      >       >       this -06 version rolls in all the changes =
requested during
>>      (and after)
>>      >       WGLC ended.
>>      >       >
>>      >       >       I'll leave it up to the chairs to decide if =
another WGLC is
>>      warranted
>>      >       or the document can
>>      >       >       progress as-is.
>>      >       >
>>      >       >       Thanks,
>>      >       >       Lars
>>      >       >
>>      >       >
>>      >       >       > On 2022-1-26, at 11:12, =
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 TCP =
Maintenance and Minor
>>      >       Extensions WG of the IETF.
>>      >       >       >
>>      >       >       >        Title           : CUBIC for Fast and =
Long-Distance
>>      Networks
>>      >       >       >        Authors         : Lisong Xu
>>      >       >       >                          Sangtae Ha
>>      >       >       >                          Injong Rhee
>>      >       >       >                          Vidhi Goel
>>      >       >       >                          Lars Eggert
>>      >       >       >       Filename        : =
draft-ietf-tcpm-rfc8312bis-06.txt
>>      >       >       >       Pages           : 35
>>      >       >       >       Date            : 2022-01-26
>>      >       >       >
>>      >       >       > Abstract:
>>      >       >       >   CUBIC is a standard TCP congestion control =
algorithm that
>>      uses a
>>      >       >       >   cubic function instead of a linear =
congestion window
>>      increase
>>      >       >       >   function to improve scalability and =
stability over fast
>>      and long-
>>      >       >       >   distance networks.  CUBIC has been adopted =
as the default
>>      TCP
>>      >       >       >   congestion control algorithm by the Linux, =
Windows, and
>>      Apple
>>      >       stacks.
>>      >       >       >
>>      >       >       >   This document updates the specification of =
CUBIC to
>>      include
>>      >       >       >   algorithmic improvements based on these =
implementations
>>      and recent
>>      >       >       >   academic work.  Based on the extensive =
deployment
>>      experience with
>>      >       >       >   CUBIC, it also moves the specification to =
the Standards
>>      Track,
>>      >       >       >   obsoleting RFC 8312.  This also requires =
updating RFC
>>      5681, to
>>      >       allow
>>      >       >       >   for CUBIC's occasionally more aggressive =
sending behavior.
>>      >       >       >
>>      >       >       >
>>      >       >       > The IETF datatracker status page for this =
draft is:
>>      >       >       > =
https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/
>>      >       >       >
>>      >       >       > There is also an HTML version available at:
>>      >       >       >
>>      =
https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html
>>      >       >       >
>>      >       >       > A diff from the previous version is available =
at:
>>      >       >       >
>>      https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06
>>      >       >       >
>>      >       >       >
>>      >       >       > Internet-Drafts are also available by rsync at
>>      >       rsync.ietf.org::internet-drafts
>>      >       >       >
>>      >       >       >
>>      >       >       > =
_______________________________________________
>>      >       >       > tcpm mailing list
>>      >       >       > tcpm@ietf.org
>>      >       >       > https://www.ietf.org/mailman/listinfo/tcpm
>>      >       >
>>      >       >       _______________________________________________
>>      >       >       tcpm mailing list
>>      >       >       tcpm@ietf.org
>>      >       >       https://www.ietf.org/mailman/listinfo/tcpm
>>      >       >
>>      >       >
>>      >       >
>>      >
>>      >
>>      >
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm


--Apple-Mail=_DAA60ABD-F9B5-4889-96E7-C9BC6BD2C5AA
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=us-ascii

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3Dus-ascii"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: after-white-space;" class=3D""><div =
class=3D""><br class=3D""></div><blockquote type=3D"cite" =
class=3D""><blockquote type=3D"cite" class=3D"">The other thing is that =
if we set a high bar for a PS doc related to CC, we should apply<br =
class=3D"">the same bar to all related docs.For example, I am personally =
wondering how much we have<br class=3D"">analyzed and performed =
experiments when we published RFC9002. (BTW, my intention is NOT to<br =
class=3D"">accuse a specific doc, there should be some more =
examples.)&nbsp;<br class=3D""></blockquote><br class=3D"">Yes, IMO the =
bar should be the same. RFC 9002 was first targetted as experimental but =
was later changed to PS. I did not have time to follow the progress much =
at all but in my opinion it would probably have been better to publish =
it also as experimental first. On the other hand, it intends to =
implement NewReno as close as possible, and therefore not to change the =
congestion control principles that we have. I cannot tell how much =
experimental effort was put to verify the behaviour of those features =
that are new in QUIC and/or differ from stds track TCP but hopefully =
good enough. (Though there seems to be one issue related to resetting =
the exponential backoff of PTO, which seems to require a more accurate =
specification to ensure unambiguous interpretation and correct outcome, =
but that's an off topic here).</blockquote><div class=3D""><br =
class=3D""></div>RFC 9002 deviates from Reno as it uses `cwnd` during =
multiplicative decrease on a congestion event -&nbsp;<a =
href=3D"https://www.rfc-editor.org/rfc/rfc9002.html#name-on-new-congestion=
-event" =
class=3D"">https://www.rfc-editor.org/rfc/rfc9002.html#name-on-new-congest=
ion-event</a>. This is definitely a strong deviation from RFC 5681 which =
clearly mentions not to use cwnd.<div class=3D"">Its good to know that =
you are hopeful about the experimental effort put in RFC 9002 without =
the evidence being explicitly presented at IETF.</div><div class=3D""><br =
class=3D""></div><div class=3D"">Thanks,</div><div class=3D"">Vidhi<br =
class=3D""><div><br class=3D""><blockquote type=3D"cite" class=3D""><div =
class=3D"">On Feb 18, 2022, at 2:23 AM, Markku Kojo &lt;<a =
href=3D"mailto:kojo=3D40cs.helsinki.fi@dmarc.ietf.org" =
class=3D"">kojo=3D40cs.helsinki.fi@dmarc.ietf.org</a>&gt; =
wrote:</div><br class=3D"Apple-interchange-newline"><div class=3D""><div =
class=3D"">Hi Yoshi,<br class=3D""><br class=3D"">On Fri, 18 Feb 2022, =
Yoshifumi Nishida wrote:<br class=3D""><br class=3D""><blockquote =
type=3D"cite" class=3D"">Hi Markku,<br class=3D"">Thanks for the =
comments. Yes, these are valid and important concerns. However, I =
have<br class=3D"">mixed feelings about them.<br class=3D"">I don't want =
to open a can&nbsp;of worms, but here's are some of my personally =
thoughts...<br class=3D"">One thing is that "long and wide deployment =
experience" might not be a very strong claim,<br class=3D"">but then, I =
am wondering what would be the strong one.<br class=3D"">One may bring =
some solid experiment results, but the others might argue it uses too<br =
class=3D"">convenient configurations, or the scale is too small or too =
short.<br class=3D""></blockquote><br class=3D"">Well, that has always =
been and IMO still is for the wg to decide when the evidence through =
experiments is good enough to proceed with publishing the doc. E.g., =
when RFC 5562 was considered for publication by the wg, the wg requested =
for more experiments before making the decision and Sally provided what =
was requested. And it was just about adding ECN capability to SYNACK but =
here we are discussing much more significant changes.<br class=3D""><br =
class=3D"">And more importantly, here we are mainly discussing whether =
there is any experimental evidence at all; for example, changing MD =
(from 0.5 to 0.7 or to whatever) must be considered separately when in =
slow start and when in congestion avoidance as I have pointed out. This =
requires tests that specifically target at studying the slow start =
overshoot phase and the following recovery phase. AFAIK, there are no =
such experimental results, no analysis nor evaluation presented to =
support the change when the sender is in slow start. So, we are not =
discussing the level of evidence but whether it exists at all, which is =
very different.<br class=3D""><br class=3D""><blockquote type=3D"cite" =
class=3D"">The other thing is that if we set a high bar for a PS doc =
related to CC, we should apply<br class=3D"">the same bar to all related =
docs.For example, I am personally wondering how much we have<br =
class=3D"">analyzed and performed experiments when we published RFC9002. =
(BTW, my intention is NOT to<br class=3D"">accuse a specific doc, there =
should be some more examples.)&nbsp;<br class=3D""></blockquote><br =
class=3D"">Yes, IMO the bar should be the same. RFC 9002 was first =
targetted as experimental but was later changed to PS. I did not have =
time to follow the progress much at all but in my opinion it would =
probably have been better to publish it also as experimental first. On =
the other hand, it intends to implement NewReno as close as possible, =
and therefore not to change the congestion control principles that we =
have. I cannot tell how much experimental effort was put to verify the =
behaviour of those features that are new in QUIC and/or differ from stds =
track TCP but hopefully good enough. (Though there seems to be one issue =
related to resetting the exponential backoff of PTO, which seems to =
require a more accurate specification to ensure unambiguous =
interpretation and correct outcome, but that's an off topic here).<br =
class=3D""><br class=3D""><blockquote type=3D"cite" class=3D"">I am not =
sure how we can maintain this kind of fairness between the docs at the =
moment.<br class=3D""></blockquote><br class=3D"">I think it is the =
responsibility for the entire tsv area to do it, and eventually the TSV =
ADs and IESG. The process to follow is clear and robust, and when =
followed it has shown to be effectual in producing good quality =
documents. AFAIK, we have not decided to change it, maybe just many of =
us are unaware of it or have forgotten its existence?<br =
class=3D"">Traditionally tsv area has been very careful in considering =
new CC proposals because congestion control is so essential for the =
stability and fairness of the Internet and I hope we are not sliping too =
much from there. AFAIK, that's one reason why the process was =
documented.<br class=3D""><br class=3D"">Thanks,<br class=3D""><br =
class=3D"">/Markku<br class=3D""><br class=3D""><blockquote type=3D"cite" =
class=3D"">Thanks,<br class=3D"">--<br class=3D"">Yoshi&nbsp;<br =
class=3D"">On Thu, Feb 17, 2022 at 6:41 AM Markku Kojo &lt;<a =
href=3D"mailto:kojo@cs.helsinki.fi" class=3D"">kojo@cs.helsinki.fi</a>&gt;=
 wrote:<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Hi Yoshi,<br =
class=3D""><br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;On Tue, 15 Feb =
2022, Yoshifumi Nishida wrote:<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; Hi&nbsp;Markku,<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; Thanks for the comments. I think =
these are very valid points.&nbsp;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; However, I would like to check =
several things as a co-chair and a doc<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;shepherd before we<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; discuss&nbsp;the points =
you've&nbsp;raised.<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; In my understanding =
(please correct me if I'm wrong), when this draft was<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;adopted as an WG<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; item, I think the goal of the doc was =
some minor updates from RFC8312 which<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;include more<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; clarifications, minor&nbsp;changes =
and bug fixes.&nbsp;<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; =
However, if we try to address your&nbsp;concerns, I think we'll need to =
invent a<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;new version of<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; CUBIC something like =
CUBIC++ or NewCUBIC in the end.&nbsp;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; I won't deny the value of such doc, =
but, this seems not to be what we agreed<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;on<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; beforehand.&nbsp;&nbsp;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; if we proceed in this direction, I =
think we will need to check the WG<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;consensus whether<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; this should be a new goal for the =
doc.<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; So, I would like to check if this is =
what you intend for the doc or you<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;think we can<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; address your points while aligning =
with the original goal.<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; Also, if someone has opinions on =
this, please share.<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;I think it is important that we remember =
the status of RFC 8312 and the<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;decades long process that has been =
followed in tsv area for new<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;TCP congestion control algorithms that =
have been proposed and submitted<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;to IETF. In order to ensure that new cc =
algos are safe and fair, the<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;process that has been followed for all =
current stds track TCP cc algos<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;has required that the cc algo is first =
accepted and published as<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;experimental RFC and only once enough =
supportive experimental evidence<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;has been gathered the doc has become a =
candidate to be forwaded to stds<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;track. We have even agreed on a relatively =
strict evaluation process to<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;follow when cc algos are brought to the =
IETF to be published as<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;experimental:<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;<a =
href=3D"https://www.ietf.org/about/groups/iesg/statements/experimental-con=
gestion-control/" =
class=3D"">https://www.ietf.org/about/groups/iesg/statements/experimental-=
congestion-control/</a><br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;RFC 8312 was published as "Informational" =
and if I recall correctly the<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;idea was "just to publish what's out =
there" for the benefit of the<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;community. RFC 8312 was never really =
evaluated, particularly not in the<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;way new cc algos are supposed to be as per =
the agreed process.<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;I do not recall what/how exactly was =
agreed when rfc8312bis was launched<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;but I would be very interested to hear the =
justification why this doc<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;does not need to follow the process =
mentioned above but we would like to<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;propose IETF to publish a non-evaluated =
Informational doc to be published<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;"with minor updates", i.e., without actual =
evaluation, as a stds track<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;RFC? If the target really remains as PS =
then the bar should be even<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;higher than what is described for =
experimental docs in the above process<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;document, i.e, what we have followod for =
experimental to be moved to stds<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;track.<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;The only justification that I have heard =
has beed "because CUBIC has long<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;and wide deployment experience" and "the =
Internet has not smelted or that<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;"we should have noticed if there were =
problems". We must, however,<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;understand that in order to have =
noticeable bad impact CUBIC should cause<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;some sort of congestion collapse. =
Congestion collapse, however, is not an<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;issue with CUBIC nor with any other CC =
algo that applies an RTO mechanisms<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;together with correctly implemented Karn's =
algo that retains the<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;backed-off RTO until an Ack is received =
for a new (not rexmitted) data<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;packet. The issue is fairness to competing =
traffic. This cannot be<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;observed by deploying and measuring the =
performance and behaviour of CUBIC<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;alone. CUBIC being more aggressive than =
current stds track TCP CC would<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;just gives good performance results that =
one running CUBIC would be happy<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;with. One must evaluate CUBIC's impact on =
the competing (Reno CC) traffic<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;in range of environments which requires =
carefully designed active<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;measurements with thoroughly-analyzed =
results (as required by the above<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;process document, RFC 5033 and RFC 2914). =
What we seem to be missing is<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;this evidence on CUBIC's impact and that =
is something the IETF must focus<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;on, not just that whether CUBIC can =
achieve better performance than other<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;existing CCs. The latter has been shown in =
many publications and is the<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;majos focus in&nbsp; many scientific =
papers proposing new algos.<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;I appreciate a lot that CUBIC has been =
implemented/developped and<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;deployed for long and I wonder whether =
those deploying CUBIC have<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;unpublished results the wg could review =
before taking the decicion?<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;I suggest everyone to read carefully RFC =
2914 Sec 3.2 and particularly<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;what it says about more aggressive (than =
RFC 5681) congestion control<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;algorithms:<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Some of these may fail to =
implement<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; the TCP =
congestion avoidance mechanisms correctly because of poor<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; implementation [RFC2525].&nbsp; =
Others may deliberately be implemented<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; with congestion avoidance =
algorithms that are more aggressive in<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; their use of bandwidth than other =
TCP implementations; this would<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; allow a vendor to claim to have a =
"faster TCP".&nbsp; The logical<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; consequence of such implementations =
would be a spiral of increasingly<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; aggressive TCP implementations, or =
increasingly aggressive transport<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; protocols, leading back to the =
point where there is effectively no<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; congestion avoidance and the =
Internet is chronically congested.<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;And:<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; It is convenient to divide flows =
into three classes: (1) TCP-<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; compatible flows, (2) unresponsive =
flows, i.e., flows that do not<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; slow down when congestion occurs, =
and (3) flows that are responsive<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; but are not TCP-compatible.&nbsp; =
The last two classes contain more<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; aggressive flows that pose =
significant threats to Internet<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; performance,<br class=3D""><br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;As I have tried to point out =
there are several features with CUBIC where<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;it is likely to be (or to me it seems it =
obviously is) more aggressive<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;than what is reguired to be =
TCP-compatible. I'm not aware of evidince<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;presented to tcpm (or IETF/IRTF) which =
shows opposite (and I happy to be<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;educated what I have missed).<br =
class=3D""><br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;You may take my =
comments to be a part of the expert review phase<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;performed by the IRTF/ICCRG for CUBIC. I'm =
not requesting to modify this<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;doc to CUBIC++ (or something) but it seems =
to be that this would be<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;necessary if this doc intends to become =
published as PS. For experimental,<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;I think it would need some addtioinal =
updates and record the areas<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;uncertainty and where more experimentation =
(clearly) is required.<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Thanks,<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;/Markku<br class=3D""><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; Thanks,<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; --<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; Yoshi<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt; On Fri, Feb 11, 2022 at 9:34 PM =
Markku Kojo &lt;<a href=3D"mailto:kojo@cs.helsinki.fi" =
class=3D"">kojo@cs.helsinki.fi</a>&gt; wrote:<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;Hi Yoshi, =
all,<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;It seems to =
me that many issues that I raised have been solved,<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;thanks.<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;However, =
there are still a number of important issues that have not<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;been<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;addressed =
adequately. At least the following:<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;#135 on =
W_max. Yoshi's observation was correct that this is not<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;resolved:<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;the =
co-authors and original developpers of CUBIC (@lisongxu and<br class=3D"">=
 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;@sangtaeha) agreed in their last message that Wmax needs =
different<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;treatment for slow start and congestion avoidance and =
plan<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;comprehensive<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;(new) evaluation of it. This is obviously an open issue but the =
issue<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;was closed?<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;#85 (&amp; =
#86 with basically same issue and these two were combined) This<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;(#85) is about ECN but the major issue is on using the same MD<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;factor in slow start and in congestion avoidance when using<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;loss-based CC. This (#85) remained closed even though I provided =
a<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;thorough explanation why it is wrong and against the original =
theory<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;and<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;design by =
Van Jacobson, against the congestion control principles (RFC<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;2914) and two co-authors agreed on this in their same last message =
to<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;#135 when they agreed on Wmax needing rework. This is an =
important<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;issue<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;that the wg should consider very carefully because it is not =
only<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;updating RFC 5681 but also in conflict with RFC 2914. How =
can<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;tcpm (and IETF) suggest and allow one CC algo to not follow =
congestion<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;control principles as set in RFC 2914 while =
requiring all other CCs to<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;follow RFC =
2914 guidelines?<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;The current draft does not provide any justification =
for using the<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;same<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;MD factor in slow start as in congestion avoidance. Nor am I<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;awere of any experimental data that would support this change.<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;The fact thet CUBIC has been long deployed does not alone provide =
any<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;supporting evidence because CUBIC is likely to give good =
performance<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;as<br class=3D"">=
 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;it is =
overagressive and thereby unfair to competing traffic and users<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;tend to be happy when measuring the performance of the sending =
CUBIC<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;only, not the competing traffic that is badly impacted. =
HyStart++ is<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;suggested as mitigation to the problem but it =
cannot; HyStart++ is<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;only<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;applicable =
during initial slow start, not during slow start after RTO!<br class=3D"">=
 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;That is, =
the "SHOULD use HYStart++" text in Sec 4.10 is impossible<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;to =
implement as I have pointed out in my comments earlier. Using a<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;proper<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;MD factor =
in slow start is even more important if loss is detectected<br class=3D"">=
 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;during a =
RTO recovery because the sender is likely to face heavy<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;congestion =
in such a case and it is very bad if the sender continues<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;sending =
with overaggressive rate, stealing the capacity from and<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;causing<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;harm to =
coexisting flows. In addition, as I have explained, HyStart++<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;does not remove the problem even for the initial slow start as it =
is<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;not<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;shown to =
work always. Instead, the results with the HyStart++ draft<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;show<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;that it =
reduces 50% of rexmits and only 36% RTOs, meaning that there<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;is<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;likely to =
be a notable percentage of cases when a sender is still<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;in slow =
start when first loss is detected (i.e., HyStart++ had no<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;effect)<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;and a =
significant number of cases where a CUBIC sender is<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;overaggressive continuing with a 1-40% larger cwnd than what is =
the<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;available capacity. Note also that any delay-based =
heuristics like<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;HyStart++ are known to work poorly in various =
wireless environmens<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;where<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;link delay =
tends to vary a lot. We may come up with some other MD<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;factor<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;that 0.5 =
when in slow start and HyStart++ is in use, but that is<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;experimental, if not research, and definitely not ready for =
stds<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;track.<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;#114, #132, =
and #143 w.r.t flightsize vs. cwnd. The current text<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&nbsp; =
&nbsp; &nbsp; does not quite correcly reflect what stacks that use cwnd =
do.<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&nbsp; &nbsp; &nbsp; I'll comment in #143 separately.<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;#96 &amp; =
#98: The text added does not address the problems raised which<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;are<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;also =
evidenced in the paper pointed by Bob in #96. Even though CUBIC<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;has<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;been =
modified a bit after the paper was published, it does not<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;automatically mean that the problem has been shown resolved:<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;experimental<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;evidence is =
required but not provided. The fact that CUBIC does not<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;change MD =
factor for fast convergence is the root of the problem<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;evidenced =
in the paper and remains so in the algo specified in this<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;draft. This =
is also a significant problem when competing with Reno CC<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;because =
CUBIC behaves much more aggressive than Reno CC when there is<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;sudden congestion and all competing flows must converge fast down =
to a<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;small fraction of the current cwnd to be fair to each =
other. This<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;again<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;cannot be evidenced not to be a problem by long deployment =
experience<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;unless experimental data that measures the impact on =
competing traffic<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;is<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;presented to back the claims. Adjusting just Wmax for fast =
convergence<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;is<br class=3D"">=
 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;not enough =
and is even likely to be ineffective because there tend to<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;be<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;several =
losses when sudden congestion is hit, and particularly if<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;NewReno<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;is in use =
the sender stays several RTTs in Fast Recovery being<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;overaggressive and then possibly continues at the same rate in =
CA<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;which<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;is unlikely =
to reach evan close to Wmax before a new loss hits<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;the sender =
again. That is, lower Wmax and lower additive increase<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;factor<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;do not =
compensate the use of larger MD factor when sudden congestion<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;is<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;encountered.<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;#93 &amp; #94 &amp; (#89) Sec 5.3 still does not address any =
difficult<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;environments, in particular buffer-bloated paths (nor does =
Sec 5.4).<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;We need evidence (results) that show CUBIC is fair towards =
other CCs<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;(Reno) also in such environments. Note that CUBIC's =
decision to leave<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;Reno-friendly region is based on the size of cwnd which tends to =
be<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;incorrect with buffer-bloated bottlenecks because with huge =
buffers<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;the<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;cwnd can be =
many times larger than what is actually needed to fully<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;utilize the =
available network bit-rate. Therefore, Reno CC has no<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;problem<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;in fully =
utilizing such bottleneck links and CUBIC must stay in<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;Reno-friendly region longer but it leaves it too early because =
the<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;same C<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;is used as =
with non-bloated environments. We lack experiments showing<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;CUBIC =
follows the congestion control principles and is fair to current<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;standard TCP CC; to my understanding no experiments with<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;buffer-bloated<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;bottlenecks =
are cited to back up the claims even though buffer bloat<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;is<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;very well =
known to be a common (difficult) environment in today's<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;Internet.<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;#90 The current text on applying undo (a response to detected =
false<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;fastrexmit) does not provide correct result if someone =
implements it.<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;I have explained the problems there in github but =
seem to have not<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;replied to latest comments by Neal. I'll reply and =
try to explain<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;more.<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;Again one major problem here is that the draft suggest a new =
response<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;algo for false fast rexmits but does not provide any =
experimental data<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;to<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;support it. Long deployment experience has been suggested as<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;justification but again without any carefully evaluated =
experimental<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;data and evidence there is no meat. The issue is =
important to solve<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;but =
is<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;not specific to CUBIC. Instead, it is general problem for all TCP =
CC<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;variants. IMHO, this is not ready for standards track but deserves =
a<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;draft<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;of its own =
so that it can be carefully evaluated and discussed ion the<br class=3D"">=
 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;list. =
AFAIK there has been no discussion on this on the tcpm list, so<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;those probably interested and having experience are likely to =
be<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;unaware<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;that this =
is part of CUBIC draft.<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;#88 The =
problem with correctness of the AIMD model and setting alpha<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;for<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;CUBIC =
requires further consideration. Bob provided an analysis that<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;leaves things still open. It seems that I never had time to review =
and<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;comment the analysis and clarify why the model does not =
work. I'll do<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;that<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;separately as it is important to ensure CUBIC behaves fairly as<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;intended<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;for the =
Reno-friendly region.<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;Best =
regards,<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;/Markku<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;On Mon, 31 =
Jan 2022, Yoshifumi Nishida wrote:<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt; =
Hello,<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt; After =
some discussions among chairs, we decided to run the 2nd WGLC<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;on<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;draft-ietf-tcpm-rfc8312bis in<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt; =
consideration of the importance of the draft.&nbsp;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt; We'll =
be grateful if you could send your feedback to the ML. The<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;WGLC runs<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;until *Feb =
11*.<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt; If =
interested, you can check in-depth past discussions in the<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;following URL.<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt; <a =
href=3D"https://github.com/NTAP/rfc8312bis/" =
class=3D"">https://github.com/NTAP/rfc8312bis/</a><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt; Thank you so much!<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt; --<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt; tcpm co-chairs<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert =
&lt;<a href=3D"mailto:lars@eggert.org" class=3D"">lars@eggert.org</a>&gt; =
wrote:<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;Hi,<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;this -06 version rolls in all the =
changes requested during<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;(and after)<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;WGLC =
ended.<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;I'll leave it up to the chairs to decide if another =
WGLC is<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;warranted<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;or the document can<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;progress as-is.<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;Thanks,<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;Lars<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt; On 2022-1-26, at =
11:12, <a href=3D"mailto:internet-drafts@ietf.org" =
class=3D"">internet-drafts@ietf.org</a> wrote:<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; A New Internet-Draft is available from the =
on-line<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Internet-Drafts<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;directories.<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; This draft is a work item of the TCP =
Maintenance and Minor<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;Extensions =
WG of the IETF.<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp; Title&nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp;: CUBIC for Fast and Long-Distance<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Networks<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp; Authors&nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp;: Lisong Xu<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; Sangtae Ha<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; Injong Rhee<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; Vidhi Goel<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; Lars Eggert<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;Filename&nbsp; &nbsp; =
&nbsp; &nbsp; : draft-ietf-tcpm-rfc8312bis-06.txt<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;Pages&nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp;: 35<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;Date&nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; : 2022-01-26<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; Abstract:<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;CUBIC is a standard TCP congestion =
control algorithm that<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;uses =
a<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;cubic function =
instead of a linear congestion window<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;increase<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;function to improve scalability and =
stability over fast<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;and =
long-<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;distance =
networks.&nbsp; CUBIC has been adopted as the default<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;TCP<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;congestion control algorithm by the =
Linux, Windows, and<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Apple<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;stacks.<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;This document updates the =
specification of CUBIC to<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;include<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;algorithmic improvements based on =
these implementations<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;and =
recent<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;academic =
work.&nbsp; Based on the extensive deployment<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;experience with<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;CUBIC, it also moves the =
specification to the Standards<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Track,<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;obsoleting RFC 8312.&nbsp; This =
also requires updating RFC<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;5681, to<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;allow<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; &nbsp;for CUBIC's =
occasionally more aggressive sending behavior.<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; The IETF datatracker status page for this draft =
is:<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt; <a =
href=3D"https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/" =
class=3D"">https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/</a=
><br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; There is also an HTML version available at:<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;<a =
href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html=
" =
class=3D"">https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.h=
tml</a><br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; A diff from the previous version is available =
at:<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;<a =
href=3D"https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06"=
 =
class=3D"">https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-=
06</a><br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; =
&nbsp; &nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; Internet-Drafts are also available by rsync =
at<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;<a href=3D"http://rsync.ietf.org" =
class=3D"">rsync.ietf.org</a>::internet-drafts<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; =
_______________________________________________<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; tcpm mailing list<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; <a href=3D"mailto:tcpm@ietf.org" =
class=3D"">tcpm@ietf.org</a><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt; <a =
href=3D"https://www.ietf.org/mailman/listinfo/tcpm" =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm</a><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;_______________________________________________<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;tcpm mailing list<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;<a href=3D"mailto:tcpm@ietf.org" =
class=3D"">tcpm@ietf.org</a><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;<a href=3D"https://www.ietf.org/mailman/listinfo/tcpm"=
 class=3D"">https://www.ietf.org/mailman/listinfo/tcpm</a><br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; &nbsp;&gt;<br =
class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; &nbsp; &nbsp; =
&nbsp;&gt;<br class=3D""> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;&nbsp; =
&nbsp; &nbsp; &nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br class=3D""> =
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&gt;<br =
class=3D""></blockquote>_______________________________________________<br=
 class=3D"">tcpm mailing list<br class=3D""><a =
href=3D"mailto:tcpm@ietf.org" class=3D"">tcpm@ietf.org</a><br =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm<br =
class=3D""></div></div></blockquote></div><br =
class=3D""></div></body></html>=

--Apple-Mail=_DAA60ABD-F9B5-4889-96E7-C9BC6BD2C5AA--

--Apple-Mail=_5612D5A8-FDCE-4B25-BE35-46A10DD7B51A
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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=
--Apple-Mail=_5612D5A8-FDCE-4B25-BE35-46A10DD7B51A--


From nobody Fri Feb 18 06:46:57 2022
Return-Path: <ncardwell@google.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 03FA03A118F for <tcpm@ietfa.amsl.com>; Fri, 18 Feb 2022 06:46:49 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -22.599
X-Spam-Level: 
X-Spam-Status: No, score=-22.599 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_MED=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, ENV_AND_HDR_SPF_MATCH=-0.5, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001, USER_IN_DEF_DKIM_WL=-7.5, USER_IN_DEF_SPF_WL=-7.5] autolearn=unavailable autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=google.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hqUpHnpmF1PV for <tcpm@ietfa.amsl.com>; Fri, 18 Feb 2022 06:46:44 -0800 (PST)
Received: from mail-qk1-x733.google.com (mail-qk1-x733.google.com [IPv6:2607:f8b0:4864:20::733]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B76153A11A5 for <tcpm@ietf.org>; Fri, 18 Feb 2022 06:46:44 -0800 (PST)
Received: by mail-qk1-x733.google.com with SMTP id t21so1771633qkg.6 for <tcpm@ietf.org>; Fri, 18 Feb 2022 06:46:44 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=LVMacv+AUrAJ4y/8xS3E9+jCZxqcG81/7KrFqWwR6tM=; b=EfZUJgJltq8ikNTKujshX9bKTqyRnSnl0Se/8yYmEu+tI1wI9gbKC4IjlLuwuHbbCh sNP4fciEiPDLkarIXGU2LArZZ+HsQpuGQb0CrL2bH2NQOVfd76OkdCvT+68a+Qjd6bOS uZbI+fMdLoANJ6Fkm7xWXpkOSgJI2jmHHn6Y3rRFiGhFT32HtRXfehoYEd3X0ScOcC0C f8eTVh1cMUnVffdPvQieN+bH21U4Z/IDELGT5lr9lFBE6i0+ke+A5t6Iby0er2rO+6Og YxWz7Rt9oDI79I5bHEmOvds4ZLo7m2QXlUnogMk6V54KtRjezjG24ll4l9J7KWwMaZrE fXpA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=LVMacv+AUrAJ4y/8xS3E9+jCZxqcG81/7KrFqWwR6tM=; b=enAvp7Q/STTRg2iP8dre6qXDCxAcEPcdnSdauZI299Ao5nHP94mLO1Wd84Ow9m6Rqy /igLzM0y6XR914UlzZFTquda8pPNDsHmyQq9GAk1yOWiqAA8RLUNLxh6eabL18Wt6lSE bMuzRBHjZFhJaQTz86KoKtihNOiYx66UsBqwULgUMgE/5WMLA0wq7laR6JAH8UpXajDc f/Puaahu80dZJn/MbvqYbn9ijp82Mne6y2CQzYLwKp/fGzRj0S1ZJAccIMHt6Wq7uOyh 447ebPV3XagBEu68RMBTI7tGRcikjxijBMyyqCNzDjmixvutSOenZgo9S79Ir6IhDAZI Q1iA==
X-Gm-Message-State: AOAM5331RhKVIRatkuY3ioqWyI9uKPFT59mX6DgqYfwmK2q9gEQXfM0i KudStUqr1UR0CaGJL4J1qTBm1WD5MJMlXOAeWDk17A==
X-Google-Smtp-Source: ABdhPJyBBKDdq1brpdJgwuNQsCV0SjFluFunSHzPOxkxZHCLhtwSnN9vabi4xq4ifaQOE0xGTbjs+VNn0+VNsIXONBc=
X-Received: by 2002:a37:6943:0:b0:5e9:6ed4:2295 with SMTP id e64-20020a376943000000b005e96ed42295mr4754530qkc.259.1645195603017; Fri, 18 Feb 2022 06:46:43 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CADVnQymwCdm0NoD95SX8JpRXFok=qMygPrwaeTJ9z29dqH=Avw@mail.gmail.com> <alpine.DEB.2.21.2202181031280.4019@hp8x-60.cs.helsinki.fi>
In-Reply-To: <alpine.DEB.2.21.2202181031280.4019@hp8x-60.cs.helsinki.fi>
From: Neal Cardwell <ncardwell@google.com>
Date: Fri, 18 Feb 2022 09:46:26 -0500
Message-ID: <CADVnQyn55Ki3CCw=mshFQ=PAoQyR6MOpgdOaDgaN+JBtfNA57Q@mail.gmail.com>
To: Markku Kojo <kojo=40cs.helsinki.fi@dmarc.ietf.org>
Cc: Yoshifumi Nishida <nsd.ietf@gmail.com>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>,  tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: multipart/alternative; boundary="00000000000081919e05d84bf3b0"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/8Mwn7TjjrDMhBFEOYCO-xuPNEEo>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 18 Feb 2022 14:46:52 -0000

--00000000000081919e05d84bf3b0
Content-Type: text/plain; charset="UTF-8"

On Fri, Feb 18, 2022 at 5:26 AM Markku Kojo <kojo=
40cs.helsinki.fi@dmarc.ietf.org> wrote:

> Hi Neal,
>
> On Thu, 17 Feb 2022, Neal Cardwell wrote:
>
...

> > Hi Markku,
> >
> > Thanks for your careful argument above. Given your argument, in order to
> continue to make
> > constructive progress on rfc8312bis without blocking for new research
> and experimentation,
> > I would propose that we change rfc8312bis so that it remains
> "Informational" (like
> > RFC8312) rather than the "Standards Track" status currently listed at:
> >
> >   https://ntap.github.io/rfc8312bis/draft-ietf-tcpm-rfc8312bis.html
> >
> > As far as I can tell, that would allow all of the valuable contributions
> in rfc8312bis to
> > carry forward and be published, while still respecting all of the
> considerations you
> > enumerated above, which center around concerns with calling CUBIC
> "standards track".
> >
> > How does that sound?
>
> Sure, that could be one possible path forward. In my opinion, publishing
> it as experimental would be more logical and would also provide an
> inherent path forward to PS once enough evidence has been gathered and wg
> considers it is mature enough to proceed.  I don't have a strong opinion
> on this. Either way, IMO, it would be important to clearly point out and
> discuss the areas that require more evaluation so that it encourages
> interested parties to focus their experimentation to those important
> areas.
>

Great. Slating rfc8312bis as "Experimental" sounds like a great solution to
me.

And I agree it would be great to "clearly point out and discuss the areas
that require more evaluation", for the reason you note. To speed forward
progress, would you like to propose a github pull request to add an
appendix that enumerates those areas?

thanks,
neal

--00000000000081919e05d84bf3b0
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote">=
<div dir=3D"ltr" class=3D"gmail_attr">On Fri, Feb 18, 2022 at 5:26 AM Markk=
u Kojo &lt;kojo=3D<a href=3D"mailto:40cs.helsinki.fi@dmarc.ietf.org">40cs.h=
elsinki.fi@dmarc.ietf.org</a>&gt; wrote:<br></div><blockquote class=3D"gmai=
l_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,20=
4,204);padding-left:1ex">Hi Neal,<br>
<br>
On Thu, 17 Feb 2022, Neal Cardwell wrote:<br></blockquote><div>...=C2=A0</d=
iv><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;bord=
er-left:1px solid rgb(204,204,204);padding-left:1ex">
&gt; Hi Markku,<br>
&gt; <br>
&gt; Thanks for your careful argument above. Given your argument, in order =
to continue to make<br>
&gt; constructive progress on rfc8312bis without blocking for new research =
and experimentation,<br>
&gt; I would propose that we change rfc8312bis so that it remains &quot;Inf=
ormational&quot; (like<br>
&gt; RFC8312) rather than the &quot;Standards Track&quot; status currently =
listed at:<br>
&gt; <br>
&gt; =C2=A0 <a href=3D"https://ntap.github.io/rfc8312bis/draft-ietf-tcpm-rf=
c8312bis.html" rel=3D"noreferrer" target=3D"_blank">https://ntap.github.io/=
rfc8312bis/draft-ietf-tcpm-rfc8312bis.html</a><br>
&gt; <br>
&gt; As far as I can tell, that would allow all of the valuable contributio=
ns in rfc8312bis to<br>
&gt; carry forward and be published, while still respecting all of the cons=
iderations you<br>
&gt; enumerated above, which center around concerns with calling CUBIC &quo=
t;standards track&quot;.<br>
&gt; <br>
&gt; How does that sound?<br>
<br>
Sure, that could be one possible path forward. In my opinion, publishing <b=
r>
it as experimental would be more logical and would also provide an <br>
inherent path forward to PS once enough evidence has been gathered and wg <=
br>
considers it is mature enough to proceed.=C2=A0 I don&#39;t have a strong o=
pinion <br>
on this. Either way, IMO, it would be important to clearly point out and <b=
r>
discuss the areas that require more evaluation so that it encourages <br>
interested parties to focus their experimentation to those important <br>
areas.<br></blockquote><br>Great. Slating rfc8312bis as &quot;Experimental&=
quot; sounds like a great solution to me.</div><div class=3D"gmail_quote"><=
br></div><div class=3D"gmail_quote">And I agree it would be great to &quot;=
clearly point out and discuss the areas that require more evaluation&quot;,=
 for the reason you note. To speed forward progress, would you like to prop=
ose a github pull request to add an appendix that enumerates those areas?</=
div><div class=3D"gmail_quote"><br></div><div class=3D"gmail_quote">thanks,=
</div><div class=3D"gmail_quote">neal</div><div class=3D"gmail_quote"><br><=
/div></div>

--00000000000081919e05d84bf3b0--


From nobody Sun Feb 20 00:36:02 2022
Return-Path: <do_not_reply@mnot.net>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2DF133A0F5E for <tcpm@ietfa.amsl.com>; Sun, 20 Feb 2022 00:36:00 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.7
X-Spam-Level: 
X-Spam-Status: No, score=-1.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_INVALID=0.1, DKIM_SIGNED=0.1, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=no autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=fail (2048-bit key) reason="fail (message has been altered)" header.d=mnot.net header.b=lSFyLHaW; dkim=fail (2048-bit key) reason="fail (message has been altered)" header.d=messagingengine.com header.b=GxoaJ5+b
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eDNKs4JHz7sA for <tcpm@ietfa.amsl.com>; Sun, 20 Feb 2022 00:35:55 -0800 (PST)
Received: from wout2-smtp.messagingengine.com (wout2-smtp.messagingengine.com [64.147.123.25]) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C36FC3A0F48 for <tcpm@ietf.org>; Sun, 20 Feb 2022 00:35:55 -0800 (PST)
Received: from compute2.internal (compute2.nyi.internal [10.202.2.46]) by mailout.west.internal (Postfix) with ESMTP id E6FA13200E83 for <tcpm@ietf.org>; Sun, 20 Feb 2022 02:39:24 -0500 (EST)
Received: from mailfrontend2 ([10.202.2.163]) by compute2.internal (MEProxy); Sun, 20 Feb 2022 02:39:25 -0500
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mnot.net; h=cc :content-type:date:from:from:in-reply-to:mime-version:reply-to :sender:subject:subject:to:to; s=fm2; bh=AZdJWd/I97w6HTtGOYp5h4B kcYejyOOPsVj7bl0vxjo=; b=lSFyLHaWuGd3iuNyYbmluWs0WL2oT/bcrAEzWoE ZQxlJuWJtscYllBKb48mS76v9uVB1JHmMViGLSWW2epH8j+CEXai4SvnD1V3rC1O Q+6Oo4bC2ZX0N5cwXziwh40lH2h8Xz6Cv5vozxdbHKUnAqox5BlraS0xa2RFDkwV ZjVUiD1hBJ8LS5C12s7/5CDquG4rALlYluzmBVQmdr3d9B80xRI3Y2GDWKXmUaJV T4xlspbWUf1NT50EYyKD2XbBRs2QySyPp2NdzVh1s0wFn2r9hw9CEw/XpIIomlMB BAvNYqZTA7AKbE5i9IUQgVzF6vK2gLTfy7VTuzOb8sWEhoQ==
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=cc:content-type:date:from:from :in-reply-to:mime-version:reply-to:sender:subject:subject:to:to :x-me-proxy:x-me-proxy:x-me-sender:x-me-sender:x-sasl-enc; s= fm2; bh=AZdJWd/I97w6HTtGOYp5h4BkcYejyOOPsVj7bl0vxjo=; b=GxoaJ5+b DYENp1UIWwaozBrts4ztofvGsxrNq32Veg7SD3DcLqsLCBTheZ7e1y9pDs1braMc hzhxdIgm48oBnC4RhuJSBWwWtzsKarj0rgKfxT3Y9Fd9vQs8xueiJNxz2QDKDOcI 3G90C2DTSsMLbkh18jViaPkS8+YJYw17BGNUeOnZAoHgKfNR68em3tnWy9HnJZf/ fy4MQ3KuhUZcxGSVrZzO/LLBb8By0YJ88IAo5lIrikDaPgOEdKDeGdFkljhOjBg/ ZQ6cKiTfnwIybQFTNndGziEfbMK803ODEGzglX/77W8YavexWrhsgW1PzFX6rPVG whfW8TMZNl687w==
X-ME-Sender: <xms:LPARYt-9Dun8GUD879aHNZrOUibzqHHhNlz8cXdJRtxG9H6ZeqvzVg> <xme:LPARYhtOZy1c8aSd3KtXrxW05Nyxr3Vkx1lvMcX5J2Ct81JMH3sse0BcpH6aBpkwD YFURbfPRP7ZEMLAzQ>
X-ME-Received: <xmr:LPARYrDbbIcAcaaVdfLJMdKwbhmfv9mUZw5vnr1Q7G_DYMt103j6lqudZHiGQTMpkG7u5C2SMTQqLINq_qwSs9dTtmoCgdaTW1508jDlutIxSI9RKIC5BgfIYkMtQJAWljg>
X-ME-Proxy-Cause: gggruggvucftvghtrhhoucdtuddrgedvvddrkeefgddutdelucetufdoteggodetrfdotf fvucfrrhhofhhilhgvmecuhfgrshhtofgrihhlpdfqfgfvpdfurfetoffkrfgpnffqhgen uceurghilhhouhhtmecufedttdenucfpohcuuggrthgvuchfihgvlhguucdlgeelmdenuc fjughrpegtggfhvffusegrtddtredttdejnecuhfhrohhmpeftvghpohhsihhtohhrhicu tegtthhivhhithihucfuuhhmmhgrrhihuceuohhtuceoughopghnohhtpghrvghplhihse hmnhhothdrnhgvtheqnecuggftrfgrthhtvghrnhepkeefvdduteejvdefkeehieevuefg fefhteetveegffekffefteffvdelheduieetnecuffhomhgrihhnpehgihhthhhusgdrtg homhenucevlhhushhtvghrufhiiigvpeehnecurfgrrhgrmhepmhgrihhlfhhrohhmpegu ohgpnhhothgprhgvphhlhiesmhhnohhtrdhnvght
X-ME-Proxy: <xmx:LPARYhcW3eE7x3W3GnGiH8Iyk7Nu8PE-LJ5IVir9gNu9RQOZG6agFw> <xmx:LPARYiPWclhtuTZiOMtqPtVJDnJlYrEN9gISFJFss9iZdpj9cDCD2w> <xmx:LPARYjk7cxjSqKyx8Odrz_lkzoh_W5JDXAbMU5HmyIo7hjgqyKgTeg> <xmx:LPARYrZkpyZKe2A5-uunRl_2FPN_0sjxQs-3Z3tEWhX_w3Pu5_X7aQ>
Received: by mail.messagingengine.com (Postfix) with ESMTPA for <tcpm@ietf.org>; Sun, 20 Feb 2022 02:39:24 -0500 (EST)
Content-Type: multipart/alternative; boundary="===============5528044144094898345=="
MIME-Version: 1.0
From: Repository Activity Summary Bot <do_not_reply@mnot.net>
To: tcpm@ietf.org
Message-Id: <20220220083555.C36FC3A0F48@ietfa.amsl.com>
Date: Sun, 20 Feb 2022 00:35:55 -0800 (PST)
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/OImTdqB391Bp80tD_kVNsrBdR4I>
Subject: [tcpm] Weekly github digest (RFC8312bis Activity Summary)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 20 Feb 2022 08:36:01 -0000

--===============5528044144094898345==
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="utf-8"; format="flowed"




Events=20

Issues
------
* NTAP/rfc8312bis (+0/-0/=F0=9F=92=AC1)
  1 issues received 1 new comments:
  - #143 Clarification about cwnd and flight size when referring to RFC 766=
1 (1 by goelvidhi)
    https://github.com/NTAP/rfc8312bis/issues/143=20



Pull requests
-------------
* NTAP/rfc8312bis (+1/-0/=F0=9F=92=AC0)
  1 pull requests submitted:
  - add text to explain why 7661 is safe to use even when cwnd > rwnd (by g=
oelvidhi)
    https://github.com/NTAP/rfc8312bis/pull/144=20


Repositories tracked by this digest:
-----------------------------------
* https://github.com/NTAP/rfc8312bis

--===============5528044144094898345==
Content-Type: text/html; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable

<!doctype html>
<html lang=3D"en">
<head>
<meta charset=3D"utf-8">
<title>Weekly github digest (RFC8312bis Activity Summary)</title>
<style>
body { font-family: Gotham, "Helvetica Neue", Helvetica, Arial, sans-serif;=
 font-size: 14px; }
h2 { margin-top: 3em; color: #A52A2A; font-style: italic; font-weight: norm=
al; }
h3 { margin-bottom:0; margin-top: 2em; font-size: 1.2em; }
h1+h2 { margin-top: 1em; }
a { color: #bb6219; text-decoration: none; }
li { margin-bottom: .35em; }
.repos { margin-bottom: 0; margin-top:0; line-height: 1.2; }
.new { color: red; }
.label { display: inline;
	padding: .2em .6em .3em;
	font-size: 75%;
	font-weight: 700;
	line-height: 1;
	color: #fff;
	text-align: center;
	white-space: nowrap;
	vertical-align: baseline;
	border-radius: .25em;
}
details {
	margin-top: 8em;
	}
summary {
	margin-bottom: 1em;
	cursor: pointer;
}
</style>
</head>

<body>
<h1>Sunday February 20, 2022</h1>

<p>Events </p>

<h2>Issues</h2>

<h3>NTAP/rfc8312bis (+0/-0/=F0=9F=92=AC1)</h3>

  <p>1 issues received 1 new comments:</p>
  <ul>
  <li>#143 <a href=3D"https://github.com/NTAP/rfc8312bis/issues/143">Clarif=
ication about cwnd and flight size when referring to RFC 7661</a> (1 by goe=
lvidhi) </li>
  </ul>




<h2>Pull requests</h2>
<h3>NTAP/rfc8312bis (+1/-0/=F0=9F=92=AC0)</h3>
  <p class=3D"new">1 pull requests submitted:</p>
  <ul>
  <li>#144 <a href=3D"https://github.com/NTAP/rfc8312bis/pull/144">add text=
 to explain why 7661 is safe to use even when cwnd &gt; rwnd</a> (by goelvi=
dhi) </li>
  </ul>




  <details>
    <summary>Repositories tracked by this digest:</summary>
<ul class=3D"repos">
  <li><a href=3D"https://github.com/NTAP/rfc8312bis">https://github.com/NTA=
P/rfc8312bis</a></li>
</ul>
</details>
</body>
</html>

--===============5528044144094898345==--


From nobody Mon Feb 21 02:43:25 2022
Return-Path: <nsd.ietf@gmail.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 14BE43A0E6D; Mon, 21 Feb 2022 02:43:23 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.096
X-Spam-Level: 
X-Spam-Status: No, score=-2.096 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fAB6uGqGRA3t; Mon, 21 Feb 2022 02:43:17 -0800 (PST)
Received: from mail-qv1-xf2b.google.com (mail-qv1-xf2b.google.com [IPv6:2607:f8b0:4864:20::f2b]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 305F03A0E6B; Mon, 21 Feb 2022 02:43:17 -0800 (PST)
Received: by mail-qv1-xf2b.google.com with SMTP id f19so30953931qvb.6; Mon, 21 Feb 2022 02:43:17 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112;  h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=GVvlkaBum0hBRfH+LBmTG8LFMZ0px5JUP3FhhnJTyag=; b=aEu6qd/a9j4MiuE9L9AGLSRbg6UZyBjbdvTtOuhHzw18Png3IWuRnsCIELKCdv1NtR FnobERtsFfPj6uutIv3pMttjXINaOoMuYVTbTqgQ+7u8OtPYCNIQFxfEFUdlL66LeaAV FK+lnhotA7hhRSUAFBNOkDqAh4zWkU10LyIcaPGH3zXrKJMthoFFCiciIDP+Ks+u0283 tRl+CXPkZbCrzLQSKGxZlCaFY+MF7D2Y+VdjfQuWUA9ej/dPJ/rU2NWrgzFKvT61qp5m mcz7dYu7X0C1ZFaNnp0kDENUCOMxvNtFp0voffHF2Q9eboC2ebK6o81x5zYG/Roz9Y8W 4k4Q==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=GVvlkaBum0hBRfH+LBmTG8LFMZ0px5JUP3FhhnJTyag=; b=pQqCagUNw3/zWAw6Cwb6tDKR0feP9zaNLAtFTZ8eVK6ENrGba4EqinAG2zSLzyilri quai3BrbrDJaGk2LH6ucVZo0qJjpJmOT57flY/7lop0SnNmLHAI/Irt+Lq2GN/5CAtlV cZ64UEEN4ONeG20PmLjvEK7sSRv1OJj+GEwy8r6a6lFnGFeu38lL3TlslX58i5vrNqaV m+31PDJ2p5JNhnpyCuCw3q282DBLGAF04cLsSymCHhPNKD720cbrvulclFpTiuUTfWS2 jhe1Trcyh2/qKVH3BF4Dhl7lucw08N/WN4Wqf8YREJRSWkqEbBZPht/BORxwMd4QjW6v dk9Q==
X-Gm-Message-State: AOAM532TtAI5jYq6Tfmny9BTgM3XyVYxc/WmMbcHj6IynTeOjqV2XTzM g6qWo3YJ1SUrJSnsPBlnMYE1Ra8VSKqOUt2YfSpaTuHN0OY=
X-Google-Smtp-Source: ABdhPJyIQxPTdVJ2HsoLm6KYE/2uqrS/hHI9UBV6XWhV3ZVgktd3SFWK4ppmZxtkQYFaCjqRdtGM/tUrQVddoxYWUN8=
X-Received: by 2002:ac8:7d90:0:b0:2c8:5e5c:aeb5 with SMTP id c16-20020ac87d90000000b002c85e5caeb5mr17005621qtd.528.1645440195865; Mon, 21 Feb 2022 02:43:15 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi>
In-Reply-To: <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi>
From: Yoshifumi Nishida <nsd.ietf@gmail.com>
Date: Mon, 21 Feb 2022 02:43:04 -0800
Message-ID: <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com>
To: Markku Kojo <kojo@cs.helsinki.fi>
Cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: multipart/alternative; boundary="0000000000005fadfc05d884e636"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/69LFo79OB_MCxzz6FpZ7Mx6iLoI>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 21 Feb 2022 10:43:24 -0000

--0000000000005fadfc05d884e636
Content-Type: text/plain; charset="UTF-8"

Hi Markku,

I've thought about this over the weekend. Changing the intended status of
the doc may look simple, but it might affect other docs as a precedent.
So, I would like to think about it a bit more.

I think it may be possible to do some experiments before we publish the doc
as a PS if it's not too complicated.
But, in this case, I think we'll need to delineate what we want to
evaluate.
One thing I am wondering is whether points are if there is a risk for
congestion collapse, or if this is too aggressive than NewReno or something
else.

If we think there's a risk for congestion collapse, we should be able to
describe which scenarios could lead such situations.
If we think CUBIC is too aggressive, I think we will need to define some
bars for it.
I think 9002 is more aggressive than TCP NewReno in some points.
So, if we reject CUBIC to be a PS due to its aggressiveness, I think there
should be a border for it.

Thanks,
--
Yoshi

On Fri, Feb 18, 2022 at 2:23 AM Markku Kojo <kojo@cs.helsinki.fi> wrote:

> Hi Yoshi,
>
> On Fri, 18 Feb 2022, Yoshifumi Nishida wrote:
>
> > Hi Markku,
> >
> > Thanks for the comments. Yes, these are valid and important concerns.
> However, I have
> > mixed feelings about them.
> > I don't want to open a can of worms, but here's are some of my
> personally thoughts...
> >
> > One thing is that "long and wide deployment experience" might not be a
> very strong claim,
> > but then, I am wondering what would be the strong one.
> > One may bring some solid experiment results, but the others might argue
> it uses too
> > convenient configurations, or the scale is too small or too short.
>
> Well, that has always been and IMO still is for the wg to decide when the
> evidence through experiments is good enough to proceed with publishing the
> doc. E.g., when RFC 5562 was considered for publication by the wg, the wg
> requested for more experiments before making the decision and Sally
> provided what was requested. And it was just about adding ECN capability
> to SYNACK but here we are discussing much more significant changes.
>
> And more importantly, here we are mainly discussing whether there is any
> experimental evidence at all; for example, changing MD (from 0.5 to 0.7 or
> to whatever) must be considered separately when in slow start and when in
> congestion avoidance as I have pointed out. This requires tests that
> specifically target at studying the slow start overshoot phase and the
> following recovery phase. AFAIK, there are no such experimental results,
> no analysis nor evaluation presented to support the change when
> the sender is in slow start. So, we are not discussing the level of
> evidence but whether it exists at all, which is very different.
>
> > The other thing is that if we set a high bar for a PS doc related to CC,
> we should apply
> > the same bar to all related docs.For example, I am personally wondering
> how much we have
> > analyzed and performed experiments when we published RFC9002. (BTW, my
> intention is NOT to
> > accuse a specific doc, there should be some more examples.)
>
> Yes, IMO the bar should be the same. RFC 9002 was first targetted as
> experimental but was later changed to PS. I did not have time to follow
> the progress much at all but in my opinion it would probably have been
> better to publish it also as experimental first. On the other hand, it
> intends to implement NewReno as close as possible, and therefore not to
> change the congestion control principles that we have. I cannot tell how
> much experimental effort was put to verify the behaviour of those features
> that are new in QUIC and/or differ from stds track TCP but hopefully
> good enough. (Though there seems to be one issue related to resetting the
> exponential backoff of PTO, which seems to require a more accurate
> specification to ensure unambiguous interpretation and correct outcome,
> but that's an off topic here).
>
> > I am not sure how we can maintain this kind of fairness between the docs
> at the moment.
>
> I think it is the responsibility for the entire tsv area to do it, and
> eventually the TSV ADs and IESG. The process to follow is clear and
> robust, and when followed it has shown to be effectual in producing good
> quality documents. AFAIK, we have not decided to change it, maybe just
> many of us are unaware of it or have forgotten its existence?
> Traditionally tsv area has been very careful in considering new CC
> proposals because congestion control is so essential for the stability
> and fairness of the Internet and I hope we are not sliping too much from
> there. AFAIK, that's one reason why the process was documented.
>
> Thanks,
>
> /Markku
>
> > Thanks,
> > --
> > Yoshi
> >
> >
> > On Thu, Feb 17, 2022 at 6:41 AM Markku Kojo <kojo@cs.helsinki.fi> wrote:
> >       Hi Yoshi,
> >
> >       On Tue, 15 Feb 2022, Yoshifumi Nishida wrote:
> >
> >       > Hi Markku,
> >       >
> >       > Thanks for the comments. I think these are very valid points.
> >       > However, I would like to check several things as a co-chair and
> a doc
> >       shepherd before we
> >       > discuss the points you've raised.
> >       >
> >       > In my understanding (please correct me if I'm wrong), when this
> draft was
> >       adopted as an WG
> >       > item, I think the goal of the doc was some minor updates from
> RFC8312 which
> >       include more
> >       > clarifications, minor changes and bug fixes.
> >       > However, if we try to address your concerns, I think we'll need
> to invent a
> >       new version of
> >       > CUBIC something like CUBIC++ or NewCUBIC in the end.
> >       > I won't deny the value of such doc, but, this seems not to be
> what we agreed
> >       on
> >       > beforehand.
> >       > if we proceed in this direction, I think we will need to check
> the WG
> >       consensus whether
> >       > this should be a new goal for the doc.
> >       >
> >       > So, I would like to check if this is what you intend for the doc
> or you
> >       think we can
> >       > address your points while aligning with the original goal.
> >       > Also, if someone has opinions on this, please share.
> >
> >       I think it is important that we remember the status of RFC 8312
> and the
> >       decades long process that has been followed in tsv area for new
> >       TCP congestion control algorithms that have been proposed and
> submitted
> >       to IETF. In order to ensure that new cc algos are safe and fair,
> the
> >       process that has been followed for all current stds track TCP cc
> algos
> >       has required that the cc algo is first accepted and published as
> >       experimental RFC and only once enough supportive experimental
> evidence
> >       has been gathered the doc has become a candidate to be forwaded to
> stds
> >       track. We have even agreed on a relatively strict evaluation
> process to
> >       follow when cc algos are brought to the IETF to be published as
> >       experimental:
> >
> >
> https://www.ietf.org/about/groups/iesg/statements/experimental-congestion-control/
> >
> >       RFC 8312 was published as "Informational" and if I recall
> correctly the
> >       idea was "just to publish what's out there" for the benefit of the
> >       community. RFC 8312 was never really evaluated, particularly not
> in the
> >       way new cc algos are supposed to be as per the agreed process.
> >
> >       I do not recall what/how exactly was agreed when rfc8312bis was
> launched
> >       but I would be very interested to hear the justification why this
> doc
> >       does not need to follow the process mentioned above but we would
> like to
> >       propose IETF to publish a non-evaluated Informational doc to be
> published
> >       "with minor updates", i.e., without actual evaluation, as a stds
> track
> >       RFC? If the target really remains as PS then the bar should be even
> >       higher than what is described for experimental docs in the above
> process
> >       document, i.e, what we have followod for experimental to be moved
> to stds
> >       track.
> >
> >       The only justification that I have heard has beed "because CUBIC
> has long
> >       and wide deployment experience" and "the Internet has not smelted
> or that
> >       "we should have noticed if there were problems". We must, however,
> >       understand that in order to have noticeable bad impact CUBIC
> should cause
> >       some sort of congestion collapse. Congestion collapse, however, is
> not an
> >       issue with CUBIC nor with any other CC algo that applies an RTO
> mechanisms
> >       together with correctly implemented Karn's algo that retains the
> >       backed-off RTO until an Ack is received for a new (not rexmitted)
> data
> >       packet. The issue is fairness to competing traffic. This cannot be
> >       observed by deploying and measuring the performance and behaviour
> of CUBIC
> >       alone. CUBIC being more aggressive than current stds track TCP CC
> would
> >       just gives good performance results that one running CUBIC would
> be happy
> >       with. One must evaluate CUBIC's impact on the competing (Reno CC)
> traffic
> >       in range of environments which requires carefully designed active
> >       measurements with thoroughly-analyzed results (as required by the
> above
> >       process document, RFC 5033 and RFC 2914). What we seem to be
> missing is
> >       this evidence on CUBIC's impact and that is something the IETF
> must focus
> >       on, not just that whether CUBIC can achieve better performance
> than other
> >       existing CCs. The latter has been shown in many publications and
> is the
> >       majos focus in  many scientific papers proposing new algos.
> >       I appreciate a lot that CUBIC has been implemented/developped and
> >       deployed for long and I wonder whether those deploying CUBIC have
> >       unpublished results the wg could review before taking the decicion?
> >
> >       I suggest everyone to read carefully RFC 2914 Sec 3.2 and
> particularly
> >       what it says about more aggressive (than RFC 5681) congestion
> control
> >       algorithms:
> >
> >         Some of these may fail to implement
> >         the TCP congestion avoidance mechanisms correctly because of poor
> >         implementation [RFC2525].  Others may deliberately be implemented
> >         with congestion avoidance algorithms that are more aggressive in
> >         their use of bandwidth than other TCP implementations; this would
> >         allow a vendor to claim to have a "faster TCP".  The logical
> >         consequence of such implementations would be a spiral of
> increasingly
> >         aggressive TCP implementations, or increasingly aggressive
> transport
> >         protocols, leading back to the point where there is effectively
> no
> >         congestion avoidance and the Internet is chronically congested.
> >
> >       And:
> >
> >         It is convenient to divide flows into three classes: (1) TCP-
> >         compatible flows, (2) unresponsive flows, i.e., flows that do not
> >         slow down when congestion occurs, and (3) flows that are
> responsive
> >         but are not TCP-compatible.  The last two classes contain more
> >         aggressive flows that pose significant threats to Internet
> >         performance,
> >
> >       As I have tried to point out there are several features with CUBIC
> where
> >       it is likely to be (or to me it seems it obviously is) more
> aggressive
> >       than what is reguired to be TCP-compatible. I'm not aware of
> evidince
> >       presented to tcpm (or IETF/IRTF) which shows opposite (and I happy
> to be
> >       educated what I have missed).
> >
> >       You may take my comments to be a part of the expert review phase
> >       performed by the IRTF/ICCRG for CUBIC. I'm not requesting to
> modify this
> >       doc to CUBIC++ (or something) but it seems to be that this would be
> >       necessary if this doc intends to become published as PS. For
> experimental,
> >       I think it would need some addtioinal updates and record the areas
> >       uncertainty and where more experimentation (clearly) is required.
> >
> >       Thanks,
> >
> >       /Markku
> >
> >       > Thanks,
> >       > --
> >       > Yoshi
> >       >
> >       >
> >       > On Fri, Feb 11, 2022 at 9:34 PM Markku Kojo <kojo@cs.helsinki.fi>
> wrote:
> >       >       Hi Yoshi, all,
> >       >
> >       >       It seems to me that many issues that I raised have been
> solved,
> >       thanks.
> >       >       However, there are still a number of important issues that
> have not
> >       been
> >       >       addressed adequately. At least the following:
> >       >
> >       >       #135 on W_max. Yoshi's observation was correct that this
> is not
> >       resolved:
> >       >       the co-authors and original developpers of CUBIC
> (@lisongxu and
> >       >       @sangtaeha) agreed in their last message that Wmax needs
> different
> >       >       treatment for slow start and congestion avoidance and plan
> >       comprehensive
> >       >       (new) evaluation of it. This is obviously an open issue
> but the issue
> >       >       was closed?
> >       >
> >       >       #85 (& #86 with basically same issue and these two were
> combined) This
> >       >       (#85) is about ECN but the major issue is on using the
> same MD
> >       >       factor in slow start and in congestion avoidance when using
> >       >       loss-based CC. This (#85) remained closed even though I
> provided a
> >       >       thorough explanation why it is wrong and against the
> original theory
> >       and
> >       >       design by Van Jacobson, against the congestion control
> principles (RFC
> >       >       2914) and two co-authors agreed on this in their same last
> message to
> >       >       #135 when they agreed on Wmax needing rework. This is an
> important
> >       issue
> >       >       that the wg should consider very carefully because it is
> not only
> >       >       updating RFC 5681 but also in conflict with RFC 2914. How
> can
> >       >       tcpm (and IETF) suggest and allow one CC algo to not
> follow congestion
> >       >       control principles as set in RFC 2914 while requiring all
> other CCs to
> >       >       follow RFC 2914 guidelines?
> >       >       The current draft does not provide any justification for
> using the
> >       same
> >       >       MD factor in slow start as in congestion avoidance. Nor am
> I
> >       >       awere of any experimental data that would support this
> change.
> >       >       The fact thet CUBIC has been long deployed does not alone
> provide any
> >       >       supporting evidence because CUBIC is likely to give good
> performance
> >       as
> >       >       it is overagressive and thereby unfair to competing
> traffic and users
> >       >       tend to be happy when measuring the performance of the
> sending CUBIC
> >       >       only, not the competing traffic that is badly impacted.
> HyStart++ is
> >       >       suggested as mitigation to the problem but it cannot;
> HyStart++ is
> >       only
> >       >       applicable during initial slow start, not during slow
> start after RTO!
> >       >       That is, the "SHOULD use HYStart++" text in Sec 4.10 is
> impossible
> >       >       to implement as I have pointed out in my comments earlier.
> Using a
> >       proper
> >       >       MD factor in slow start is even more important if loss is
> detectected
> >       >       during a RTO recovery because the sender is likely to face
> heavy
> >       >       congestion in such a case and it is very bad if the sender
> continues
> >       >       sending with overaggressive rate, stealing the capacity
> from and
> >       causing
> >       >       harm to coexisting flows. In addition, as I have
> explained, HyStart++
> >       >       does not remove the problem even for the initial slow
> start as it is
> >       not
> >       >       shown to work always. Instead, the results with the
> HyStart++ draft
> >       show
> >       >       that it reduces 50% of rexmits and only 36% RTOs, meaning
> that there
> >       is
> >       >       likely to be a notable percentage of cases when a sender
> is still
> >       >       in slow start when first loss is detected (i.e., HyStart++
> had no
> >       effect)
> >       >       and a significant number of cases where a CUBIC sender is
> >       >       overaggressive continuing with a 1-40% larger cwnd than
> what is the
> >       >       available capacity. Note also that any delay-based
> heuristics like
> >       >       HyStart++ are known to work poorly in various wireless
> environmens
> >       where
> >       >       link delay tends to vary a lot. We may come up with some
> other MD
> >       factor
> >       >       that 0.5 when in slow start and HyStart++ is in use, but
> that is
> >       >       experimental, if not research, and definitely not ready
> for stds
> >       track.
> >       >
> >       >       #114, #132, and #143 w.r.t flightsize vs. cwnd. The
> current text
> >       >             does not quite correcly reflect what stacks that use
> cwnd do.
> >       >             I'll comment in #143 separately.
> >       >
> >       >       #96 & #98: The text added does not address the problems
> raised which
> >       are
> >       >       also evidenced in the paper pointed by Bob in #96. Even
> though CUBIC
> >       has
> >       >       been modified a bit after the paper was published, it does
> not
> >       >       automatically mean that the problem has been shown
> resolved:
> >       experimental
> >       >       evidence is required but not provided. The fact that CUBIC
> does not
> >       >       change MD factor for fast convergence is the root of the
> problem
> >       >       evidenced in the paper and remains so in the algo
> specified in this
> >       >       draft. This is also a significant problem when competing
> with Reno CC
> >       >       because CUBIC behaves much more aggressive than Reno CC
> when there is
> >       >       sudden congestion and all competing flows must converge
> fast down to a
> >       >       small fraction of the current cwnd to be fair to each
> other. This
> >       again
> >       >       cannot be evidenced not to be a problem by long deployment
> experience
> >       >       unless experimental data that measures the impact on
> competing traffic
> >       is
> >       >       presented to back the claims. Adjusting just Wmax for fast
> convergence
> >       is
> >       >       not enough and is even likely to be ineffective because
> there tend to
> >       be
> >       >       several losses when sudden congestion is hit, and
> particularly if
> >       NewReno
> >       >       is in use the sender stays several RTTs in Fast Recovery
> being
> >       >       overaggressive and then possibly continues at the same
> rate in CA
> >       which
> >       >       is unlikely to reach evan close to Wmax before a new loss
> hits
> >       >       the sender again. That is, lower Wmax and lower additive
> increase
> >       factor
> >       >       do not compensate the use of larger MD factor when sudden
> congestion
> >       is
> >       >       encountered.
> >       >
> >       >       #93 & #94 & (#89) Sec 5.3 still does not address any
> difficult
> >       >       environments, in particular buffer-bloated paths (nor does
> Sec 5.4).
> >       >       We need evidence (results) that show CUBIC is fair towards
> other CCs
> >       >       (Reno) also in such environments. Note that CUBIC's
> decision to leave
> >       >       Reno-friendly region is based on the size of cwnd which
> tends to be
> >       >       incorrect with buffer-bloated bottlenecks because with
> huge buffers
> >       the
> >       >       cwnd can be many times larger than what is actually needed
> to fully
> >       >       utilize the available network bit-rate. Therefore, Reno CC
> has no
> >       problem
> >       >       in fully utilizing such bottleneck links and CUBIC must
> stay in
> >       >       Reno-friendly region longer but it leaves it too early
> because the
> >       same C
> >       >       is used as with non-bloated environments. We lack
> experiments showing
> >       >       CUBIC follows the congestion control principles and is
> fair to current
> >       >       standard TCP CC; to my understanding no experiments with
> >       buffer-bloated
> >       >       bottlenecks are cited to back up the claims even though
> buffer bloat
> >       is
> >       >       very well known to be a common (difficult) environment in
> today's
> >       >       Internet.
> >       >
> >       >       #90 The current text on applying undo (a response to
> detected false
> >       >       fastrexmit) does not provide correct result if someone
> implements it.
> >       >       I have explained the problems there in github but seem to
> have not
> >       >       replied to latest comments by Neal. I'll reply and try to
> explain
> >       more.
> >       >       Again one major problem here is that the draft suggest a
> new response
> >       >       algo for false fast rexmits but does not provide any
> experimental data
> >       to
> >       >       support it. Long deployment experience has been suggested
> as
> >       >       justification but again without any carefully evaluated
> experimental
> >       >       data and evidence there is no meat. The issue is important
> to solve
> >       but is
> >       >       not specific to CUBIC. Instead, it is general problem for
> all TCP CC
> >       >       variants. IMHO, this is not ready for standards track but
> deserves a
> >       draft
> >       >       of its own so that it can be carefully evaluated and
> discussed ion the
> >       >       list. AFAIK there has been no discussion on this on the
> tcpm list, so
> >       >       those probably interested and having experience are likely
> to be
> >       unaware
> >       >       that this is part of CUBIC draft.
> >       >
> >       >       #88 The problem with correctness of the AIMD model and
> setting alpha
> >       for
> >       >       CUBIC requires further consideration. Bob provided an
> analysis that
> >       >       leaves things still open. It seems that I never had time
> to review and
> >       >       comment the analysis and clarify why the model does not
> work. I'll do
> >       that
> >       >       separately as it is important to ensure CUBIC behaves
> fairly as
> >       intended
> >       >       for the Reno-friendly region.
> >       >
> >       >       Best regards,
> >       >
> >       >       /Markku
> >       >
> >       >       On Mon, 31 Jan 2022, Yoshifumi Nishida wrote:
> >       >
> >       >       > Hello,
> >       >       >
> >       >       > After some discussions among chairs, we decided to run
> the 2nd WGLC
> >       on
> >       >       draft-ietf-tcpm-rfc8312bis in
> >       >       > consideration of the importance of the draft.
> >       >       > We'll be grateful if you could send your feedback to the
> ML. The
> >       WGLC runs
> >       >       until *Feb 11*.
> >       >       >
> >       >       > If interested, you can check in-depth past discussions
> in the
> >       following URL.
> >       >       > https://github.com/NTAP/rfc8312bis/
> >       >       >
> >       >       > Thank you so much!
> >       >       > --
> >       >       > tcpm co-chairs
> >       >       >
> >       >       >
> >       >       > On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <
> lars@eggert.org> wrote:
> >       >       >       Hi,
> >       >       >
> >       >       >       this -06 version rolls in all the changes
> requested during
> >       (and after)
> >       >       WGLC ended.
> >       >       >
> >       >       >       I'll leave it up to the chairs to decide if
> another WGLC is
> >       warranted
> >       >       or the document can
> >       >       >       progress as-is.
> >       >       >
> >       >       >       Thanks,
> >       >       >       Lars
> >       >       >
> >       >       >
> >       >       >       > On 2022-1-26, at 11:12, 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 TCP Maintenance
> and Minor
> >       >       Extensions WG of the IETF.
> >       >       >       >
> >       >       >       >        Title           : CUBIC for Fast and
> Long-Distance
> >       Networks
> >       >       >       >        Authors         : Lisong Xu
> >       >       >       >                          Sangtae Ha
> >       >       >       >                          Injong Rhee
> >       >       >       >                          Vidhi Goel
> >       >       >       >                          Lars Eggert
> >       >       >       >       Filename        :
> draft-ietf-tcpm-rfc8312bis-06.txt
> >       >       >       >       Pages           : 35
> >       >       >       >       Date            : 2022-01-26
> >       >       >       >
> >       >       >       > Abstract:
> >       >       >       >   CUBIC is a standard TCP congestion control
> algorithm that
> >       uses a
> >       >       >       >   cubic function instead of a linear congestion
> window
> >       increase
> >       >       >       >   function to improve scalability and stability
> over fast
> >       and long-
> >       >       >       >   distance networks.  CUBIC has been adopted as
> the default
> >       TCP
> >       >       >       >   congestion control algorithm by the Linux,
> Windows, and
> >       Apple
> >       >       stacks.
> >       >       >       >
> >       >       >       >   This document updates the specification of
> CUBIC to
> >       include
> >       >       >       >   algorithmic improvements based on these
> implementations
> >       and recent
> >       >       >       >   academic work.  Based on the extensive
> deployment
> >       experience with
> >       >       >       >   CUBIC, it also moves the specification to the
> Standards
> >       Track,
> >       >       >       >   obsoleting RFC 8312.  This also requires
> updating RFC
> >       5681, to
> >       >       allow
> >       >       >       >   for CUBIC's occasionally more aggressive
> sending behavior.
> >       >       >       >
> >       >       >       >
> >       >       >       > The IETF datatracker status page for this draft
> is:
> >       >       >       >
> https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/
> >       >       >       >
> >       >       >       > There is also an HTML version available at:
> >       >       >       >
> >       https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html
> >       >       >       >
> >       >       >       > A diff from the previous version is available at:
> >       >       >       >
> >       https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rfc8312bis-06
> >       >       >       >
> >       >       >       >
> >       >       >       > Internet-Drafts are also available by rsync at
> >       >       rsync.ietf.org::internet-drafts
> >       >       >       >
> >       >       >       >
> >       >       >       > _______________________________________________
> >       >       >       > tcpm mailing list
> >       >       >       > tcpm@ietf.org
> >       >       >       > https://www.ietf.org/mailman/listinfo/tcpm
> >       >       >
> >       >       >       _______________________________________________
> >       >       >       tcpm mailing list
> >       >       >       tcpm@ietf.org
> >       >       >       https://www.ietf.org/mailman/listinfo/tcpm
> >       >       >
> >       >       >
> >       >       >
> >       >
> >       >
> >       >
> >
> >
> >

--0000000000005fadfc05d884e636
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi=C2=A0Markku,</div><div><br></div><div>I&#39;ve tho=
ught about this over the weekend. Changing the intended status of the doc m=
ay look simple, but it might affect other=C2=A0docs as a precedent.=C2=A0</=
div><div>So, I would like to think about it a bit more.</div><div><br></div=
><div>I think it may be possible to do some experiments before we=C2=A0publ=
ish the doc as a PS if it&#39;s not too complicated.</div><div>But, in this=
 case, I think we&#39;ll need to delineate=C2=A0what we want to evaluate.=
=C2=A0<br></div><div>One thing I am wondering is whether points are=C2=A0if=
 there is a risk for congestion collapse, or if this is too aggressive than=
 NewReno or something else.</div><div><br></div><div>If we think there&#39;=
s a risk for congestion collapse, we should be able to describe which scena=
rios could lead such situations.</div><div>If we think CUBIC is too aggress=
ive, I think we will need to define some bars for it.</div><div>I think 900=
2 is more aggressive than TCP NewReno in some points.=C2=A0</div><div>So, i=
f we reject CUBIC to be a PS due to its aggressiveness, I think there shoul=
d be a border for it.</div><div><br></div><div>Thanks,</div><div>--<br></di=
v><div>Yoshi</div><div><br></div><div class=3D"gmail_quote"><div dir=3D"ltr=
" class=3D"gmail_attr">On Fri, Feb 18, 2022 at 2:23 AM Markku Kojo &lt;<a h=
ref=3D"mailto:kojo@cs.helsinki.fi">kojo@cs.helsinki.fi</a>&gt; wrote:<br></=
div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;bor=
der-left:1px solid rgb(204,204,204);padding-left:1ex">Hi Yoshi,<br>
<br>
On Fri, 18 Feb 2022, Yoshifumi Nishida wrote:<br>
<br>
&gt; Hi Markku,<br>
&gt; <br>
&gt; Thanks for the comments. Yes, these are valid and important concerns. =
However, I have<br>
&gt; mixed feelings about them.<br>
&gt; I don&#39;t want to open a can=C2=A0of worms, but here&#39;s are some =
of my personally thoughts...<br>
&gt; <br>
&gt; One thing is that &quot;long and wide deployment experience&quot; migh=
t not be a very strong claim,<br>
&gt; but then, I am wondering what would be the strong one.<br>
&gt; One may bring some solid experiment results, but the others might argu=
e it uses too<br>
&gt; convenient configurations, or the scale is too small or too short.<br>
<br>
Well, that has always been and IMO still is for the wg to decide when the <=
br>
evidence through experiments is good enough to proceed with publishing the =
<br>
doc. E.g., when RFC 5562 was considered for publication by the wg, the wg <=
br>
requested for more experiments before making the decision and Sally <br>
provided what was requested. And it was just about adding ECN capability <b=
r>
to SYNACK but here we are discussing much more significant changes.<br>
<br>
And more importantly, here we are mainly discussing whether there is any <b=
r>
experimental evidence at all; for example, changing MD (from 0.5 to 0.7 or =
<br>
to whatever) must be considered separately when in slow start and when in <=
br>
congestion avoidance as I have pointed out. This requires tests that <br>
specifically target at studying the slow start overshoot phase and the <br>
following recovery phase. AFAIK, there are no such experimental results, <b=
r>
no analysis nor evaluation presented to support the change when <br>
the sender is in slow start. So, we are not discussing the level of <br>
evidence but whether it exists at all, which is very different.<br>
<br>
&gt; The other thing is that if we set a high bar for a PS doc related to C=
C, we should apply<br>
&gt; the same bar to all related docs.For example, I am personally wonderin=
g how much we have<br>
&gt; analyzed and performed experiments when we published RFC9002. (BTW, my=
 intention is NOT to<br>
&gt; accuse a specific doc, there should be some more examples.)=C2=A0<br>
<br>
Yes, IMO the bar should be the same. RFC 9002 was first targetted as <br>
experimental but was later changed to PS. I did not have time to follow <br=
>
the progress much at all but in my opinion it would probably have been <br>
better to publish it also as experimental first. On the other hand, it <br>
intends to implement NewReno as close as possible, and therefore not to <br=
>
change the congestion control principles that we have. I cannot tell how <b=
r>
much experimental effort was put to verify the behaviour of those features =
<br>
that are new in QUIC and/or differ from stds track TCP but hopefully <br>
good enough. (Though there seems to be one issue related to resetting the <=
br>
exponential backoff of PTO, which seems to require a more accurate <br>
specification to ensure unambiguous interpretation and correct outcome, <br=
>
but that&#39;s an off topic here).<br>
<br>
&gt; I am not sure how we can maintain this kind of fairness between the do=
cs at the moment.<br>
<br>
I think it is the responsibility for the entire tsv area to do it, and <br>
eventually the TSV ADs and IESG. The process to follow is clear and <br>
robust, and when followed it has shown to be effectual in producing good <b=
r>
quality documents. AFAIK, we have not decided to change it, maybe just <br>
many of us are unaware of it or have forgotten its existence?<br>
Traditionally tsv area has been very careful in considering new CC <br>
proposals because congestion control is so essential for the stability <br>
and fairness of the Internet and I hope we are not sliping too much from <b=
r>
there. AFAIK, that&#39;s one reason why the process was documented.<br>
<br>
Thanks,<br>
<br>
/Markku<br>
<br>
&gt; Thanks,<br>
&gt; --<br>
&gt; Yoshi=C2=A0<br>
&gt; <br>
&gt; <br>
&gt; On Thu, Feb 17, 2022 at 6:41 AM Markku Kojo &lt;<a href=3D"mailto:kojo=
@cs.helsinki.fi" target=3D"_blank">kojo@cs.helsinki.fi</a>&gt; wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Hi Yoshi,<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0On Tue, 15 Feb 2022, Yoshifumi Nishida wrote=
:<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Hi=C2=A0Markku,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Thanks for the comments. I think these =
are very valid points.=C2=A0<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; However, I would like to check several =
things as a co-chair and a doc<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0shepherd before we<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; discuss=C2=A0the points you&#39;ve=C2=
=A0raised.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; In my understanding (please correct me =
if I&#39;m wrong), when this draft was<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0adopted as an WG<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; item, I think the goal of the doc was s=
ome minor updates from RFC8312 which<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0include more<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; clarifications, minor=C2=A0changes and =
bug fixes.=C2=A0<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; However, if we try to address your=C2=
=A0concerns, I think we&#39;ll need to invent a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0new version of<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; CUBIC something like CUBIC++ or NewCUBI=
C in the end.=C2=A0<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; I won&#39;t deny the value of such doc,=
 but, this seems not to be what we agreed<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0on<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; beforehand.=C2=A0=C2=A0<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; if we proceed in this direction, I thin=
k we will need to check the WG<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0consensus whether<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; this should be a new goal for the doc.<=
br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; So, I would like to check if this is wh=
at you intend for the doc or you<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0think we can<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; address your points while aligning with=
 the original goal.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Also, if someone has opinions on this, =
please share.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0I think it is important that we remember the=
 status of RFC 8312 and the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0decades long process that has been followed =
in tsv area for new<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0TCP congestion control algorithms that have =
been proposed and submitted<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0to IETF. In order to ensure that new cc algo=
s are safe and fair, the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0process that has been followed for all curre=
nt stds track TCP cc algos<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0has required that the cc algo is first accep=
ted and published as<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0experimental RFC and only once enough suppor=
tive experimental evidence<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0has been gathered the doc has become a candi=
date to be forwaded to stds<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0track. We have even agreed on a relatively s=
trict evaluation process to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0follow when cc algos are brought to the IETF=
 to be published as<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0experimental:<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"https://www.ietf.org/about/groups=
/iesg/statements/experimental-congestion-control/" rel=3D"noreferrer" targe=
t=3D"_blank">https://www.ietf.org/about/groups/iesg/statements/experimental=
-congestion-control/</a><br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0RFC 8312 was published as &quot;Informationa=
l&quot; and if I recall correctly the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0idea was &quot;just to publish what&#39;s ou=
t there&quot; for the benefit of the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0community. RFC 8312 was never really evaluat=
ed, particularly not in the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0way new cc algos are supposed to be as per t=
he agreed process.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0I do not recall what/how exactly was agreed =
when rfc8312bis was launched<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0but I would be very interested to hear the j=
ustification why this doc<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0does not need to follow the process mentione=
d above but we would like to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0propose IETF to publish a non-evaluated Info=
rmational doc to be published<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&quot;with minor updates&quot;, i.e., withou=
t actual evaluation, as a stds track<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0RFC? If the target really remains as PS then=
 the bar should be even<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0higher than what is described for experiment=
al docs in the above process<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0document, i.e, what we have followod for exp=
erimental to be moved to stds<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0track.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0The only justification that I have heard has=
 beed &quot;because CUBIC has long<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0and wide deployment experience&quot; and &qu=
ot;the Internet has not smelted or that<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&quot;we should have noticed if there were p=
roblems&quot;. We must, however,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0understand that in order to have noticeable =
bad impact CUBIC should cause<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0some sort of congestion collapse. Congestion=
 collapse, however, is not an<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0issue with CUBIC nor with any other CC algo =
that applies an RTO mechanisms<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0together with correctly implemented Karn&#39=
;s algo that retains the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0backed-off RTO until an Ack is received for =
a new (not rexmitted) data<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0packet. The issue is fairness to competing t=
raffic. This cannot be<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0observed by deploying and measuring the perf=
ormance and behaviour of CUBIC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0alone. CUBIC being more aggressive than curr=
ent stds track TCP CC would<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0just gives good performance results that one=
 running CUBIC would be happy<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0with. One must evaluate CUBIC&#39;s impact o=
n the competing (Reno CC) traffic<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0in range of environments which requires care=
fully designed active<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0measurements with thoroughly-analyzed result=
s (as required by the above<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0process document, RFC 5033 and RFC 2914). Wh=
at we seem to be missing is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0this evidence on CUBIC&#39;s impact and that=
 is something the IETF must focus<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0on, not just that whether CUBIC can achieve =
better performance than other<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0existing CCs. The latter has been shown in m=
any publications and is the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0majos focus in=C2=A0 many scientific papers =
proposing new algos.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0I appreciate a lot that CUBIC has been imple=
mented/developped and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0deployed for long and I wonder whether those=
 deploying CUBIC have<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0unpublished results the wg could review befo=
re taking the decicion?<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0I suggest everyone to read carefully RFC 291=
4 Sec 3.2 and particularly<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0what it says about more aggressive (than RFC=
 5681) congestion control<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0algorithms:<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 Some of these may fail to implement<b=
r>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 the TCP congestion avoidance mechanis=
ms correctly because of poor<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 implementation [RFC2525].=C2=A0 Other=
s may deliberately be implemented<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 with congestion avoidance algorithms =
that are more aggressive in<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 their use of bandwidth than other TCP=
 implementations; this would<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 allow a vendor to claim to have a &qu=
ot;faster TCP&quot;.=C2=A0 The logical<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 consequence of such implementations w=
ould be a spiral of increasingly<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 aggressive TCP implementations, or in=
creasingly aggressive transport<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 protocols, leading back to the point =
where there is effectively no<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 congestion avoidance and the Internet=
 is chronically congested.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0And:<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 It is convenient to divide flows into=
 three classes: (1) TCP-<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 compatible flows, (2) unresponsive fl=
ows, i.e., flows that do not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 slow down when congestion occurs, and=
 (3) flows that are responsive<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 but are not TCP-compatible.=C2=A0 The=
 last two classes contain more<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 aggressive flows that pose significan=
t threats to Internet<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 performance,<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0As I have tried to point out there are sever=
al features with CUBIC where<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0it is likely to be (or to me it seems it obv=
iously is) more aggressive<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0than what is reguired to be TCP-compatible. =
I&#39;m not aware of evidince<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0presented to tcpm (or IETF/IRTF) which shows=
 opposite (and I happy to be<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0educated what I have missed).<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0You may take my comments to be a part of the=
 expert review phase<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0performed by the IRTF/ICCRG for CUBIC. I&#39=
;m not requesting to modify this<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0doc to CUBIC++ (or something) but it seems t=
o be that this would be<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0necessary if this doc intends to become publ=
ished as PS. For experimental,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0I think it would need some addtioinal update=
s and record the areas<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0uncertainty and where more experimentation (=
clearly) is required.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Thanks,<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0/Markku<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Thanks,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; --<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Yoshi<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; On Fri, Feb 11, 2022 at 9:34 PM Markku =
Kojo &lt;<a href=3D"mailto:kojo@cs.helsinki.fi" target=3D"_blank">kojo@cs.h=
elsinki.fi</a>&gt; wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Hi Yoshi, all=
,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0It seems to m=
e that many issues that I raised have been solved,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0thanks.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0However, ther=
e are still a number of important issues that have not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0been<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0addressed ade=
quately. At least the following:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#135 on W_max=
. Yoshi&#39;s observation was correct that this is not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0resolved:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0the co-author=
s and original developpers of CUBIC (@lisongxu and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0@sangtaeha) a=
greed in their last message that Wmax needs different<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0treatment for=
 slow start and congestion avoidance and plan<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0comprehensive<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0(new) evaluat=
ion of it. This is obviously an open issue but the issue<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0was closed?<b=
r>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#85 (&amp; #8=
6 with basically same issue and these two were combined) This<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0(#85) is abou=
t ECN but the major issue is on using the same MD<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0factor in slo=
w start and in congestion avoidance when using<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0loss-based CC=
. This (#85) remained closed even though I provided a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0thorough expl=
anation why it is wrong and against the original theory<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0design by Van=
 Jacobson, against the congestion control principles (RFC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A02914) and two=
 co-authors agreed on this in their same last message to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#135 when the=
y agreed on Wmax needing rework. This is an important<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0issue<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0that the wg s=
hould consider very carefully because it is not only<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0updating RFC =
5681 but also in conflict with RFC 2914. How can<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0tcpm (and IET=
F) suggest and allow one CC algo to not follow congestion<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0control princ=
iples as set in RFC 2914 while requiring all other CCs to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0follow RFC 29=
14 guidelines?<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0The current d=
raft does not provide any justification for using the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0same<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0MD factor in =
slow start as in congestion avoidance. Nor am I<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0awere of any =
experimental data that would support this change.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0The fact thet=
 CUBIC has been long deployed does not alone provide any<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0supporting ev=
idence because CUBIC is likely to give good performance<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0as<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0it is overagr=
essive and thereby unfair to competing traffic and users<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0tend to be ha=
ppy when measuring the performance of the sending CUBIC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0only, not the=
 competing traffic that is badly impacted. HyStart++ is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0suggested as =
mitigation to the problem but it cannot; HyStart++ is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0only<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0applicable du=
ring initial slow start, not during slow start after RTO!<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0That is, the =
&quot;SHOULD use HYStart++&quot; text in Sec 4.10 is impossible<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0to implement =
as I have pointed out in my comments earlier. Using a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0proper<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0MD factor in =
slow start is even more important if loss is detectected<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0during a RTO =
recovery because the sender is likely to face heavy<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0congestion in=
 such a case and it is very bad if the sender continues<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0sending with =
overaggressive rate, stealing the capacity from and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0causing<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0harm to coexi=
sting flows. In addition, as I have explained, HyStart++<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0does not remo=
ve the problem even for the initial slow start as it is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0shown to work=
 always. Instead, the results with the HyStart++ draft<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0show<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0that it reduc=
es 50% of rexmits and only 36% RTOs, meaning that there<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0likely to be =
a notable percentage of cases when a sender is still<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0in slow start=
 when first loss is detected (i.e., HyStart++ had no<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0effect)<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0and a signifi=
cant number of cases where a CUBIC sender is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0overaggressiv=
e continuing with a 1-40% larger cwnd than what is the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0available cap=
acity. Note also that any delay-based heuristics like<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0HyStart++ are=
 known to work poorly in various wireless environmens<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0where<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0link delay te=
nds to vary a lot. We may come up with some other MD<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0factor<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0that 0.5 when=
 in slow start and HyStart++ is in use, but that is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0experimental,=
 if not research, and definitely not ready for stds<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0track.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#114, #132, a=
nd #143 w.r.t flightsize vs. cwnd. The current text<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 =C2=A0=
 =C2=A0 does not quite correcly reflect what stacks that use cwnd do.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0 =C2=A0=
 =C2=A0 I&#39;ll comment in #143 separately.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#96 &amp; #98=
: The text added does not address the problems raised which<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0are<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0also evidence=
d in the paper pointed by Bob in #96. Even though CUBIC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0has<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0been modified=
 a bit after the paper was published, it does not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0automatically=
 mean that the problem has been shown resolved:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0experimental<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0evidence is r=
equired but not provided. The fact that CUBIC does not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0change MD fac=
tor for fast convergence is the root of the problem<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0evidenced in =
the paper and remains so in the algo specified in this<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0draft. This i=
s also a significant problem when competing with Reno CC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0because CUBIC=
 behaves much more aggressive than Reno CC when there is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0sudden conges=
tion and all competing flows must converge fast down to a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0small fractio=
n of the current cwnd to be fair to each other. This<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0again<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0cannot be evi=
denced not to be a problem by long deployment experience<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0unless experi=
mental data that measures the impact on competing traffic<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0presented to =
back the claims. Adjusting just Wmax for fast convergence<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0not enough an=
d is even likely to be ineffective because there tend to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0be<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0several losse=
s when sudden congestion is hit, and particularly if<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0NewReno<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is in use the=
 sender stays several RTTs in Fast Recovery being<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0overaggressiv=
e and then possibly continues at the same rate in CA<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0which<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is unlikely t=
o reach evan close to Wmax before a new loss hits<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0the sender ag=
ain. That is, lower Wmax and lower additive increase<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0factor<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0do not compen=
sate the use of larger MD factor when sudden congestion<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0encountered.<=
br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#93 &amp; #94=
 &amp; (#89) Sec 5.3 still does not address any difficult<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0environments,=
 in particular buffer-bloated paths (nor does Sec 5.4).<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0We need evide=
nce (results) that show CUBIC is fair towards other CCs<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0(Reno) also i=
n such environments. Note that CUBIC&#39;s decision to leave<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Reno-friendly=
 region is based on the size of cwnd which tends to be<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0incorrect wit=
h buffer-bloated bottlenecks because with huge buffers<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0cwnd can be m=
any times larger than what is actually needed to fully<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0utilize the a=
vailable network bit-rate. Therefore, Reno CC has no<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0problem<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0in fully util=
izing such bottleneck links and CUBIC must stay in<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Reno-friendly=
 region longer but it leaves it too early because the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0same C<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is used as wi=
th non-bloated environments. We lack experiments showing<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0CUBIC follows=
 the congestion control principles and is fair to current<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0standard TCP =
CC; to my understanding no experiments with<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0buffer-bloated<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0bottlenecks a=
re cited to back up the claims even though buffer bloat<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0very well kno=
wn to be a common (difficult) environment in today&#39;s<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Internet.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#90 The curre=
nt text on applying undo (a response to detected false<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0fastrexmit) d=
oes not provide correct result if someone implements it.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0I have explai=
ned the problems there in github but seem to have not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0replied to la=
test comments by Neal. I&#39;ll reply and try to explain<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0more.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Again one maj=
or problem here is that the draft suggest a new response<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0algo for fals=
e fast rexmits but does not provide any experimental data<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0support it. L=
ong deployment experience has been suggested as<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0justification=
 but again without any carefully evaluated experimental<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0data and evid=
ence there is no meat. The issue is important to solve<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0but is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0not specific =
to CUBIC. Instead, it is general problem for all TCP CC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0variants. IMH=
O, this is not ready for standards track but deserves a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0draft<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0of its own so=
 that it can be carefully evaluated and discussed ion the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0list. AFAIK t=
here has been no discussion on this on the tcpm list, so<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0those probabl=
y interested and having experience are likely to be<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0unaware<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0that this is =
part of CUBIC draft.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0#88 The probl=
em with correctness of the AIMD model and setting alpha<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0for<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0CUBIC require=
s further consideration. Bob provided an analysis that<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0leaves things=
 still open. It seems that I never had time to review and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0comment the a=
nalysis and clarify why the model does not work. I&#39;ll do<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0that<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0separately as=
 it is important to ensure CUBIC behaves fairly as<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0intended<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0for the Reno-=
friendly region.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Best regards,=
<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0/Markku<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0On Mon, 31 Ja=
n 2022, Yoshifumi Nishida wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Hello,<b=
r>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; After so=
me discussions among chairs, we decided to run the 2nd WGLC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0on<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0draft-ietf-tc=
pm-rfc8312bis in<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; consider=
ation of the importance of the draft.=C2=A0<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; We&#39;l=
l be grateful if you could send your feedback to the ML. The<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0WGLC runs<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0until *Feb 11=
*.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; If inter=
ested, you can check in-depth past discussions in the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0following URL.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; <a href=
=3D"https://github.com/NTAP/rfc8312bis/" rel=3D"noreferrer" target=3D"_blan=
k">https://github.com/NTAP/rfc8312bis/</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; Thank yo=
u so much!<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; --<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; tcpm co-=
chairs<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt; On Wed, =
Jan 26, 2022 at 2:50 AM Lars Eggert &lt;<a href=3D"mailto:lars@eggert.org" =
target=3D"_blank">lars@eggert.org</a>&gt; wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0Hi,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0this -06 version rolls in all the changes requested dur=
ing<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0(and after)<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0WGLC ended.<b=
r>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0I&#39;ll leave it up to the chairs to decide if another=
 WGLC is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0warranted<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0or the docume=
nt can<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0progress as-is.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0Thanks,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0Lars<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; On 2022-1-26, at 11:12, <a href=3D"mailto:internet=
-drafts@ietf.org" target=3D"_blank">internet-drafts@ietf.org</a> wrote:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; A New Internet-Draft is available from the on-line=
<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Internet-Drafts<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0directories.<=
br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; This draft is a work item of the TCP Maintenance a=
nd Minor<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Extensions WG=
 of the IETF.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Title=C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0: CUBIC for Fast and Long-Distance<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Networks<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Authors=C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0: Lisong Xu<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Sangtae Ha<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Injong Rhee<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Vidhi Goel<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Lars Eggert<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Filename=C2=A0 =C2=A0 =
=C2=A0 =C2=A0 : draft-ietf-tcpm-rfc8312bis-06.txt<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Pages=C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0: 35<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Date=C2=A0 =C2=A0 =C2=A0=
 =C2=A0 =C2=A0 =C2=A0 : 2022-01-26<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; Abstract:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0CUBIC is a standard TCP congestion con=
trol algorithm that<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0uses a<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0cubic function instead of a linear con=
gestion window<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0increase<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0function to improve scalability and st=
ability over fast<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0and long-<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0distance networks.=C2=A0 CUBIC has bee=
n adopted as the default<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0TCP<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0congestion control algorithm by the Li=
nux, Windows, and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Apple<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0stacks.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0This document updates the specificatio=
n of CUBIC to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0include<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0algorithmic improvements based on thes=
e implementations<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0and recent<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0academic work.=C2=A0 Based on the exte=
nsive deployment<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0experience with<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0CUBIC, it also moves the specification=
 to the Standards<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Track,<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0obsoleting RFC 8312.=C2=A0 This also r=
equires updating RFC<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A05681, to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0allow<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0for CUBIC&#39;s occasionally more aggr=
essive sending behavior.<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; The IETF datatracker status page for this draft is=
:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://datatracker.ietf.org/doc/draft-=
ietf-tcpm-rfc8312bis/" rel=3D"noreferrer" target=3D"_blank">https://datatra=
cker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; There is also an HTML version available at:<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"https://www.ietf.org/archive/id/d=
raft-ietf-tcpm-rfc8312bis-06.html" rel=3D"noreferrer" target=3D"_blank">htt=
ps://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; A diff from the previous version is available at:<=
br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"https://www.ietf.org/rfcdiff?url2=
=3Ddraft-ietf-tcpm-rfc8312bis-06" rel=3D"noreferrer" target=3D"_blank">http=
s://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312bis-06</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; Internet-Drafts are also available by rsync at<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0rsync.ietf.or=
g::internet-drafts<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; _______________________________________________<br=
>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; tcpm mailing list<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"mailto:tcpm@ietf.org" target=3D"_blank"=
>tcpm@ietf.org</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/t=
cpm" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/mailman/list=
info/tcpm</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0_______________________________________________<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0tcpm mailing list<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0<a href=3D"mailto:tcpm@ietf.org" target=3D"_blank">tcpm=
@ietf.org</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =
=C2=A0 =C2=A0 =C2=A0<a href=3D"https://www.ietf.org/mailman/listinfo/tcpm" =
rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/mailman/listinfo/=
tcpm</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0&gt;<br>
&gt; <br>
&gt; <br>
&gt;</blockquote></div></div>

--0000000000005fadfc05d884e636--


From nobody Mon Feb 21 06:52:39 2022
Return-Path: <lars@eggert.org>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D0DF23A1129; Mon, 21 Feb 2022 06:52:37 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.099
X-Spam-Level: 
X-Spam-Status: No, score=-2.099 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=eggert.org
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NPTFx_wGd5vD; Mon, 21 Feb 2022 06:52:33 -0800 (PST)
Received: from mail.eggert.org (mail.eggert.org [IPv6:2a00:ac00:4000:400:211:32ff:fe22:186f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3DA6F3A1121; Mon, 21 Feb 2022 06:52:33 -0800 (PST)
Received: from smtpclient.apple (unknown [IPv6:2a00:ac00:4000:400:897b:a2e4:e787:3438]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.eggert.org (Postfix) with ESMTPSA id B586D1D242D; Mon, 21 Feb 2022 16:52:14 +0200 (EET)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=eggert.org; s=dkim; t=1645455134; bh=hZj5QQ6bHv+zJ2jqYr/0tz45VkG6qlzI45CFjdM2K5I=; h=Subject:From:In-Reply-To:Date:Cc:References:To; b=i4F2OCy+s2OtppjZWtkedhqEV6EYyeusqIe2KHFj3frAgsa5rQ14bj4FGk8gzF/cI bFQH5FXzBDDPDyiK6lwVyadH2g6ceSPPm1rAtcYifelgx1ryj4VHcDMCdZVcC4uXi7 fTW7H71Eckqh+ODeY3/mctkeCIxO8L7cy/hplI3c=
Content-Type: multipart/signed; boundary="Apple-Mail=_42D0908D-B566-4B53-9416-3140786226F0"; protocol="application/pgp-signature"; micalg=pgp-sha512
Mime-Version: 1.0 (Mac OS X Mail 15.0 \(3693.60.0.1.1\))
From: Lars Eggert <lars@eggert.org>
In-Reply-To: <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com>
Date: Mon, 21 Feb 2022 16:52:13 +0200
Cc: Markku Kojo <kojo@cs.helsinki.fi>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
Message-Id: <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi> <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com>
To: Yoshifumi Nishida <nsd.ietf@gmail.com>
X-MailScanner-ID: B586D1D242D.A29CE
X-MailScanner: Found to be clean
X-MailScanner-From: lars@eggert.org
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/ZnDpCtxKWMhiv9mrMudpNGadcCo>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 21 Feb 2022 14:52:38 -0000

--Apple-Mail=_42D0908D-B566-4B53-9416-3140786226F0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

Hi,

first, as an editor (but not wearing other hats) I do object against =
reclassifying the document as Experimental or Informational, esp. this =
late in the process.

CUBIC is not a new CC proposal. It's been ubiquitously deployed on the =
Internet for many years. All major TCP stacks use it by default. Given =
that, I don't see how the IESG statement on "Experimental Specification =
of New Congestion Control Algorithms" applies here.

(As an aside, that statement was specifically written when CUBIC, =
Compound and Hamilton were all jostling for RFC publication. The =
community converged on CUBIC in the meantime.)

CUBIC *is* the default CC mechanism used on the Internet and in other IP =
networks. It should be published on the standards track.

Second, this is the second time that an individual raised a long list of =
issues after the WGLC has ended. We did try to work through all the =
issues that were raised after the first WGLC, but the individual often =
failed to participate in the discussion in a timely manner. I now see at =
least several of those same issues re-raised by the same individual. =
Before going down the same path a second time, I would actually ask the =
WG to confirm there is consensus that the raised issues are valid (and =
if they are valid, commit to contributing to a resolution in a timely =
manner.)

Thanks,
Lars


--Apple-Mail=_42D0908D-B566-4B53-9416-3140786226F0
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
	filename=signature.asc
Content-Type: application/pgp-signature;
	name=signature.asc
Content-Description: Message signed with OpenPGP

-----BEGIN PGP SIGNATURE-----
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=YRcD
-----END PGP SIGNATURE-----

--Apple-Mail=_42D0908D-B566-4B53-9416-3140786226F0--


From nobody Mon Feb 21 16:11:25 2022
Return-Path: <in@bobbriscoe.net>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8B5BF3A0DE6; Mon, 21 Feb 2022 16:11:23 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.812
X-Spam-Level: 
X-Spam-Status: No, score=-2.812 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, NICE_REPLY_A=-0.714, RCVD_IN_DNSWL_BLOCKED=0.001, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=bobbriscoe.net
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jU-43uJ1NHqv; Mon, 21 Feb 2022 16:11:18 -0800 (PST)
Received: from mail-ssdrsserver2.hostinginterface.eu (mail-ssdrsserver2.hostinginterface.eu [185.185.85.90]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 78B013A0DD4; Mon, 21 Feb 2022 16:11:16 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=bobbriscoe.net; s=default; h=In-Reply-To:From:References:Cc:To:Subject: MIME-Version:Date:Message-ID:Content-Type:Sender:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id: List-Help:List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=9lc0pYopr+ntFs5A9nk6BGOpoD53hr3UtH2Kgrd1ps4=; b=PbiZge0uJfe8HhsRMyvjDCE+Dx 5qjXUsUtiSYcJj2lf6w2fAjF014cKcH5FXBk4APGMB8YkmAuBs2FLbCpl5p2EHBcZ8wpBunvQuqqZ OToQQbVfqzbblCDOmTPnFnyp3lAnFgL/JAi8Z13qQXlBMNQ+A1sp2N++ye7VALBcb2OkWOtyvGpQe 1KBf4R3Qdc0phy3R4exzgcC5vA/y1EXvnd3QJ0batJgGIKdrkY161qa5r05nq9m7+HlevOep0k6cD x7hAOO+ImrQTSnAW5YDOH/qJgZrBt+ROVvc7dvOsRfjmjthK357JAHA7CIgiM11t0rmqI7kinhz04 CnC1i2Gw==;
Received: from 67.153.238.178.in-addr.arpa ([178.238.153.67]:55542 helo=[192.168.1.11]) by ssdrsserver2.hostinginterface.eu with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 (Exim 4.94.2) (envelope-from <in@bobbriscoe.net>) id 1nMIlx-00060P-Pb; Tue, 22 Feb 2022 00:11:14 +0000
Content-Type: multipart/alternative; boundary="------------xHSz3W9QleOe4XSBoUUSbSLJ"
Message-ID: <e8d7113c-47fa-7f91-54db-f7571840cd85@bobbriscoe.net>
Date: Tue, 22 Feb 2022 00:11:07 +0000
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.5.0
Content-Language: en-GB
To: Yoshifumi Nishida <nsd.ietf@gmail.com>, Markku Kojo <kojo@cs.helsinki.fi>
Cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi> <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com> <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org>
From: Bob Briscoe <in@bobbriscoe.net>
In-Reply-To: <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org>
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - ssdrsserver2.hostinginterface.eu
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - bobbriscoe.net
X-Get-Message-Sender-Via: ssdrsserver2.hostinginterface.eu: authenticated_id: in@bobbriscoe.net
X-Authenticated-Sender: ssdrsserver2.hostinginterface.eu: in@bobbriscoe.net
X-Source: 
X-Source-Args: 
X-Source-Dir: 
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/m1dht7W4qcADkTvY_EnI6GMz2hE>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 22 Feb 2022 00:11:24 -0000

This is a multi-part message in MIME format.
--------------xHSz3W9QleOe4XSBoUUSbSLJ
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Yoshi, Markku,

Many of the technical points Markku raised were important omissions from 
the rfc8312bis draft. Whatever status we give the eventual RFC, they 
should be highlighted as areas that need further work, either known 
weaknesses or just areas where we don't have enough data to know whether 
they are weaknesses. I tried to help find useful references for 
experiments already done around those points, and I tried to help write 
such text.

However, I cannot agree with some of Markku's statements about process.


      #1: The rules can change

It's OK to quote RFCs, but it's also OK for us to reach consensus that 
certain aspects of old RFCs are outdated, as long as we give our 
reasons, and there is rough consensus around that. We are all consenting 
adults - we don't have to be limited by RFCs from the past if we 
collectively don't want to be.

I have not noticed a need for the above "rule #1" in this discussion 
yet, but I'm just putting it out there, so if anyone needs to invoke it 
later, they can't be accused of inventing a rule for their own convenience.


      #2: There is no TCP-compatibility requirement

Yes, CUBIC is not TCP-compatible. But that's a feature, not a bug. Words 
like 'TCP-compatible' or 'fair' sound nice, but do not be fooled into 
thinking that an RFC recommends or requires you to behave in a certain 
way just because it has used a nice-sounding word for it.

For an RFC to recommend or require something, it has to say SHOULD or 
MUST. So when BCP 41 [RFC2914] defines 'TCP-compatible', and talks about 
what is necessary to be 'fair', it implies nothing about whether you 
should or must be TCP-compatible, or fair; unless it actually says you 
SHOULD or MUST be TCP-compatible, or fair. *No RFC says that.*

BCP 133 [RFC5033] and others have written careful words about straying 
from equal flow rates (quoted below). That gives a checklist for people 
like us who come along later. But it also respects our right to make a 
decision on whether/how to publish a CC RFC based on our own present-day 
knowledge and experience. To quote the relevant guideline in BCP 133 
[RFC5033]:

    (1) Impact on Standard TCP, SCTP [RFC2960  <https://www.rfc-editor.org/rfc/rfc2960>], and DCCP [RFC4340  <https://www.rfc-editor.org/rfc/rfc4340>].

        Proposed congestion control mechanisms should be evaluated when
        competing with standard IETF congestion control [RFC2581,
        RFC2960  <https://www.rfc-editor.org/rfc/rfc2960>,RFC4340  <https://www.rfc-editor.org/rfc/rfc4340>].  Alternate congestion controllers that have a
        significantly negative impact on traffic using standard
        congestion control may be suspect and this aspect should be part
        of the community's decision making with regards to the
        suitability of the alternate congestion control mechanism.

        We note that this bullet is not a requirement for strict TCP-
        friendliness as a prerequisite for an alternate congestion
        control mechanism to advance to Experimental.  As an example,
        HighSpeed TCP is a congestion control mechanism that is
        Experimental, but that is not TCP-friendly in all environments.
        We also note that this guideline does not constrain the fairness
        offered for non-best-effort traffic.

        As an example from an Experimental RFC, fairness with standard
        TCP is discussed in Sections4  <https://www.rfc-editor.org/rfc/rfc5033.html#section-4>  and6  <https://www.rfc-editor.org/rfc/rfc5033.html#section-6>  of [RFC3649  <https://www.rfc-editor.org/rfc/rfc3649>] (HighSpeed TCP)
        and*using spare capacity*  is discussed in Sections6  <https://www.rfc-editor.org/rfc/rfc5033.html#section-6>,11.1  <https://www.rfc-editor.org/rfc/rfc5033.html#section-11.1>, and12  <https://www.rfc-editor.org/rfc/rfc5033.html#section-12>
        of [RFC3649  <https://www.rfc-editor.org/rfc/rfc3649>].

The rfc8312bis draft has a section on 'using spare capacity' precisely 
because that's what was considered acceptable about CUBIC and the new 
generation of hi-BDP CCs developed at that time, as I've emphasized in 
the last para above.


      #3: A process for publishing EXP RFCs doesn't cover publishing
      non-EXP RFCs (by definition)

This IESG statement on "Experimental Specification of New Congestion 
Control Algorithms 
<https://www.ietf.org/about/groups/iesg/statements/experimental-congestion-control/>" 
says it defines the process for a CC to be published as an Experimental 
RFC. It does not say it is the process that /all/ CCs have to follow to 
enter the standards track.

Before invoking that process, one has to determine whether it is 
applicable to this case. The IESG statement contains a sentence about 
how to determine its applicability:

    The decision of whether a specific new congestion control schemes
    goes beyond the principles of [RFC2914] and hence will undergo the
    process described in this document lies with the transport area
    directors, who will use the expertise of the transport area
    directorate and other congestion control experts to aid their decision.

Since RFC2914, we've had RFC3246 explaining why Reno is unscalable, and 
RFC5033 (quoted above) giving guidelines for designers of new CCs. So we 
can debate whether the process that Markku quoted is relevant to this 
case, to help the AD to decide whether to follow the process or not.


      #4 EXP or STD? My view

Reno is worse than other CCs in some respects. Nearly everyone has 
abandoned it. So, if someone asked you why Reno is the IETF's only stds 
track CC, what would you say ...?

As I already said, we have to stop thinking that the stds track requires 
a CC algorithm to be perfect. A CC certainly mustn't have failure or 
collapse modes, but it's performance doesn't have to be perfect in all 
cases. For instance, CUBIC has known poor utilization over radio links. 
That didn't stop Reno from being on the stds track. For me, the IETF 
needs a new CC on the stds track, and CUBIC is the most natural candidate.

However, we must not set a precedent for any old CC to move to the stds 
track. So we must document it's weaknesses and failings, and explain why 
they are not serious (if that is the case).

New CCs need to interoperate with the most prevalent CC on the Internet 
(CUBIC). The IETF cannot even say that as long as it hasn't formally 
recognized CUBIC on the stds track. That is a failing of *us* in the 
tcpm WG. We can move CUBIC to stds track via the exp track if we have 
to, or we can go straight to stds track. The only difference is that the 
first option makes us look like jobsworths for longer.

jobsworth
/ˈdʒɒbzwəːθ/
/noun/
  derogatory•informal
an official who upholds petty rules even at the expense of humanity or 
common sense.
1970s: from "I can't do that, it's more than my job's worth."


Bob

-- 
________________________________________________________________
Bob Briscoehttp://bobbriscoe.net/

--------------xHSz3W9QleOe4XSBoUUSbSLJ
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    Yoshi, Markku,<br>
    <br>
    Many of the technical points Markku raised were important omissions
    from the rfc8312bis draft. Whatever status we give the eventual RFC,
    they should be highlighted as areas that need further work, either
    known weaknesses or just areas where we don't have enough data to
    know whether they are weaknesses. I tried to help find useful
    references for experiments already done around those points, and I
    tried to help write such text.<br>
    <br>
    However, I cannot agree with some of Markku's statements about
    process. <br>
    <br>
    <h3>#1: The rules can change </h3>
    It's OK to quote RFCs, but it's also OK for us to reach consensus
    that certain aspects of old RFCs are outdated, as long as we give
    our reasons, and there is rough consensus around that. We are all
    consenting adults - we don't have to be limited by RFCs from the
    past if we collectively don't want to be.<br>
    <br>
    I have not noticed a need for the above "rule #1" in this discussion
    yet, but I'm just putting it out there, so if anyone needs to invoke
    it later, they can't be accused of inventing a rule for their own
    convenience.<br>
    <br>
    <h3>#2: There is no TCP-compatibility requirement </h3>
    Yes, CUBIC is not TCP-compatible. But that's a feature, not a bug.
    Words like 'TCP-compatible' or 'fair' sound nice, but do not be
    fooled into thinking that an RFC recommends or requires you to
    behave in a certain way just because it has used a nice-sounding
    word for it. <br>
    <br>
    For an RFC to recommend or require something, it has to say SHOULD
    or MUST. So when BCP 41 [RFC2914] defines 'TCP-compatible', and
    talks about what is necessary to be 'fair', it implies nothing about
    whether you should or must be TCP-compatible, or fair; unless it
    actually says you SHOULD or MUST be TCP-compatible, or fair. <b>No
      RFC says that.</b> <br>
    <br>
    BCP 133 [RFC5033] and others have written careful words about
    straying from equal flow rates (quoted below). That gives a
    checklist for people like us who come along later. But it also
    respects our right to make a decision on whether/how to publish a CC
    RFC based on our own present-day knowledge and experience. To quote
    the relevant guideline in BCP 133 [RFC5033]:<br>
    <pre class="newpage">   (1) Impact on Standard TCP, SCTP [<a href="https://www.rfc-editor.org/rfc/rfc2960" title="&quot;Stream Control Transmission Protocol&quot;">RFC2960</a>], and DCCP [<a href="https://www.rfc-editor.org/rfc/rfc4340" title="&quot;Datagram Congestion Control Protocol (DCCP)&quot;">RFC4340</a>].

       Proposed congestion control mechanisms should be evaluated when
       competing with standard IETF congestion control [RFC2581,
       <a href="https://www.rfc-editor.org/rfc/rfc2960">RFC2960</a>, <a href="https://www.rfc-editor.org/rfc/rfc4340">RFC4340</a>].  Alternate congestion controllers that have a
       significantly negative impact on traffic using standard
       congestion control may be suspect and this aspect should be part
       of the community's decision making with regards to the
       suitability of the alternate congestion control mechanism.

       We note that this bullet is not a requirement for strict TCP-
       friendliness as a prerequisite for an alternate congestion
       control mechanism to advance to Experimental.  As an example,
       HighSpeed TCP is a congestion control mechanism that is
       Experimental, but that is not TCP-friendly in all environments.
       We also note that this guideline does not constrain the fairness
       offered for non-best-effort traffic.

       As an example from an Experimental RFC, fairness with standard
       TCP is discussed in Sections <a href="https://www.rfc-editor.org/rfc/rfc5033.html#section-4">4</a> and <a href="https://www.rfc-editor.org/rfc/rfc5033.html#section-6">6</a> of [<a href="https://www.rfc-editor.org/rfc/rfc3649" title="&quot;HighSpeed TCP for Large Congestion Windows&quot;">RFC3649</a>] (HighSpeed TCP)
       and <b>using spare capacity</b> is discussed in Sections <a href="https://www.rfc-editor.org/rfc/rfc5033.html#section-6">6</a>, <a href="https://www.rfc-editor.org/rfc/rfc5033.html#section-11.1">11.1</a>, and <a href="https://www.rfc-editor.org/rfc/rfc5033.html#section-12">12</a>
       of [<a href="https://www.rfc-editor.org/rfc/rfc3649" title="&quot;HighSpeed TCP for Large Congestion Windows&quot;">RFC3649</a>].</pre>
    The rfc8312bis draft has a section on 'using spare capacity'
    precisely because that's what was considered acceptable about CUBIC
    and the new generation of hi-BDP CCs developed at that time, as I've
    emphasized in the last para above. <br>
    <br>
    <h3>#3: A process for publishing EXP RFCs doesn't cover publishing
      non-EXP RFCs (by definition) <br>
    </h3>
    This IESG statement on "<a
href="https://www.ietf.org/about/groups/iesg/statements/experimental-congestion-control/">Experimental
      Specification of New Congestion Control Algorithms</a>" says it
    defines the process for a CC to be published as an Experimental RFC.
    It does not say it is the process that /all/ CCs have to follow to
    enter the standards track. <br>
    <br>
    Before invoking that process, one has to determine whether it is
    applicable to this case. The IESG statement contains a sentence
    about how to determine its applicability:<br>
    <blockquote><font face="monospace">The decision of whether a
        specific new congestion control schemes goes beyond the
        principles of [RFC2914] and hence will undergo the process
        described in this document lies with the transport area
        directors, who will use the expertise of the transport area
        directorate and other congestion control experts to aid their
        decision.</font><br>
    </blockquote>
    Since RFC2914, we've had RFC3246 explaining why Reno is unscalable,
    and RFC5033 (quoted above) giving guidelines for designers of new
    CCs. So we can debate whether the process that Markku quoted is
    relevant to this case, to help the AD to decide whether to follow
    the process or not.<br>
    <br>
    <h3>#4 EXP or STD? My view <br>
    </h3>
    Reno is worse than other CCs in some respects. Nearly everyone has
    abandoned it. So, if someone asked you why Reno is the IETF's only
    stds track CC, what would you say ...?<br>
    <br>
    As I already said, we have to stop thinking that the stds track
    requires a CC algorithm to be
    perfect. A CC certainly mustn't have failure or collapse modes, but
    it's performance doesn't have to be perfect in all cases. For
    instance, CUBIC has known poor utilization over radio links. That
    didn't stop Reno from being on the stds track. For me, the IETF
    needs a new CC on the stds track, and CUBIC is the most natural
    candidate. <br>
    <br>
    However, we must not set a precedent for any old CC to move to the
    stds track. So we must document it's weaknesses and failings, and
    explain why they are not serious (if that is the case). <br>
    <br>
    New CCs need to interoperate with the most prevalent CC on the
    Internet (CUBIC). The IETF cannot even say that as long as it hasn't
    formally recognized CUBIC on the stds track. That is a failing of
    *us* in the tcpm WG. We can move CUBIC to stds track via the exp
    track if we have to, or we can go straight to stds track. The only
    difference is that the first option makes us look like jobsworths
    for longer. <br>
    <br>
    <div class="WI9k4c">
      <div class="jY7QFf" style="margin-bottom:0;line-height:normal">
        <div class="c8d6zd DgZBFd ya2TWb"><span data-dobid="hdw">jobsworth</span></div>
      </div>
      <div class="S23sjd"><span class="LTKOO">/<span>ˈdʒɒbzwəːθ</span>/</span></div>
    </div>
    <div class="lW8rQd">
      <div style="display:inline" data-dobid="dfn">
        <div class="pgRvse YrbPuc vdBwhd"><i><span class="YrbPuc">noun</span></i>
          <div style="display:inline-block"><span class="mQo3nc aztjNb
              emHvJd"><span> derogatory</span><span class="BkZwPe">•</span><span>informal</span></span></div>
        </div>
        <span>an official who upholds petty rules even at the expense of
          humanity or common sense.</span></div>
    </div>
    <span><span>1970s: from "I can't do that, it's more than my job's
        worth."<br>
        <br>
      </span></span><br>
    Bob<br>
    <br>
    <pre class="moz-signature" cols="72">-- 
________________________________________________________________
Bob Briscoe                               <a class="moz-txt-link-freetext" href="http://bobbriscoe.net/">http://bobbriscoe.net/</a></pre>
  </body>
</html>

--------------xHSz3W9QleOe4XSBoUUSbSLJ--


From nobody Tue Feb 22 02:03:51 2022
Return-Path: <Michael.Scharf@hs-esslingen.de>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 733023A0C38; Tue, 22 Feb 2022 02:03:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.097
X-Spam-Level: 
X-Spam-Status: No, score=-2.097 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, SPF_HELO_NONE=0.001, SPF_NONE=0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=hs-esslingen.de
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id cL62eJxbMhBQ; Tue, 22 Feb 2022 02:03:43 -0800 (PST)
Received: from mail.hs-esslingen.de (mail.hs-esslingen.de [134.108.32.78]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4ADE73A0C52; Tue, 22 Feb 2022 02:03:41 -0800 (PST)
Received: from localhost (localhost.localdomain [127.0.0.1]) by mail.hs-esslingen.de (Postfix) with ESMTP id 7B42A25A13; Tue, 22 Feb 2022 11:03:39 +0100 (CET)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=hs-esslingen.de; s=mail; t=1645524219; bh=rhIx6CO4g2itifNGy+B2D6RWyHA6gAm2cC45+GcJONQ=; h=From:To:CC:Subject:Date:References:In-Reply-To:From; b=nZ95eJSFCNtAzjJF94exnpP9q4Ky0uGraa0CZ21nXZSksGfAxUX910iq0fHGYYe9j Kj7e28lu5GWje3p4R9fLN8DTQSS8gFtjYDORJ9TnWBWYTrytsVAK7OQ0QmfKZ0k2Qq Id/YaMyHDIqwcndslX9Hv+u4S2YpwPVMJKhdEQnc=
X-Virus-Scanned: by amavisd-new-2.7.1 (20120429) (Debian) at hs-esslingen.de
Received: from mail.hs-esslingen.de ([127.0.0.1]) by localhost (hs-esslingen.de [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id rWsZoVglDOaO; Tue, 22 Feb 2022 11:03:38 +0100 (CET)
Received: from rznt8202.rznt.rzdir.fht-esslingen.de (rznt8202.hs-esslingen.de [134.108.48.165]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.hs-esslingen.de (Postfix) with ESMTPS; Tue, 22 Feb 2022 11:03:38 +0100 (CET)
Received: from rznt8202.rznt.rzdir.fht-esslingen.de (134.108.48.165) by rznt8202.rznt.rzdir.fht-esslingen.de (134.108.48.165) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.18; Tue, 22 Feb 2022 11:03:38 +0100
Received: from rznt8202.rznt.rzdir.fht-esslingen.de ([fe80::aca4:171a:3ee1:57e0]) by rznt8202.rznt.rzdir.fht-esslingen.de ([fe80::aca4:171a:3ee1:57e0%3]) with mapi id 15.01.2375.018; Tue, 22 Feb 2022 11:03:38 +0100
From: "Scharf, Michael" <Michael.Scharf@hs-esslingen.de>
To: Lars Eggert <lars@eggert.org>, Yoshifumi Nishida <nsd.ietf@gmail.com>
CC: Markku Kojo <kojo@cs.helsinki.fi>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
Thread-Topic: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
Thread-Index: AQHYFz9qjcNEwGI41kmv6ftBUhB7tqyPZSAAgAZtPoCAAgdFgIABLUIAgAAdDACABLxoAIAARZ2AgAFKj+A=
Date: Tue, 22 Feb 2022 10:03:38 +0000
Message-ID: <56bc8db8ee5d438699eafd60650453ba@hs-esslingen.de>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi> <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com> <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org>
In-Reply-To: <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [134.108.140.249]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5pQ6LrDE>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 22 Feb 2022 10:03:49 -0000

Hi all,

As always in a discussion about EXP vs. PS, I suggest having a look at the =
definition of Proposed Standard, e.g., in RFC 7127:

   A Proposed Standard specification is stable, has resolved known
   design choices, has received significant community review, and
   appears to enjoy enough community interest to be considered valuable.

   Usually, neither implementation nor operational experience is
   required for the designation of a specification as a Proposed
   Standard.  However, such experience is highly desirable and will
   usually represent a strong argument in favor of a Proposed Standard
   designation.

   The IESG may require implementation and/or operational experience
   prior to granting Proposed Standard status to a specification that
   materially affects the core Internet protocols or that specifies
   behavior that may have significant operational impact on the
   Internet.

   A Proposed Standard will have no known technical omissions with
   respect to the requirements placed upon it.  Proposed Standards are
   of such quality that implementations can be deployed in the Internet.
   However, as with all technical specifications, Proposed Standards may
   be revised if problems are found or better solutions are identified,
   when experiences with deploying implementations of such technologies
   at scale is gathered.

As an individual contributor to TCP and congestion control research, I beli=
eve that a CUBIC specification can meet these requirements.

Maybe the "no known technical omissions with respect to the requirements pl=
aced upon it" could benefit from some text inside rfc8312bis to make clear =
what the requirements are, i.e., what the standard tries to achieve (and wh=
at not). But, at least to me, the current section 5 already covers that. I =
am not against adding some more text on the scope of rfc8312bis, but I'll c=
ertainly not insist in that.

Also, another repetition of what I wrote already in the past:

In my observation, TCPM is extremely conservative when it comes to standard=
s track documents, as compared to other WGs in the IETF (also outside TSV, =
e.g., in RTG or OPS). Probably that is not a bug, but a feature in a standa=
rdization body. And at least I have not tried to change that thinking as ch=
air in the last 10 years. Nonetheless, if a document can meet the bar for a=
 PS, I believe TCPM should try to go for a PS.

Michael (not wearing any hat)


> -----Original Message-----
> From: Lars Eggert <lars@eggert.org>
> Sent: Monday, February 21, 2022 3:52 PM
> To: Yoshifumi Nishida <nsd.ietf@gmail.com>
> Cc: Markku Kojo <kojo@cs.helsinki.fi>; tcpm@ietf.org Extensions
> <tcpm@ietf.org>; tcpm-chairs <tcpm-chairs@ietf.org>
> Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
>=20
> Hi,
>=20
> first, as an editor (but not wearing other hats) I do object against recl=
assifying
> the document as Experimental or Informational, esp. this late in the proc=
ess.
>=20
> CUBIC is not a new CC proposal. It's been ubiquitously deployed on the
> Internet for many years. All major TCP stacks use it by default. Given th=
at, I
> don't see how the IESG statement on "Experimental Specification of New
> Congestion Control Algorithms" applies here.
>=20
> (As an aside, that statement was specifically written when CUBIC, Compoun=
d
> and Hamilton were all jostling for RFC publication. The community converg=
ed
> on CUBIC in the meantime.)
>=20
> CUBIC *is* the default CC mechanism used on the Internet and in other IP
> networks. It should be published on the standards track.
>=20
> Second, this is the second time that an individual raised a long list of =
issues
> after the WGLC has ended. We did try to work through all the issues that
> were raised after the first WGLC, but the individual often failed to part=
icipate
> in the discussion in a timely manner. I now see at least several of those=
 same
> issues re-raised by the same individual. Before going down the same path =
a
> second time, I would actually ask the WG to confirm there is consensus th=
at
> the raised issues are valid (and if they are valid, commit to contributin=
g to a
> resolution in a timely manner.)
>=20
> Thanks,
> Lars


From nobody Tue Feb 22 03:13:36 2022
Return-Path: <in@bobbriscoe.net>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 533BC3A0DF7; Tue, 22 Feb 2022 03:13:34 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -7.814
X-Spam-Level: 
X-Spam-Status: No, score=-7.814 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, NICE_REPLY_A=-0.714, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=bobbriscoe.net
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oaT61HN5ZJB4; Tue, 22 Feb 2022 03:13:29 -0800 (PST)
Received: from mail-ssdrsserver2.hostinginterface.eu (mail-ssdrsserver2.hostinginterface.eu [185.185.85.90]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 293673A0DFA; Tue, 22 Feb 2022 03:13:26 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=bobbriscoe.net; s=default; h=Content-Transfer-Encoding:Content-Type: In-Reply-To:References:Cc:To:From:Subject:MIME-Version:Date:Message-ID:Sender :Reply-To:Content-ID:Content-Description:Resent-Date:Resent-From: Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id:List-Help: List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=DLNJrx8te0P6guI8Mie9Dz7ye+KsmgSFB9FIHMA1GgA=; b=Vdy72IrXS9KIjhqmdbcuzRdeiR 3Li/buv1rKFPkkVT0a3Vg0HaNAfaGMC9b/Q8Rjdm6TIY21QA8hAMcWwwgPwMPufRQcmEGe9b/UDJ4 13MesdsMd82+yuV+pKQIqZeprDUVOxq403dfoikttI4/BpYLxDPeZZ4DrU260MbbbS3++ERfHaB8T PKd/ca1O0VRFjgoMKKZShwl9Z2UEPIqXeu2lBYZbp4Tu7y8Ef4OiLMqSKAIwlIDGEDU4MYU5lYnEa Bt8CrIrSM+KFJvykvsuoOhGI5mflwVea/eHK7uuWqYNpx5W88Zz854M/2NoMWzp+s6knD5/MCzGoz za6RBMIA==;
Received: from 67.153.238.178.in-addr.arpa ([178.238.153.67]:55578 helo=[192.168.1.11]) by ssdrsserver2.hostinginterface.eu with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 (Exim 4.94.2) (envelope-from <in@bobbriscoe.net>) id 1nMT6m-00027P-B7; Tue, 22 Feb 2022 11:13:24 +0000
Message-ID: <a51a6d8e-a22f-e023-2b16-99a49c3246ed@bobbriscoe.net>
Date: Tue, 22 Feb 2022 11:13:22 +0000
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.5.0
Content-Language: en-GB
From: Bob Briscoe <in@bobbriscoe.net>
To: Yoshifumi Nishida <nsd.ietf@gmail.com>, Markku Kojo <kojo@cs.helsinki.fi>
Cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi> <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com> <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org> <e8d7113c-47fa-7f91-54db-f7571840cd85@bobbriscoe.net>
In-Reply-To: <e8d7113c-47fa-7f91-54db-f7571840cd85@bobbriscoe.net>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - ssdrsserver2.hostinginterface.eu
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - bobbriscoe.net
X-Get-Message-Sender-Via: ssdrsserver2.hostinginterface.eu: authenticated_id: in@bobbriscoe.net
X-Authenticated-Sender: ssdrsserver2.hostinginterface.eu: in@bobbriscoe.net
X-Source: 
X-Source-Args: 
X-Source-Dir: 
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/J5WWpNDsJmDFlWAZ3zY30BVF9os>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 22 Feb 2022 11:13:35 -0000

All,

On 22/02/2022 00:11, Bob Briscoe wrote:
> Since RFC2914, we've had RFC3246 explaining why Reno is unscalable,

Sry, I meant RFC3649 "HighSpeed TCP for Large Congestion Windows" not 
RFC3246 (Expedited Forwarding)!


Bob

-- 
________________________________________________________________
Bob Briscoe                               http://bobbriscoe.net/


From nobody Tue Feb 22 21:04:38 2022
Return-Path: <internet-drafts@ietf.org>
X-Original-To: tcpm@ietf.org
Delivered-To: tcpm@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 26C373A08B6; Tue, 22 Feb 2022 21:04:30 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <i-d-announce@ietf.org>
Cc: tcpm@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 7.45.0
Auto-Submitted: auto-generated
Precedence: bulk
Reply-To: tcpm@ietf.org
Message-ID: <164559267002.12452.17995648680457765047@ietfa.amsl.com>
Date: Tue, 22 Feb 2022 21:04:30 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/E1BZhiHIqVZTjEbjAfLzcKOCk90>
Subject: [tcpm] I-D Action: draft-ietf-tcpm-rfc793bis-27.txt
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Feb 2022 05:04:31 -0000

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the TCP Maintenance and Minor Extensions WG of the IETF.

        Title           : Transmission Control Protocol (TCP) Specification
        Author          : Wesley M. Eddy
	Filename        : draft-ietf-tcpm-rfc793bis-27.txt
	Pages           : 114
	Date            : 2022-02-22

Abstract:
   This document specifies the Transmission Control Protocol (TCP).  TCP
   is an important transport layer protocol in the Internet protocol
   stack, and has continuously evolved over decades of use and growth of
   the Internet.  Over this time, a number of changes have been made to
   TCP as it was specified in RFC 793, though these have only been
   documented in a piecemeal fashion.  This document collects and brings
   those changes together with the protocol specification from RFC 793.
   This document obsoletes RFC 793, as well as RFCs 879, 2873, 6093,
   6429, 6528, and 6691 that updated parts of RFC 793.  It updates RFCs
   1011 and 1122, and should be considered as a replacement for the
   portions of those document dealing with TCP requirements.  It also
   updates RFC 5961 by adding a small clarification in reset handling
   while in the SYN-RECEIVED state.  The TCP header control bits from
   RFC 793 have also been updated based on RFC 3168.

   RFC EDITOR NOTE: If approved for publication as an RFC, this should
   be marked additionally as "STD: 7" and replace RFC 793 in that role.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc793bis/

There is also an htmlized version available at:
https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-rfc793bis-27

A diff from the previous version is available at:
https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rfc793bis-27


Internet-Drafts are also available by rsync at rsync.ietf.org::internet-drafts



From nobody Wed Feb 23 01:17:04 2022
Return-Path: <nsd.ietf@gmail.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E85903A0EE1; Wed, 23 Feb 2022 01:17:01 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.097
X-Spam-Level: 
X-Spam-Status: No, score=-2.097 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id y2CNQfQvlILF; Wed, 23 Feb 2022 01:16:57 -0800 (PST)
Received: from mail-qv1-xf2d.google.com (mail-qv1-xf2d.google.com [IPv6:2607:f8b0:4864:20::f2d]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CDEDA3A0EDF; Wed, 23 Feb 2022 01:16:56 -0800 (PST)
Received: by mail-qv1-xf2d.google.com with SMTP id e22so7142780qvf.9; Wed, 23 Feb 2022 01:16:56 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112;  h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=8j2O7tw0kLQZ3Oyb3WF9ntiCkffdq/sJWUJb3dO3VnM=; b=GXjb9jkHUOxSwmJjswb2TpYbmr/UTOvsNlChts4fE3ryN60AH+d5NQpSxMn3xnVNsC clVYExcJccHf7h0+aFQPVWkF37osxg79Yb5f3GRtBAu5fPIbaTAN1cjjToqEGJ/SIYx0 6guW9yrzgnfDWhA5EQG3OT6yfc6a3SLvTVAgxeo74GbHmkkmF79wEOs9drVTZgvCI7+e o0nX0S1faslsXuS2Daw6MYdpSIqRFvdH8I9LtXKs4HW/A+rA6KrVAwkf4Tp3Dixm9LPK 945KyvaRE0e4+aAJyQA6tSZQkrgAARDA4CLO9sJeCJfqFt/TTFVDFxp6uTMRI8Ufhdtx P/nQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=8j2O7tw0kLQZ3Oyb3WF9ntiCkffdq/sJWUJb3dO3VnM=; b=siPgugcQ7tpYSVenx+C/Ns8iYxednrXGavnV8D0QSn0xoXxTmMCmgY0SBLp4Pa8jXx BtH+5lejJlFnlyMIcbTYgs7l7VBW60yHdDkpj7H13hDoe5TGBYgfsDzDK/Xs7mGMAtOd uIhk5M8zUuQkof0z1jFr5RnbD9lj0YDPZzA4nhb/uTaa6tJU21m8EKwXp0CObXt8v1T3 ViIun4w06Cqm5vukz/w5PErcu5Ict01rEErF9BJUa9KigEOr/g6iSFsvBquewR+EgoQk ozvtGTA9xxsHM6Dmk7XORN+8iPfJy9WzPTwEg0JHNlpAaWRqqMQrGskmWUzlcWbr8Ofn RYKg==
X-Gm-Message-State: AOAM532TuSuMfOfS4d/zAY+Ywla93ih89iJ8pbPzaBiKlebt9/bjZcFV ZfEcS2tgKBWv3QFKSyqia5FjD1pCZFQkwxf71jOUh/sf
X-Google-Smtp-Source: ABdhPJwcHwM8FrgfyqbBralUv5icDce8JlhkyTI2wlg3HtXyaLPcLND+FPzbyUybcHjB0JjVbmcNIEL8Xa+61sBn77c=
X-Received: by 2002:ad4:5de2:0:b0:42c:452b:351d with SMTP id jn2-20020ad45de2000000b0042c452b351dmr22987828qvb.111.1645607815520; Wed, 23 Feb 2022 01:16:55 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi> <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com> <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org> <e8d7113c-47fa-7f91-54db-f7571840cd85@bobbriscoe.net>
In-Reply-To: <e8d7113c-47fa-7f91-54db-f7571840cd85@bobbriscoe.net>
From: Yoshifumi Nishida <nsd.ietf@gmail.com>
Date: Wed, 23 Feb 2022 01:16:44 -0800
Message-ID: <CAAK044RCR6zdVKb602+W0V13_fbmDW5PaEtP+Bgz6OAEUtOifA@mail.gmail.com>
To: Bob Briscoe <in@bobbriscoe.net>
Cc: Markku Kojo <kojo@cs.helsinki.fi>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>,  tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: multipart/alternative; boundary="00000000000048a4cf05d8abedc5"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/OArT4VCGKLUEISIdv92y__IUQvE>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Feb 2022 09:17:02 -0000

--00000000000048a4cf05d8abedc5
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi Bob, everyone,

I also think the points raised by Markku are valid and should be
highlighted for future work. But, at the same time, I am not sure it is
reasonable to delay producing standard docs.

What I am wondering is if it would make sense to publish these points as an
RFC to indicate "we believe they are acceptable for PS, but will need long
term analysis and evaluations as the Internet evolves."
Because we know they are a bit more aggressive than what have been done
before, but they are the feature, not bugs as Bob mentioned.
In this way, we don't have to hold the doc for a long time.

I think we can add more points in this doc, for example, initial window.
I believe we'll discuss promoting RFC6928 as a PS in the near future, but
if we apply "wide deployments and experiences are not enough" policy, it
may take a very long time while RFC9002 already allows it.
I am thinking this approach could allow us to treat IW10 for TCP and IW10
for QUIC more or less equally.
--
Yoshi

On Mon, Feb 21, 2022 at 4:11 PM Bob Briscoe <in@bobbriscoe.net> wrote:

> Yoshi, Markku,
>
> Many of the technical points Markku raised were important omissions from
> the rfc8312bis draft. Whatever status we give the eventual RFC, they shou=
ld
> be highlighted as areas that need further work, either known weaknesses o=
r
> just areas where we don't have enough data to know whether they are
> weaknesses. I tried to help find useful references for experiments alread=
y
> done around those points, and I tried to help write such text.
>
> However, I cannot agree with some of Markku's statements about process.
>
> #1: The rules can change It's OK to quote RFCs, but it's also OK for us
> to reach consensus that certain aspects of old RFCs are outdated, as long
> as we give our reasons, and there is rough consensus around that. We are
> all consenting adults - we don't have to be limited by RFCs from the past
> if we collectively don't want to be.
>
> I have not noticed a need for the above "rule #1" in this discussion yet,
> but I'm just putting it out there, so if anyone needs to invoke it later,
> they can't be accused of inventing a rule for their own convenience.
>
> #2: There is no TCP-compatibility requirement Yes, CUBIC is not
> TCP-compatible. But that's a feature, not a bug. Words like
> 'TCP-compatible' or 'fair' sound nice, but do not be fooled into thinking
> that an RFC recommends or requires you to behave in a certain way just
> because it has used a nice-sounding word for it.
>
> For an RFC to recommend or require something, it has to say SHOULD or
> MUST. So when BCP 41 [RFC2914] defines 'TCP-compatible', and talks about
> what is necessary to be 'fair', it implies nothing about whether you shou=
ld
> or must be TCP-compatible, or fair; unless it actually says you SHOULD or
> MUST be TCP-compatible, or fair. *No RFC says that.*
>
> BCP 133 [RFC5033] and others have written careful words about straying
> from equal flow rates (quoted below). That gives a checklist for people
> like us who come along later. But it also respects our right to make a
> decision on whether/how to publish a CC RFC based on our own present-day
> knowledge and experience. To quote the relevant guideline in BCP 133
> [RFC5033]:
>
>    (1) Impact on Standard TCP, SCTP [RFC2960 <https://www.rfc-editor.org/=
rfc/rfc2960>], and DCCP [RFC4340 <https://www.rfc-editor.org/rfc/rfc4340>].
>
>        Proposed congestion control mechanisms should be evaluated when
>        competing with standard IETF congestion control [RFC2581,
>        RFC2960 <https://www.rfc-editor.org/rfc/rfc2960>, RFC4340 <https:/=
/www.rfc-editor.org/rfc/rfc4340>].  Alternate congestion controllers that h=
ave a
>        significantly negative impact on traffic using standard
>        congestion control may be suspect and this aspect should be part
>        of the community's decision making with regards to the
>        suitability of the alternate congestion control mechanism.
>
>        We note that this bullet is not a requirement for strict TCP-
>        friendliness as a prerequisite for an alternate congestion
>        control mechanism to advance to Experimental.  As an example,
>        HighSpeed TCP is a congestion control mechanism that is
>        Experimental, but that is not TCP-friendly in all environments.
>        We also note that this guideline does not constrain the fairness
>        offered for non-best-effort traffic.
>
>        As an example from an Experimental RFC, fairness with standard
>        TCP is discussed in Sections 4 <https://www.rfc-editor.org/rfc/rfc=
5033.html#section-4> and 6 <https://www.rfc-editor.org/rfc/rfc5033.html#sec=
tion-6> of [RFC3649 <https://www.rfc-editor.org/rfc/rfc3649>] (HighSpeed TC=
P)
>        and *using spare capacity* is discussed in Sections 6 <https://www=
.rfc-editor.org/rfc/rfc5033.html#section-6>, 11.1 <https://www.rfc-editor.o=
rg/rfc/rfc5033.html#section-11.1>, and 12 <https://www.rfc-editor.org/rfc/r=
fc5033.html#section-12>
>        of [RFC3649 <https://www.rfc-editor.org/rfc/rfc3649>].
>
> The rfc8312bis draft has a section on 'using spare capacity' precisely
> because that's what was considered acceptable about CUBIC and the new
> generation of hi-BDP CCs developed at that time, as I've emphasized in th=
e
> last para above.
>
> #3: A process for publishing EXP RFCs doesn't cover publishing non-EXP
> RFCs (by definition)
> This IESG statement on "Experimental Specification of New Congestion
> Control Algorithms
> <https://www.ietf.org/about/groups/iesg/statements/experimental-congestio=
n-control/>"
> says it defines the process for a CC to be published as an Experimental
> RFC. It does not say it is the process that /all/ CCs have to follow to
> enter the standards track.
>
> Before invoking that process, one has to determine whether it is
> applicable to this case. The IESG statement contains a sentence about how
> to determine its applicability:
>
> The decision of whether a specific new congestion control schemes goes
> beyond the principles of [RFC2914] and hence will undergo the process
> described in this document lies with the transport area directors, who wi=
ll
> use the expertise of the transport area directorate and other congestion
> control experts to aid their decision.
>
> Since RFC2914, we've had RFC3246 explaining why Reno is unscalable, and
> RFC5033 (quoted above) giving guidelines for designers of new CCs. So we
> can debate whether the process that Markku quoted is relevant to this cas=
e,
> to help the AD to decide whether to follow the process or not.
>
> #4 EXP or STD? My view
> Reno is worse than other CCs in some respects. Nearly everyone has
> abandoned it. So, if someone asked you why Reno is the IETF's only stds
> track CC, what would you say ...?
>
> As I already said, we have to stop thinking that the stds track requires =
a
> CC algorithm to be perfect. A CC certainly mustn't have failure or collap=
se
> modes, but it's performance doesn't have to be perfect in all cases. For
> instance, CUBIC has known poor utilization over radio links. That didn't
> stop Reno from being on the stds track. For me, the IETF needs a new CC o=
n
> the stds track, and CUBIC is the most natural candidate.
>
> However, we must not set a precedent for any old CC to move to the stds
> track. So we must document it's weaknesses and failings, and explain why
> they are not serious (if that is the case).
>
> New CCs need to interoperate with the most prevalent CC on the Internet
> (CUBIC). The IETF cannot even say that as long as it hasn't formally
> recognized CUBIC on the stds track. That is a failing of *us* in the tcpm
> WG. We can move CUBIC to stds track via the exp track if we have to, or w=
e
> can go straight to stds track. The only difference is that the first opti=
on
> makes us look like jobsworths for longer.
>
> jobsworth
> /=CB=88d=CA=92=C9=92bzw=C9=99=CB=90=CE=B8/
> *noun*
>  derogatory=E2=80=A2informal
> an official who upholds petty rules even at the expense of humanity or
> common sense.
> 1970s: from "I can't do that, it's more than my job's worth."
>
>
> Bob
>
> --
> ________________________________________________________________
> Bob Briscoe                               http://bobbriscoe.net/
>
>

--00000000000048a4cf05d8abedc5
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi Bob, everyone,</div><div><br></div><div>I also thi=
nk the points raised by Markku are valid and should be highlighted for futu=
re work. But, at the same time, I am not sure it is reasonable=C2=A0to dela=
y producing standard docs.=C2=A0 =C2=A0</div><div><br></div><div>What I am =
wondering is if it would make sense to publish these points as an RFC to in=
dicate &quot;we believe they are acceptable for PS, but will need long term=
 analysis and evaluations as the Internet evolves.&quot;=C2=A0</div><div>Be=
cause we know they are a bit more aggressive than what=C2=A0have been done =
before, but they are the feature,=C2=A0not bugs as=C2=A0Bob mentioned.</div=
><div>In this way, we don&#39;t have to hold the doc for a long time.<br></=
div><div><br></div><div>I think we can add more points in this doc, for exa=
mple, initial window.<br></div><div>I believe we&#39;ll discuss promoting R=
FC6928 as a PS in the near future, but if we apply &quot;wide deployments a=
nd experiences are not enough&quot; policy, it may take a very long time wh=
ile RFC9002 already allows it.</div><div>I am thinking=C2=A0this approach c=
ould allow us=C2=A0to treat IW10 for TCP and IW10 for QUIC more or less equ=
ally.</div><div>--</div><div>Yoshi</div><br><div class=3D"gmail_quote"><div=
 dir=3D"ltr" class=3D"gmail_attr">On Mon, Feb 21, 2022 at 4:11 PM Bob Brisc=
oe &lt;<a href=3D"mailto:in@bobbriscoe.net">in@bobbriscoe.net</a>&gt; wrote=
:<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.=
8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
 =20
   =20
 =20
  <div>
    Yoshi, Markku,<br>
    <br>
    Many of the technical points Markku raised were important omissions
    from the rfc8312bis draft. Whatever status we give the eventual RFC,
    they should be highlighted as areas that need further work, either
    known weaknesses or just areas where we don&#39;t have enough data to
    know whether they are weaknesses. I tried to help find useful
    references for experiments already done around those points, and I
    tried to help write such text.<br>
    <br>
    However, I cannot agree with some of Markku&#39;s statements about
    process. <br>
    <br>
    <h3>#1: The rules can change </h3>
    It&#39;s OK to quote RFCs, but it&#39;s also OK for us to reach consens=
us
    that certain aspects of old RFCs are outdated, as long as we give
    our reasons, and there is rough consensus around that. We are all
    consenting adults - we don&#39;t have to be limited by RFCs from the
    past if we collectively don&#39;t want to be.<br>
    <br>
    I have not noticed a need for the above &quot;rule #1&quot; in this dis=
cussion
    yet, but I&#39;m just putting it out there, so if anyone needs to invok=
e
    it later, they can&#39;t be accused of inventing a rule for their own
    convenience.<br>
    <br>
    <h3>#2: There is no TCP-compatibility requirement </h3>
    Yes, CUBIC is not TCP-compatible. But that&#39;s a feature, not a bug.
    Words like &#39;TCP-compatible&#39; or &#39;fair&#39; sound nice, but d=
o not be
    fooled into thinking that an RFC recommends or requires you to
    behave in a certain way just because it has used a nice-sounding
    word for it. <br>
    <br>
    For an RFC to recommend or require something, it has to say SHOULD
    or MUST. So when BCP 41 [RFC2914] defines &#39;TCP-compatible&#39;, and
    talks about what is necessary to be &#39;fair&#39;, it implies nothing =
about
    whether you should or must be TCP-compatible, or fair; unless it
    actually says you SHOULD or MUST be TCP-compatible, or fair. <b>No
      RFC says that.</b> <br>
    <br>
    BCP 133 [RFC5033] and others have written careful words about
    straying from equal flow rates (quoted below). That gives a
    checklist for people like us who come along later. But it also
    respects our right to make a decision on whether/how to publish a CC
    RFC based on our own present-day knowledge and experience. To quote
    the relevant guideline in BCP 133 [RFC5033]:<br>
    <pre>   (1) Impact on Standard TCP, SCTP [<a href=3D"https://www.rfc-ed=
itor.org/rfc/rfc2960" title=3D"&quot;Stream Control Transmission Protocol&q=
uot;" target=3D"_blank">RFC2960</a>], and DCCP [<a href=3D"https://www.rfc-=
editor.org/rfc/rfc4340" title=3D"&quot;Datagram Congestion Control Protocol=
 (DCCP)&quot;" target=3D"_blank">RFC4340</a>].

       Proposed congestion control mechanisms should be evaluated when
       competing with standard IETF congestion control [RFC2581,
       <a href=3D"https://www.rfc-editor.org/rfc/rfc2960" target=3D"_blank"=
>RFC2960</a>, <a href=3D"https://www.rfc-editor.org/rfc/rfc4340" target=3D"=
_blank">RFC4340</a>].  Alternate congestion controllers that have a
       significantly negative impact on traffic using standard
       congestion control may be suspect and this aspect should be part
       of the community&#39;s decision making with regards to the
       suitability of the alternate congestion control mechanism.

       We note that this bullet is not a requirement for strict TCP-
       friendliness as a prerequisite for an alternate congestion
       control mechanism to advance to Experimental.  As an example,
       HighSpeed TCP is a congestion control mechanism that is
       Experimental, but that is not TCP-friendly in all environments.
       We also note that this guideline does not constrain the fairness
       offered for non-best-effort traffic.

       As an example from an Experimental RFC, fairness with standard
       TCP is discussed in Sections <a href=3D"https://www.rfc-editor.org/r=
fc/rfc5033.html#section-4" target=3D"_blank">4</a> and <a href=3D"https://w=
ww.rfc-editor.org/rfc/rfc5033.html#section-6" target=3D"_blank">6</a> of [<=
a href=3D"https://www.rfc-editor.org/rfc/rfc3649" title=3D"&quot;HighSpeed =
TCP for Large Congestion Windows&quot;" target=3D"_blank">RFC3649</a>] (Hig=
hSpeed TCP)
       and <b>using spare capacity</b> is discussed in Sections <a href=3D"=
https://www.rfc-editor.org/rfc/rfc5033.html#section-6" target=3D"_blank">6<=
/a>, <a href=3D"https://www.rfc-editor.org/rfc/rfc5033.html#section-11.1" t=
arget=3D"_blank">11.1</a>, and <a href=3D"https://www.rfc-editor.org/rfc/rf=
c5033.html#section-12" target=3D"_blank">12</a>
       of [<a href=3D"https://www.rfc-editor.org/rfc/rfc3649" title=3D"&quo=
t;HighSpeed TCP for Large Congestion Windows&quot;" target=3D"_blank">RFC36=
49</a>].</pre>
    The rfc8312bis draft has a section on &#39;using spare capacity&#39;
    precisely because that&#39;s what was considered acceptable about CUBIC
    and the new generation of hi-BDP CCs developed at that time, as I&#39;v=
e
    emphasized in the last para above. <br>
    <br>
    <h3>#3: A process for publishing EXP RFCs doesn&#39;t cover publishing
      non-EXP RFCs (by definition) <br>
    </h3>
    This IESG statement on &quot;<a href=3D"https://www.ietf.org/about/grou=
ps/iesg/statements/experimental-congestion-control/" target=3D"_blank">Expe=
rimental
      Specification of New Congestion Control Algorithms</a>&quot; says it
    defines the process for a CC to be published as an Experimental RFC.
    It does not say it is the process that /all/ CCs have to follow to
    enter the standards track. <br>
    <br>
    Before invoking that process, one has to determine whether it is
    applicable to this case. The IESG statement contains a sentence
    about how to determine its applicability:<br>
    <blockquote><font face=3D"monospace">The decision of whether a
        specific new congestion control schemes goes beyond the
        principles of [RFC2914] and hence will undergo the process
        described in this document lies with the transport area
        directors, who will use the expertise of the transport area
        directorate and other congestion control experts to aid their
        decision.</font><br>
    </blockquote>
    Since RFC2914, we&#39;ve had RFC3246 explaining why Reno is unscalable,
    and RFC5033 (quoted above) giving guidelines for designers of new
    CCs. So we can debate whether the process that Markku quoted is
    relevant to this case, to help the AD to decide whether to follow
    the process or not.<br>
    <br>
    <h3>#4 EXP or STD? My view <br>
    </h3>
    Reno is worse than other CCs in some respects. Nearly everyone has
    abandoned it. So, if someone asked you why Reno is the IETF&#39;s only
    stds track CC, what would you say ...?<br>
    <br>
    As I already said, we have to stop thinking that the stds track
    requires a CC algorithm to be
    perfect. A CC certainly mustn&#39;t have failure or collapse modes, but
    it&#39;s performance doesn&#39;t have to be perfect in all cases. For
    instance, CUBIC has known poor utilization over radio links. That
    didn&#39;t stop Reno from being on the stds track. For me, the IETF
    needs a new CC on the stds track, and CUBIC is the most natural
    candidate. <br>
    <br>
    However, we must not set a precedent for any old CC to move to the
    stds track. So we must document it&#39;s weaknesses and failings, and
    explain why they are not serious (if that is the case). <br>
    <br>
    New CCs need to interoperate with the most prevalent CC on the
    Internet (CUBIC). The IETF cannot even say that as long as it hasn&#39;=
t
    formally recognized CUBIC on the stds track. That is a failing of
    *us* in the tcpm WG. We can move CUBIC to stds track via the exp
    track if we have to, or we can go straight to stds track. The only
    difference is that the first option makes us look like jobsworths
    for longer. <br>
    <br>
    <div>
      <div style=3D"margin-bottom:0px;line-height:normal">
        <div><span>jobsworth</span></div>
      </div>
      <div><span>/<span>=CB=88d=CA=92=C9=92bzw=C9=99=CB=90=CE=B8</span>/</s=
pan></div>
    </div>
    <div>
      <div style=3D"display:inline">
        <div><i><span>noun</span></i>
          <div style=3D"display:inline-block"><span><span>=C2=A0derogatory<=
/span><span>=E2=80=A2</span><span>informal</span></span></div>
        </div>
        <span>an official who upholds petty rules even at the expense of
          humanity or common sense.</span></div>
    </div>
    <span><span>1970s: from &quot;I can&#39;t do that, it&#39;s more than m=
y job&#39;s
        worth.&quot;<br>
        <br>
      </span></span><br>
    Bob<br>
    <br>
    <pre cols=3D"72">--=20
________________________________________________________________
Bob Briscoe                               <a href=3D"http://bobbriscoe.net/=
" target=3D"_blank">http://bobbriscoe.net/</a></pre>
  </div>

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

--00000000000048a4cf05d8abedc5--


From nobody Wed Feb 23 07:17:51 2022
Return-Path: <touch@strayalpha.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 681253A1093; Wed, 23 Feb 2022 07:17:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.318
X-Spam-Level: 
X-Spam-Status: No, score=-6.318 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_HELO_NONE=0.001, SPF_NEUTRAL=0.779, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=strayalpha.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Ka3Hr6vbJrCu; Wed, 23 Feb 2022 07:17:43 -0800 (PST)
Received: from server217-1.web-hosting.com (server217-1.web-hosting.com [198.54.114.226]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 065503A117F; Wed, 23 Feb 2022 07:17:22 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=strayalpha.com; s=default; h=To:References:Message-Id:Cc:Date:In-Reply-To: From:Subject:Mime-Version:Content-Type:Sender:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id: List-Help:List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=0+6mfbPP2Xz/uO84BnmcNwACIgU0J2e6Il1Cl9e8ZNE=; b=cTDOr/bNYdSHo/orAZc0yPepTD LUjmIiTlm5/4/Zb2TYbo3Z89n+Ir4csQzomB4MwBQIZHNwJ+ftxJZLnK569ODwUA54P/4LbTEfD/y dVA/+/8siw123XqMYhr0OAhCGC4Yld3bVyEiSmsbbRwvw4hft572hKry4XVXYe3zNgElHV0bRGid4 ryIUcdCLpgkCDbK0vCpgWb8VW4NK9A5QZj6HQ2w/Z56EtRspvuIKBkbBvCHr0jPOQpxyzJ9xXb5hD hEfZ8dw53vx0dbejGvZEq/sQvMLkLPHZLHXXWxdUIYm7ZxRcRMI3tmbFBwkEIHZsVNY2jG9VNlgVQ R/GSLPWg==;
Received: from cpe-172-114-237-88.socal.res.rr.com ([172.114.237.88]:49159 helo=smtpclient.apple) by server217.web-hosting.com with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from <touch@strayalpha.com>) id 1nMtOE-00E3I9-47; Wed, 23 Feb 2022 10:17:19 -0500
Content-Type: multipart/alternative; boundary="Apple-Mail=_50612ECD-B1AC-4D72-87C1-D7990C238E12"
Mime-Version: 1.0 (Mac OS X Mail 15.0 \(3693.60.0.1.1\))
From: "touch@strayalpha.com" <touch@strayalpha.com>
In-Reply-To: <e8d7113c-47fa-7f91-54db-f7571840cd85@bobbriscoe.net>
Date: Wed, 23 Feb 2022 07:17:07 -0800
Cc: Yoshifumi Nishida <nsd.ietf@gmail.com>, Markku Kojo <kojo@cs.helsinki.fi>,  "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
Message-Id: <3D137052-588B-4479-93EC-B3F7CA502D1C@strayalpha.com>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi> <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com> <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org> <e8d7113c-47fa-7f91-54db-f7571840cd85@bobbriscoe.net>
To: Bob Briscoe <in@bobbriscoe.net>
X-Mailer: Apple Mail (2.3693.60.0.1.1)
X-OutGoing-Spam-Status: No, score=-1.0
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - server217.web-hosting.com
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - strayalpha.com
X-Get-Message-Sender-Via: server217.web-hosting.com: authenticated_id: touch@strayalpha.com
X-Authenticated-Sender: server217.web-hosting.com: touch@strayalpha.com
X-Source: 
X-Source-Args: 
X-Source-Dir: 
X-From-Rewrite: unmodified, already matched
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/vmYjUMLaLM0-EoHN41_PMZo0Enk>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Feb 2022 15:17:49 -0000

--Apple-Mail=_50612ECD-B1AC-4D72-87C1-D7990C238E12
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

Hi, Bob (et al.),

> On Feb 21, 2022, at 4:11 PM, Bob Briscoe <in@bobbriscoe.net> wrote:
>=20
> Yoshi, Markku,
>=20
> Many of the technical points Markku raised were important omissions =
from the rfc8312bis draft. Whatever status we give the eventual RFC,     =
they should be highlighted as areas that need further work, either known =
weaknesses or just areas where we don't have enough data to know whether =
they are weaknesses. I tried to help find useful references for =
experiments already done around those points, and I tried to help write =
such text.
>=20
> However, I cannot agree with some of Markku's statements about =
process.=20
>=20
> #1: The rules can change
>=20
> ...
> #2: There is no TCP-compatibility requirement
>=20
> Yes, CUBIC is not TCP-compatible. But that's a feature, not a bug. =
Words like 'TCP-compatible' or 'fair' sound nice, but do not be fooled =
into thinking that an RFC recommends or requires you to behave in a =
certain way just because it has used a nice-sounding word for it.=20
>=20
> For an RFC to recommend or require something, it has to say SHOULD or =
MUST. So when BCP 41 [RFC2914] defines 'TCP-compatible', and talks about =
what is necessary to be 'fair', it implies nothing about whether you =
should or must be TCP-compatible, or fair; unless it actually says you =
SHOULD or MUST be TCP-compatible, or fair. No RFC says that.=20

RFC 5681 "TCP Congestion Control=E2=80=9D (standards-track) says =
(emphasis mine below):

3 <https://datatracker.ietf.org/doc/html/rfc5681#section-3>.  Congestion =
Control Algorithms

   This section defines the four congestion control algorithms: slow
   start, congestion avoidance, fast retransmit, and fast recovery,
   developed in [Jac88 =
<https://datatracker.ietf.org/doc/html/rfc5681#ref-Jac88>] and [Jac90 =
<https://datatracker.ietf.org/doc/html/rfc5681#ref-Jac90>].  In some =
situations, it may be
   beneficial for a TCP sender to be more conservative than the
   algorithms allow; however, a TCP MUST NOT be more aggressive than the
   following algorithms allow (that is, MUST NOT send data when the
   value of cwnd computed by the following algorithms would not allow
   the data to be sent).

That looks like a MUST that defines fairness to me.

Joe


--Apple-Mail=_50612ECD-B1AC-4D72-87C1-D7990C238E12
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=utf-8

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3Dutf-8"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: after-white-space;" class=3D"">Hi, =
Bob (et al.),
<div><br class=3D""><blockquote type=3D"cite" class=3D""><div =
class=3D"">On Feb 21, 2022, at 4:11 PM, Bob Briscoe &lt;<a =
href=3D"mailto:in@bobbriscoe.net" class=3D"">in@bobbriscoe.net</a>&gt; =
wrote:</div><br class=3D"Apple-interchange-newline"><div class=3D"">
 =20
    <meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3DUTF-8" class=3D"">
 =20
  <div class=3D"">
    Yoshi, Markku,<br class=3D"">
    <br class=3D"">
    Many of the technical points Markku raised were important omissions
    from the rfc8312bis draft. Whatever status we give the eventual RFC,
    they should be highlighted as areas that need further work, either
    known weaknesses or just areas where we don't have enough data to
    know whether they are weaknesses. I tried to help find useful
    references for experiments already done around those points, and I
    tried to help write such text.<br class=3D"">
    <br class=3D"">
    However, I cannot agree with some of Markku's statements about
    process. <br class=3D"">
    <br class=3D"">
    <h3 class=3D"">#1: The rules can change </h3>...<br class=3D"">
    <h3 class=3D"">#2: There is no TCP-compatibility requirement </h3>
    Yes, CUBIC is not TCP-compatible. But that's a feature, not a bug.
    Words like 'TCP-compatible' or 'fair' sound nice, but do not be
    fooled into thinking that an RFC recommends or requires you to
    behave in a certain way just because it has used a nice-sounding
    word for it. <br class=3D"">
    <br class=3D"">
    For an RFC to recommend or require something, it has to say SHOULD
    or MUST. So when BCP 41 [RFC2914] defines 'TCP-compatible', and
    talks about what is necessary to be 'fair', it implies nothing about
    whether you should or must be TCP-compatible, or fair; unless it
    actually says you SHOULD or MUST be TCP-compatible, or fair. <b =
class=3D"">No
      RFC says that.</b> <br class=3D""></div></div></blockquote><div><br =
class=3D""></div><div>RFC&nbsp;5681 "<span style=3D"font-size: 1em;" =
class=3D""><b class=3D"">TCP Congestion Control</b></span><b =
class=3D"">=E2=80=9D</b><span style=3D"font-size: 1em;" class=3D""><b =
class=3D"">&nbsp;</b></span>(standards-track) says (emphasis mine =
below):</div><div><br class=3D""></div><div><pre class=3D"newpage" =
style=3D"font-size: 13.333333015441895px; margin-top: 0px; =
margin-bottom: 0px; break-before: page;"><span class=3D"h2" =
style=3D"display: inline; font-size: 1em; font-weight: bold;"><a =
class=3D"selflink" id=3D"section-3" =
href=3D"https://datatracker.ietf.org/doc/html/rfc5681#section-3">3</a>.  =
Congestion Control Algorithms</span>

   This section defines the four congestion control algorithms: slow
   start, congestion avoidance, fast retransmit, and fast recovery,
   developed in [<a =
href=3D"https://datatracker.ietf.org/doc/html/rfc5681#ref-Jac88" =
title=3D"&quot;Congestion Avoidance and Control&quot;" =
class=3D"">Jac88</a>] and [<a =
href=3D"https://datatracker.ietf.org/doc/html/rfc5681#ref-Jac90" =
title=3D"&quot;Modified TCP Congestion Avoidance Algorithm&quot;" =
class=3D"">Jac90</a>].  In some situations, it may be
   beneficial for a TCP sender to be more conservative than the
   algorithms allow; however, a T<b class=3D"">CP MUST NOT be more =
aggressive than the
   following algorithms allow (that is, MUST NOT send data when the
   value of cwnd computed by the following algorithms would not allow
   the data to be sent)</b>.</pre><div class=3D""><br =
class=3D""></div><div class=3D"">That looks like a MUST that defines =
fairness to me.</div><div class=3D""><br class=3D""></div><div =
class=3D"">Joe</div><div class=3D""><br =
class=3D""></div></div></div></body></html>=

--Apple-Mail=_50612ECD-B1AC-4D72-87C1-D7990C238E12--


From nobody Wed Feb 23 10:44:39 2022
Return-Path: <ietf@bobbriscoe.net>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8EA2B3A117D; Wed, 23 Feb 2022 10:44:36 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.813
X-Spam-Level: 
X-Spam-Status: No, score=-2.813 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, NICE_REPLY_A=-0.714, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=bobbriscoe.net
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KjdTMkv9OWBK; Wed, 23 Feb 2022 10:44:31 -0800 (PST)
Received: from mail-ssdrsserver2.hostinginterface.eu (mail-ssdrsserver2.hostinginterface.eu [185.185.85.90]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 471483A11B9; Wed, 23 Feb 2022 10:44:29 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=bobbriscoe.net; s=default; h=In-Reply-To:From:References:Cc:To:Subject: MIME-Version:Date:Message-ID:Content-Type:Sender:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id: List-Help:List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=k+IsnwCoNaXOqiiuW4kLI5BvO/E9e7A4gaDIGcq/Mr8=; b=CGXqCbGglrqeMye4VY3AWHZ3w0 ImhIX/iUZ4E52FjbMe8zj+DUQlUlfA+u39K+/fmquR3lma1JBpjRmkxQpHRCN21er/012zVM+jcgz gRkjGZcg2hoDT9GSKVKDsDcCRUbYtGC1+3rezYoDzzrKVK5CDXBWLW5LCfCv8BGMuskiNmitQrnlq yVW5O9U6KdcmFZORDTmSOMeBpcjKtYIUpWtV/yP9KAMgmoPd6fy+RS8fig3k5LMEaWsfE38lHhm9R C7mrbiE8TdYrtnHIUK+l+1SOBbZhnE8f5Yf5+Xn1lpOA1S+uhgSSC0XWUZDJPalpzjrQ7TYOwZTWF RZjiY97Q==;
Received: from 67.153.238.178.in-addr.arpa ([178.238.153.67]:55606 helo=[192.168.1.11]) by ssdrsserver2.hostinginterface.eu with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 (Exim 4.94.2) (envelope-from <ietf@bobbriscoe.net>) id 1nMwcr-0001NM-Ks; Wed, 23 Feb 2022 18:44:28 +0000
Content-Type: multipart/alternative; boundary="------------h4IKHyF3ne24kFNzVgFQXRoX"
Message-ID: <4212deb0-b01e-7c96-35af-2a8414731684@bobbriscoe.net>
Date: Wed, 23 Feb 2022 18:44:23 +0000
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.5.0
Content-Language: en-GB
To: "touch@strayalpha.com" <touch@strayalpha.com>
Cc: Yoshifumi Nishida <nsd.ietf@gmail.com>, Markku Kojo <kojo@cs.helsinki.fi>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi> <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com> <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org> <e8d7113c-47fa-7f91-54db-f7571840cd85@bobbriscoe.net> <3D137052-588B-4479-93EC-B3F7CA502D1C@strayalpha.com>
From: Bob Briscoe <ietf@bobbriscoe.net>
In-Reply-To: <3D137052-588B-4479-93EC-B3F7CA502D1C@strayalpha.com>
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - ssdrsserver2.hostinginterface.eu
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - bobbriscoe.net
X-Get-Message-Sender-Via: ssdrsserver2.hostinginterface.eu: authenticated_id: in@bobbriscoe.net
X-Authenticated-Sender: ssdrsserver2.hostinginterface.eu: in@bobbriscoe.net
X-Source: 
X-Source-Args: 
X-Source-Dir: 
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/cm6zyv_R00NysMmgdUCCepPNXIE>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Feb 2022 18:44:37 -0000

This is a multi-part message in MIME format.
--------------h4IKHyF3ne24kFNzVgFQXRoX
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Joe,

On 23/02/2022 15:17, touch@strayalpha.com wrote:
> Hi, Bob (et al.),
>
>> On Feb 21, 2022, at 4:11 PM, Bob Briscoe <in@bobbriscoe.net> wrote:
>>
>> Yoshi, Markku,
>>
>> Many of the technical points Markku raised were important omissions 
>> from the rfc8312bis draft. Whatever status we give the eventual RFC, 
>> they should be highlighted as areas that need further work, either 
>> known weaknesses or just areas where we don't have enough data to 
>> know whether they are weaknesses. I tried to help find useful 
>> references for experiments already done around those points, and I 
>> tried to help write such text.
>>
>> However, I cannot agree with some of Markku's statements about process.
>>
>>
>>       #1: The rules can change
>>
>> ...
>>
>>
>>       #2: There is no TCP-compatibility requirement
>>
>> Yes, CUBIC is not TCP-compatible. But that's a feature, not a bug. 
>> Words like 'TCP-compatible' or 'fair' sound nice, but do not be 
>> fooled into thinking that an RFC recommends or requires you to behave 
>> in a certain way just because it has used a nice-sounding word for it.
>>
>> For an RFC to recommend or require something, it has to say SHOULD or 
>> MUST. So when BCP 41 [RFC2914] defines 'TCP-compatible', and talks 
>> about what is necessary to be 'fair', it implies nothing about 
>> whether you should or must be TCP-compatible, or fair; unless it 
>> actually says you SHOULD or MUST be TCP-compatible, or fair. *No RFC 
>> says that.*
>
> RFC 5681 "*TCP Congestion Control**”***(standards-track) says 
> (emphasis mine below):
>
> 3 <https://datatracker.ietf.org/doc/html/rfc5681#section-3>. 
> Congestion Control Algorithms
>
>     This section defines the four congestion control algorithms: slow
>     start, congestion avoidance, fast retransmit, and fast recovery,
>     developed in [Jac88  <https://datatracker.ietf.org/doc/html/rfc5681#ref-Jac88>] and [Jac90  <https://datatracker.ietf.org/doc/html/rfc5681#ref-Jac90>].  In some situations, it may be
>     beneficial for a TCP sender to be more conservative than the
>     algorithms allow; however, a T*CP MUST NOT be more aggressive than the following algorithms allow 
> (that is, MUST NOT send data when the value of cwnd computed by the 
> following algorithms would not allow the data to be sent)*.
>
> That looks like a MUST that defines fairness to me.

[BB] Nice point. I usually consider that requirements in an RFC mandate 
what implementers have to do to claim compliance with that RFC. But this 
one does indeed seem to be setting its reach much wider, to any 
implementation of the TCP protocol, not just implementations of this CC.

If that is how we are meant to read it, it contradicts BCP 133 [RFC5033] 
that I quoted earlier. So, I guess people will pick which one they prefer.

As far as implementation compliance is concerned, hardly any TCP 
implementation in use complies with this mandate any more. And even at 
the time it was published (2009), it already wasn't being followed by 
many/most Windows and Linux machines (Compound and Cubic were introduced 
in 2006). I notice this wording was added in RFC2581 (1999), so the 
non-compliance crept up on the IETF, and by republishing these words 
unchanged, I guess this looks like the IETF perpetuating a pious hope 
that TCPs will comply.

So, if we do make CUBIC PS, it will need to update this para in RFC5681.



Bob

>
> Joe
>

-- 
________________________________________________________________
Bob Briscoehttp://bobbriscoe.net/

--------------h4IKHyF3ne24kFNzVgFQXRoX
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    Joe,<br>
    <br>
    <div class="moz-cite-prefix">On 23/02/2022 15:17,
      <a class="moz-txt-link-abbreviated" href="mailto:touch@strayalpha.com">touch@strayalpha.com</a> wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:3D137052-588B-4479-93EC-B3F7CA502D1C@strayalpha.com">
      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
      Hi, Bob (et al.),
      <div><br class="">
        <blockquote type="cite" class="">
          <div class="">On Feb 21, 2022, at 4:11 PM, Bob Briscoe &lt;<a
              href="mailto:in@bobbriscoe.net"
              class="moz-txt-link-freetext" moz-do-not-send="true">in@bobbriscoe.net</a>&gt;
            wrote:</div>
          <br class="Apple-interchange-newline">
          <div class="">
            <meta http-equiv="Content-Type" content="text/html;
              charset=UTF-8" class="">
            <div class=""> Yoshi, Markku,<br class="">
              <br class="">
              Many of the technical points Markku raised were important
              omissions from the rfc8312bis draft. Whatever status we
              give the eventual RFC, they should be highlighted as areas
              that need further work, either known weaknesses or just
              areas where we don't have enough data to know whether they
              are weaknesses. I tried to help find useful references for
              experiments already done around those points, and I tried
              to help write such text.<br class="">
              <br class="">
              However, I cannot agree with some of Markku's statements
              about process. <br class="">
              <br class="">
              <h3 class="">#1: The rules can change </h3>
              ...<br class="">
              <h3 class="">#2: There is no TCP-compatibility requirement
              </h3>
              Yes, CUBIC is not TCP-compatible. But that's a feature,
              not a bug. Words like 'TCP-compatible' or 'fair' sound
              nice, but do not be fooled into thinking that an RFC
              recommends or requires you to behave in a certain way just
              because it has used a nice-sounding word for it. <br
                class="">
              <br class="">
              For an RFC to recommend or require something, it has to
              say SHOULD or MUST. So when BCP 41 [RFC2914] defines
              'TCP-compatible', and talks about what is necessary to be
              'fair', it implies nothing about whether you should or
              must be TCP-compatible, or fair; unless it actually says
              you SHOULD or MUST be TCP-compatible, or fair. <b
                class="">No RFC says that.</b> <br class="">
            </div>
          </div>
        </blockquote>
        <div><br class="">
        </div>
        <div>RFC 5681 "<span style="font-size: 1em;" class=""><b
              class="">TCP Congestion Control</b></span><b class="">”</b><span
            style="font-size: 1em;" class=""><b class=""> </b></span>(standards-track)
          says (emphasis mine below):</div>
        <div><br class="">
        </div>
        <div>
          <pre class="newpage" style="font-size: 13.333333015441895px; margin-top: 0px; margin-bottom: 0px; break-before: page;"><span class="h2" style="display: inline; font-size: 1em; font-weight: bold;"><a class="selflink" id="section-3" href="https://datatracker.ietf.org/doc/html/rfc5681#section-3" moz-do-not-send="true">3</a>.  Congestion Control Algorithms</span>

   This section defines the four congestion control algorithms: slow
   start, congestion avoidance, fast retransmit, and fast recovery,
   developed in [<a href="https://datatracker.ietf.org/doc/html/rfc5681#ref-Jac88" title="&quot;Congestion Avoidance and Control&quot;" class="" moz-do-not-send="true">Jac88</a>] and [<a href="https://datatracker.ietf.org/doc/html/rfc5681#ref-Jac90" title="&quot;Modified TCP Congestion Avoidance Algorithm&quot;" class="" moz-do-not-send="true">Jac90</a>].  In some situations, it may be
   beneficial for a TCP sender to be more conservative than the
   algorithms allow; however, a T<b class="">CP MUST NOT be more aggressive than the
   following algorithms allow (that is, MUST NOT send data when the
   value of cwnd computed by the following algorithms would not allow
   the data to be sent)</b>.</pre>
          <div class=""><br class="">
          </div>
          <div class="">That looks like a MUST that defines fairness to
            me.</div>
        </div>
      </div>
    </blockquote>
    <br>
    [BB] Nice point. I usually consider that requirements in an RFC
    mandate what implementers have to do to claim compliance with that
    RFC. But this one does indeed seem to be setting its reach much
    wider, to any implementation of the TCP protocol, not just
    implementations of this CC. <br>
    <br>
    If that is how we are meant to read it, it contradicts BCP 133
    [RFC5033] that I quoted earlier. So, I guess people will pick which
    one they prefer.<br>
    <br>
    As far as implementation compliance is concerned, hardly any TCP
    implementation in use complies with this mandate any more. And even
    at the time it was published (2009), it already wasn't being
    followed by many/most Windows and Linux machines (Compound and Cubic
    were introduced in 2006). I notice this wording was added in RFC2581
    (1999), so the non-compliance crept up on the IETF, and by
    republishing these words unchanged, I guess this looks like the IETF
    perpetuating a pious hope that TCPs will comply.<br>
    <br>
    So, if we do make CUBIC PS, it will need to update this para in
    RFC5681.<br>
    <br>
    <br>
    <br>
    Bob<br>
    <br>
    <blockquote type="cite"
      cite="mid:3D137052-588B-4479-93EC-B3F7CA502D1C@strayalpha.com">
      <div>
        <div>
          <div class=""><br class="">
          </div>
          <div class="">Joe</div>
          <div class=""><br class="">
          </div>
        </div>
      </div>
    </blockquote>
    <br>
    <pre class="moz-signature" cols="72">-- 
________________________________________________________________
Bob Briscoe                               <a class="moz-txt-link-freetext" href="http://bobbriscoe.net/">http://bobbriscoe.net/</a></pre>
  </body>
</html>

--------------h4IKHyF3ne24kFNzVgFQXRoX--


From nobody Wed Feb 23 10:50:38 2022
Return-Path: <lars@eggert.org>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 683D43A11CB; Wed, 23 Feb 2022 10:50:36 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -7.099
X-Spam-Level: 
X-Spam-Status: No, score=-7.099 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_HI=-5, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=eggert.org
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 11f1skRFtnJV; Wed, 23 Feb 2022 10:50:31 -0800 (PST)
Received: from mail.eggert.org (mail.eggert.org [IPv6:2a00:ac00:4000:400:211:32ff:fe22:186f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9AD943A11C0; Wed, 23 Feb 2022 10:50:31 -0800 (PST)
Received: from smtpclient.apple (unknown [IPv6:2a00:ac00:4000:400:f45e:6fa:2c6d:5105]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.eggert.org (Postfix) with ESMTPSA id 50D951D24AB; Wed, 23 Feb 2022 20:50:19 +0200 (EET)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=eggert.org; s=dkim; t=1645642219; bh=2qlYR0ed/rSQgd9p8QQDxYmVdt15k6DsZ0ohL9zUoBE=; h=Subject:From:In-Reply-To:Date:Cc:References:To; b=nAhCTJGTji7RRbnBGcEw4x7luypnSGknlQVQWEgQ6QUNsCxIPGnUQowUQowf1Z23a FLb02yGacDsL198DMEyIFExAHh811nW0e1LjGyXangy2lREio34xvPqXzeq2pROHTY KrE7VqFCknbbrFPdep1f8z15G2cyw4OFDRDbmVoQ=
Content-Type: multipart/signed; boundary="Apple-Mail=_1E3B91AF-65D4-49D1-8AFF-A74FB961DD42"; protocol="application/pgp-signature"; micalg=pgp-sha512
Mime-Version: 1.0 (Mac OS X Mail 15.0 \(3693.60.0.1.1\))
From: Lars Eggert <lars@eggert.org>
In-Reply-To: <4212deb0-b01e-7c96-35af-2a8414731684@bobbriscoe.net>
Date: Wed, 23 Feb 2022 20:50:15 +0200
Cc: Joe Touch <touch@strayalpha.com>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, Markku Kojo <kojo@cs.helsinki.fi>, tcpm-chairs <tcpm-chairs@ietf.org>
Message-Id: <A46EAECD-0CA3-405B-A24B-60FECBCECC02@eggert.org>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi> <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com> <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org> <e8d7113c-47fa-7f91-54db-f7571840cd85@bobbriscoe.net> <3D137052-588B-4479-93EC-B3F7CA502D1C@strayalpha.com> <4212deb0-b01e-7c96-35af-2a8414731684@bobbriscoe.net>
To: Bob Briscoe <ietf@bobbriscoe.net>
X-MailScanner-ID: 50D951D24AB.A2863
X-MailScanner: Found to be clean
X-MailScanner-From: lars@eggert.org
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/myvnZJrRXFdJmOXE8uBryC5CRxI>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Feb 2022 18:50:37 -0000

--Apple-Mail=_1E3B91AF-65D4-49D1-8AFF-A74FB961DD42
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

Hi,

On 2022-2-23, at 20:44, Bob Briscoe <ietf@bobbriscoe.net> wrote:
> So, if we do make CUBIC PS, it will need to update this para in =
RFC5681.

it does so already - see the last paragraph in =
https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-rfc8312bis#section-1=


Thanks,
Lars


--Apple-Mail=_1E3B91AF-65D4-49D1-8AFF-A74FB961DD42
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
	filename=signature.asc
Content-Type: application/pgp-signature;
	name=signature.asc
Content-Description: Message signed with OpenPGP

-----BEGIN PGP SIGNATURE-----
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=4Kl6
-----END PGP SIGNATURE-----

--Apple-Mail=_1E3B91AF-65D4-49D1-8AFF-A74FB961DD42--


From nobody Wed Feb 23 10:52:07 2022
Return-Path: <lars@eggert.org>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AFACD3A11CD; Wed, 23 Feb 2022 10:52:05 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -7.099
X-Spam-Level: 
X-Spam-Status: No, score=-7.099 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_HI=-5, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=eggert.org
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id b5-bNuf9IvtN; Wed, 23 Feb 2022 10:52:00 -0800 (PST)
Received: from mail.eggert.org (mail.eggert.org [91.190.195.94]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 7B16D3A11CB; Wed, 23 Feb 2022 10:52:00 -0800 (PST)
Received: from smtpclient.apple (unknown [IPv6:2a00:ac00:4000:400:f45e:6fa:2c6d:5105]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.eggert.org (Postfix) with ESMTPSA id 9A6F91D24AB; Wed, 23 Feb 2022 20:51:51 +0200 (EET)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=eggert.org; s=dkim; t=1645642311; bh=lHIeFLvmHhirNL+1+L2M0rpkoE+mfnzRG3GblQqy6NQ=; h=Subject:From:In-Reply-To:Date:Cc:References:To; b=xinN0YHckaEWZIX48LmlQPsP2jU6Y4SQXZm7fViD1gnyQH0TKZr8keg+Yx/l2b96Q mSdMRFpd/D1p3KU07eX8H2lFkzXfjib9MThie/uY/XjozCasn2H6KZfClXbtpzcAQB nP08NX2KX2iK8Q/A/abc5BxMxFESfv2D5jrw0bxY=
Content-Type: multipart/signed; boundary="Apple-Mail=_18495599-78FF-472B-BF1A-CBFB8F7E97A8"; protocol="application/pgp-signature"; micalg=pgp-sha512
Mime-Version: 1.0 (Mac OS X Mail 15.0 \(3693.60.0.1.1\))
From: Lars Eggert <lars@eggert.org>
In-Reply-To: <A46EAECD-0CA3-405B-A24B-60FECBCECC02@eggert.org>
Date: Wed, 23 Feb 2022 20:51:50 +0200
Cc: Joe Touch <touch@strayalpha.com>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, Markku Kojo <kojo@cs.helsinki.fi>, tcpm-chairs <tcpm-chairs@ietf.org>
Message-Id: <C9ABCFE0-31ED-4766-8A57-F5B51208A6D1@eggert.org>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <alpine.DEB.2.21.2202120048000.4019@hp8x-60.cs.helsinki.fi> <CAAK044SUv2pjPSi_9jitNdtTHtGR-DVhiEn77yCf8M6B=bgKwQ@mail.gmail.com> <alpine.DEB.2.21.2202171538190.4019@hp8x-60.cs.helsinki.fi> <CAAK044TwWJq0PgVSeHU7LfEwacPuix8KHqrBGXB+TTrVaFh0-Q@mail.gmail.com> <alpine.DEB.2.21.2202181052240.4019@hp8x-60.cs.helsinki.fi> <CAAK044RR6HTHKOgaoh4hkssXugSHMc+dV_2Ru3xLPsLaYRR0aA@mail.gmail.com> <718D13E5-D813-455B-8541-2396EA1F5CA8@eggert.org> <e8d7113c-47fa-7f91-54db-f7571840cd85@bobbriscoe.net> <3D137052-588B-4479-93EC-B3F7CA502D1C@strayalpha.com> <4212deb0-b01e-7c96-35af-2a8414731684@bobbriscoe.net> <A46EAECD-0CA3-405B-A24B-60FECBCECC02@eggert.org>
To: Bob Briscoe <ietf@bobbriscoe.net>
X-MailScanner-ID: 9A6F91D24AB.A42BE
X-MailScanner: Found to be clean
X-MailScanner-From: lars@eggert.org
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/6kcpoC8UQ4QMT7xZbijYtiXL-A0>
Subject: Re: [tcpm] 2nd WGLC for draft-ietf-tcpm-rfc8312bis
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 23 Feb 2022 18:52:06 -0000

--Apple-Mail=_18495599-78FF-472B-BF1A-CBFB8F7E97A8
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

On 2022-2-23, at 20:50, Lars Eggert <lars@eggert.org> wrote:
> On 2022-2-23, at 20:44, Bob Briscoe <ietf@bobbriscoe.net> wrote:
>> So, if we do make CUBIC PS, it will need to update this para in =
RFC5681.
>=20
> it does so already - see the last paragraph in =
https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-rfc8312bis#section-1=


sorry, the last paragraph on page 4 at the above URL. (Section 1 is =
longer.)

Thanks,
Lars



--Apple-Mail=_18495599-78FF-472B-BF1A-CBFB8F7E97A8
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
	filename=signature.asc
Content-Type: application/pgp-signature;
	name=signature.asc
Content-Description: Message signed with OpenPGP

-----BEGIN PGP SIGNATURE-----
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=RG6P
-----END PGP SIGNATURE-----

--Apple-Mail=_18495599-78FF-472B-BF1A-CBFB8F7E97A8--


From nobody Fri Feb 25 05:49:33 2022
Return-Path: <noreply@ietf.org>
X-Original-To: tcpm@ietf.org
Delivered-To: tcpm@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 04B213A108F; Fri, 25 Feb 2022 05:49:12 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: Stewart Bryant via Datatracker <noreply@ietf.org>
To: <gen-art@ietf.org>
Cc: draft-ietf-tcpm-yang-tcp.all@ietf.org, last-call@ietf.org, tcpm@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 7.46.0
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <164579695194.18407.4718409421052888446@ietfa.amsl.com>
Reply-To: Stewart Bryant <stewart.bryant@gmail.com>
Date: Fri, 25 Feb 2022 05:49:12 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/l24lcF0TZqGH0BMfAVGPa9_pxFY>
Subject: [tcpm] Genart last call review of draft-ietf-tcpm-yang-tcp-06
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 25 Feb 2022 13:49:12 -0000

Reviewer: Stewart Bryant
Review result: Ready

I am the assigned Gen-ART reviewer for this draft. The General Area
Review Team (Gen-ART) reviews all IETF documents being processed
by the IESG for the IETF Chair.  Please treat these comments just
like any other last call comments.

For more information, please see the FAQ at

<https://trac.ietf.org/trac/gen/wiki/GenArtfaq>.

Document: draft-ietf-tcpm-yang-tcp-06
Reviewer: Stewart Bryant
Review Date: 2022-02-25
IETF LC End Date: 2022-03-02
IESG Telechat date: Not scheduled for a telechat

Summary: A well written document ready for publication as an RFC

Major issues: None

Minor issues: None

Nits/editorial comments: None




From nobody Fri Feb 25 13:01:31 2022
Return-Path: <kaduk@mit.edu>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3500F3A0900; Fri, 25 Feb 2022 13:00:59 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.498
X-Spam-Level: 
X-Spam-Status: No, score=-1.498 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, KHOP_HELO_FCRDNS=0.399, SPF_HELO_NONE=0.001, SPF_NONE=0.001, URIBL_BLOCKED=0.001] autolearn=no autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RVreCsWuun7Q; Fri, 25 Feb 2022 13:00:53 -0800 (PST)
Received: from outgoing.mit.edu (outgoing-auth-1.mit.edu [18.9.28.11]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 828F33A08C1; Fri, 25 Feb 2022 13:00:51 -0800 (PST)
Received: from kduck.mit.edu ([24.16.140.251]) (authenticated bits=56) (User authenticated as kaduk@ATHENA.MIT.EDU) by outgoing.mit.edu (8.14.7/8.12.4) with ESMTP id 21PL0Wgk000356 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Fri, 25 Feb 2022 16:00:39 -0500
Date: Fri, 25 Feb 2022 13:00:31 -0800
From: Benjamin Kaduk <kaduk@mit.edu>
To: Wesley Eddy <wes@mti-systems.com>
Cc: The IESG <iesg@ietf.org>, draft-ietf-tcpm-rfc793bis@ietf.org, tcpm-chairs@ietf.org, tcpm@ietf.org, Michael Scharf <michael.scharf@hs-esslingen.de>
Message-ID: <20220225210031.GV12881@kduck.mit.edu>
References: <163236803976.28405.5643771942452620510@ietfa.amsl.com> <6f6e7b90-081a-74b4-b329-8879addcb8c4@mti-systems.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <6f6e7b90-081a-74b4-b329-8879addcb8c4@mti-systems.com>
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/Bx3PjaryrmbVPzyhCkiLHpb7MNc>
Subject: Re: [tcpm] Benjamin Kaduk's Discuss on draft-ietf-tcpm-rfc793bis-25: (with DISCUSS and COMMENT)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 25 Feb 2022 21:01:00 -0000

Hi Wes,

I'm also sorry for the slow response -- your note arrived during a
particularly busy time, and since I was going to have to swap in a bunch of
state on the document to reply, I needed a bigger chunk of time in order to
produce something useful.

On Sun, Jan 09, 2022 at 11:02:12PM -0500, Wesley Eddy wrote:
> Hello, I'm looping back on this thread, since we need to close the loop 
> on the DISCUSS points, and address the other COMMENT contents.  Sorry 
> for the rather slow response, but I've finally been able to process 
> everything from your review:
> 
> 
> On 9/22/2021 11:34 PM, Benjamin Kaduk via Datatracker wrote:
> > Benjamin Kaduk has entered the following ballot position for
> > draft-ietf-tcpm-rfc793bis-25: Discuss
> >
> > ...
> > ----------------------------------------------------------------------
> > DISCUSS:
> > ----------------------------------------------------------------------
> >
> > Many thanks for taking on the task of producing a roll-up update for the
> > core TCP specification!  I am sure it was a lot of work, but I am happy
> > to see it done.
> >
> > That said, I do have a few points that I would like to have a bit more
> > discussion on before the document is published; I'm happy to see that
> > Warren already linked to
> > https://www.ietf.org/blog/handling-iesg-ballot-positions/  on the topic
> > of what a DISCUSS position can (and cannot) mean.
> >
> > (1) We incorporate some long-standing enhancements that improve the
> > security and robustness of TCP (in particular, random ISN and protection
> > against off-path in-window attacks come to mind), but only at SHOULD or
> > MAY requirements level.
> >
> > For example, we currently say:
> >
> >     A TCP implementation MUST use the above type of "clock" for clock-
> >     driven selection of initial sequence numbers (MUST-8), and SHOULD
> >     generate its Initial Sequence Numbers with the expression:
> >
> >     ISN = M + F(localip, localport, remoteip, remoteport, secretkey)
> >
> > and:
> >
> >           +  RFC 5961 [37] section 5 describes a potential blind data
> >              injection attack, and mitigation that implementations MAY
> >              choose to include (MAY-12).  TCP stacks that implement
> >              RFC 5961 MUST add an input check that the ACK value is
> >              [...]
> >
> > What prevents us from making a MUST-level requirement for randomized
> > ISNs?  Is it just the fact that it was only a SHOULD in RFC 6528 and a
> > perception that promoting to a MUST would be incompatible with retaining
> > Internet Standard status?
> >
> > Likewise, what prevents using stronger normative language (e.g., MUST)
> > for the RFC 5961 protections?
> >
> > It seems to me that these mechanisms are of general applicability and
> > provide significant value for use of TCP on the internet, even though
> > they are not fully robust and do not use cryptographic mechanisms.  If
> > there are scenarios where their use is harmful or even just not
> > applicable, that seems like an exceptional case that should get
> > documented so as to strengthen the general recommendation for the
> > non-exception cases.
> 
> On this question, I don't know if you (Ben) agreed, but the final 
> message I saw was from Joe:
> 
> https://mailarchive.ietf.org/arch/msg/tcpm/vDCi0xAf1Iayzsb4BS6DwlW3nhY/

That is the final message I see as well, where Joe included the statement
"Remember that one axiom of TCP is “think twice before changing TCP, then
don’t”, as a rule", which didn't really seem like a great sign that
continuing the conversation was likely to be productive, so I wanted to
wait and hear your thoughts as well before sending further replies.

> I think your question would be a good one to bring up in TCPM for future 
> work, but the working group was trying to avoid such changes in this 
> document.

I would like to see some stronger justification for pushing this change to
future work rather than incorporating it now.

In essence, I think that we require a fairly strong justification to
publish an Internet Standard in 2022 that says it's okay to adopt a data
model where a host has a global piece of state that it freely sends to
anyone who asks, where that piece of state can be used to attack/disrupt
all new connections that host makes, as opposed to just connections on the
5-tuple that asked.

The actual scope of utility of an ISN is local to an individual 5-tuple,
not global to a host, and false sharing of ISNs across connections adds
risk.  My stance as SEC AD is that the IETF should produce protocols that
are as secure as possible *subject to any constraints on them*.  Here, we
have the constraint of retaining compatibility with the existing deployment
base (which is a very compelling constraint!), so you do not see me
advocating for mandatory tcpcrypt (for example).  But I want to see some
explanation of what harm or risk there is in saying that the ISN is always
produced with the PRF of the 5-tuple (okay, 4-tuple since the protocol is
always TCP), to motivate a divergence from the more-secure behavior and
thereby justify retaining the behavior currently in the draft.  What
constraint are we subject to that prevents doing the right (security-wise)
thing?



> 
> >
> > (2) I think this is just a process question to ensure that the IESG
> > knows what we are approving at Internet Standard maturity, though it
> > is certainly possible that I misunderstand the situation.
> >
> > In Section 3.7.3 we see the normative statement (SHLD-6) that "when the
> > when the effective MTU of an interface varies packet-to- packet, TCP
> > implementations SHOULD use the smallest effective MTU of the interface
> > to calculate the value to advertise in the MSS option".  This seems to
> > originate in RFC 6691 (being obsoleted by this document), but RFC 6691
> > is only an Informational document and has not had an opportunity to
> > "accumulate experience at Proposed Standard before progressing", to
> > paraphrase RFC 6410.
> >
> > Similarly, Section 3.9.2 has (SHLD-23) "Generally, an application SHOULD
> > NOT change the DiffServ field value during the course of a connection
> > (SHLD-23)."  This is a bit harder to track down, as the DiffServ field
> > was not always known by that name.  I actually failed to find a directly
> > analogous previous statement of this guidance (presumably my error), and
> > thus don't know if it had any experience at the PS level or not.
> >
> > RFC 6410 seems pretty clear that some revisions are okay in Internet
> > Standards without such "bake time" at PS, but it does seem like
> > something that should be done consciously rather than by accident.
> 
> If I understood the thread earlier and results of the IESG telechat, 
> this point could be cleared?

Yes, this point can be considered resolved.

> 
> >
> > (3) This is also a process point for explicit consideration by the IESG.
> >
> > Appendix A.2 appears to discuss a few (rare) scenarios in which the
> > technical mechanisms of this document fail catastrophically (e.g.,
> > getting stuck in a SYN|ACK loop and failing to complete the handshake).
> > Does this meet the "resolved known design choices" and "no known
> > technical omission" bar required by RFC 2026 even for *proposed*
> > standard?
> >
> > (Note that RFC 2026 explicitly says that the IESG may waive this
> > requirement, at least for PS.)
> >
> >
> > (AFAICT one such scenario is reported at
> > https://www.rfc-editor.org/errata_search.php?eid=3305  , which the change
> > log for this document calls out as "not applicable due to other
> > changes"; I am not sure which "other changes" are intended, for this
> > case.)
> 
> If I understood the thread earlier and results of the IESG telechat, 
> this point could be cleared?

Yes, this point can be considered resolved.

> > (4) Another point mostly just to get explicit IESG acknowledgment
> > (elevating one of Lars' comments to DISCUSS level, essentially).
> >
> > As the changelog (and gen-art reviewer!) notes:
> >
> >     Early in the process of updating RFC 793, Scott Brim mentioned that
> >     this should include a PERPASS/privacy review.  This may be something
> >     for the chairs or AD to request during WGLC or IETF LC.
> >
> > I don't see any evidence to suggest that such a review actually
> > occurred.  Do we want to seek out such a targeted review before
> > progressing?
> 
> If I understood the thread earlier and results of the IESG telechat, 
> this point could be cleared?

Yes, this point can be considered resolved.

> > ----------------------------------------------------------------------
> > COMMENT:
> > ----------------------------------------------------------------------
> >
> > Thank you for the editorial changes so that we now talk about "a TCP
> > implementation" or a "remote TCP peer" rather than just "a TCP" or
> > "a remote TCP"!
> >
> > Abstract
> >
> >                                                   It also updates RFC
> >     5961 by adding a small clarification in reset handling while in the
> >     SYN-RECEIVED state.  [...]
> >
> > I'm not sure I found what this clarification was; is SYN-RECEIVED the
> > correct state?  The ad-hoc diff I constructed between RFC 793 and this
> > document shows identical text for the "If the RST bit is set" case when
> > currently in SYN-RECEIVED STATE.
> 
> The only trick is that RFC 5961 has an inconsistency in that its Section 
> 3.2 is first is talking about the change being applied in the 
> synchronized states, but SYN-RECEIVED is one of the states that it 
> applies to and is not technically a synchronized state. The 
> clarification is to just not use that term (synchronized state) in the 
> place where this logic is added.

Ah, that makes sense.  Thank you for checking and the explanation here.

> 
> >
> > Section 3.1
> >
> >     Options: [TCP Option]; Options#Size == (DOffset-5)*32; present
> >     only when DOffset > 5.
> >
> > My (later) nit-level notation comment aside, the given expression does
> > not seem to convey the size occupied by the options, but rather the
> > combined size of the options and the padding.
> 
> True; in the working copy I've added "note that this size expression 
> also includes any padding trailing the actual options present."  Does 
> that make sense for you?

Yes, that sounds good.

> 
> >
> > Section 3.2
> >
> >     A TCP Option is one of: an End of Option List Option, a No-Operation
> >     Option, or a Maximum Segment Size Option.
> >
> > The IANA registry lists some thirty-odd option kinds, so this sentence
> > just seems false without some additional qualifier ("defined by this
> > specification", etc.)
> 
> True, in the working copy, I changed it to "A TCP Option in the 
> mandatory option set".
> 
> 
> >
> > Section 3.4
> >
> >     In response to sending data the TCP endpoint will receive
> >     acknowledgments.  The following comparisons are needed to process the
> >     acknowledgments.
> >        [...]
> >        SEG.SEQ = first sequence number of a segment
> >
> >        SEG.LEN = the number of octets occupied by the data in the segment
> >        (counting SYN and FIN)
> >
> >        SEG.SEQ+SEG.LEN-1 = last sequence number of a segment
> >
> > It seems to me that this information from the incoming segment is not
> > part of processing the *acknowledgment*, but rather part of processing
> > the data received in that segment (a procedure discussed a few
> > paragraphs later).
> 
> I see what you mean, though these are used in validating any segments, 
> including pure acknowledgements.
> 
> 
> >
> >               This clock is a 32-bit counter that typically increments at
> >     least once every roughly 4 microseconds, [...]
> >     Maximum Segment Lifetime (MSL), generated ISNs will be unique, since
> >     it cycles approximately every 4.55 hours, which is much longer than
> >     the MSL.
> >
> > Once we put in the "at least" we allow arbitrarily faster clock updates,
> > and that puts the "approximately every 4.55 hours" estimate in question.
> > Very fast clock updates would cycle correspondingly faster.  Do we need
> > to place a lower limit on the clock update interval?  (On first look, it
> > seems like we might not, since the keyed PRF F() is providing most of
> > the protection from off-path guessing, and an attacker can always use
> > direct connections to estimate the clock cycle interval.  OTOH, if it
> > cycles so fast that it repeats within O(MSL), that might be problematic.)
> >
> >     parameters and some secret data.  For discussion of the selection of
> >     a specific hash algorithm and management of the secret key data,
> >     please see Section 3 of [41].
> >
> > The guidance in the referenced document seems a bit dated (it indicates
> > that MD5 is probably still okay for this purpose).  While the known
> > attacks on MD5 do not directly translate into an attack on ISN
> > generation, collisions can be found on as little as 64 bytes of input,
> > and all of the straightforward ways to use pure MD5 as a keyed hash for
> > this purpose have some undesirable properties.  I'm happy to note that
> > FreeBSD is using siphash for this purpose, which should be more than
> > adequate.  I expect that Linux and other major TCP stacks are already
> > doing something similar, so the guidance to use MD5 may be dated in
> > practice as well as in utility.
> >
> > I don't have a great proposal for where to put some updated guidance
> > (unless there's already some work underway in tcpm?); it is probably not
> > appropriate to put it inline here, so either an appendix or a separate
> > document seem plausible.
> 
> What you say makes sense.  Properly updating RFC 6528 is something we 
> can propose in TCPM based on your comments, which seems better to do in 
> a separate document.

Yes, that seems better placed in an update to 6528 than in this document.

> 
> >
> > Section 3.7.1
> >
> >     The MSS value to be sent in an MSS option should be equal to the
> >     effective MTU minus the fixed IP and TCP headers.  By ignoring both
> >     IP and TCP options when calculating the value for the MSS option, if
> >     there are any IP or TCP options to be sent in a packet, then the
> >     sender must decrease the size of the TCP data accordingly.  RFC 6691
> >     [42] discusses this in greater detail.
> >
> > I note that RFC 6691 is obsoleted by this document; it seems to me that
> > if we think there is useful content still in that document, we should
> > include such content in this document instead of referring to a document
> > we are calling obsolete.  (This is not the only place we do so, to be
> > clear, but I will try to mention it just once.  I do see the note that
> > we only claim to incorporate the normative portions of most of the
> > obsoleted specs, leaving the informational content alone.)
> 
> Understood, though referencing is very little work and allows someone 
> who is interested in more detail "why" to find it.  In contrast,  
> incorporating all this type of information into this document would be a 
> lot more work, and that content would not be needed or really even of 
> value for many readers.
> 
> 
> >
> > Section 3.8.4
> >
> >     An implementation SHOULD send a keep-alive segment with no data
> >     (SHLD-12); however, it MAY be configurable to send a keep-alive
> >     segment containing one garbage octet (MAY-6), for compatibility with
> >     erroneous TCP implementations.
> >
> > Such misbehaved TCP impelementations were misbehaved even in 1989 when
> > RFC 1122 was published -- do we have a sense for whether they are still
> > around to any significant degree?
> 
> That's a great question; I don't know, but would highly doubt they are 
> around anymore (at least not in production on the Internet).

Might be an interesting experiment, but it sounds like we don't have enough
data to justify changing the text here (yet).

> 
> >
> > Section 3.8.5
> >
> >     As a result of implementation differences and middlebox interactions,
> >     new applications SHOULD NOT employ the TCP urgent mechanism (SHLD-
> >     13).  However, TCP implementations MUST still include support for the
> >     urgent mechanism (MUST-30).  Details can be found in RFC 6093 [38].
> >
> > This "SHOULD NOT employ" has been in force for over a decade (RFC 6093
> > is dated January 2011).  How long do we have to wait until there are
> > sufficiently few implementations employing the urgent mechanism that it
> > no longer needs to be implemented?
> 
> When it was last looked at, there was at least one fairly prominent 
> application that relied on it, though they could do so semi-safely 
> because of being tied to an OS with the expected interpretation.
> 
> 
> >
> > Section 3.9.2.3
> >
> >        An incoming SYN with an invalid source address MUST be ignored
> >        either by TCP or by the IP layer (MUST-63) (Section 3.2.1.3 of
> >        [18]).
> >
> > Requirements of the form "A or B must do X" that are ambiguous about
> > whether A or B takes the action leave the risk that both will expect the
> > other party to take the action, and the action will fail to occur.  If
> > we're in a position to specifically require one (or both!) to check,
> > that leads to a more robust and verifiable system.  (I assume we're not
> > in such a position, but it can't hurt to check.)
> 
> Great point.  I guess we hope IP would discard it, but in case not, TCP 
> should.  If the implementation is handling both protocols at once (as in 
> some small hosts), there might not be much distinction.
> 
> 
> >
> > Section 4
> >
> >     Destination Address
> >             The network layer address of the remote endpoint.
> >     [...]
> >     Source Address
> >             The network layer address of the sending endpoint.
> >
> > These definitions don't seem to work in the context of a receiver
> > validating the TCP checksum, where the destination address is the local
> > endpoint's address and the source address is the remote endpoint's
> > address.  (I note that these definitions are different from what RFC 793
> > itself used.)
> 
> Interesting point on the destination (though the source one seems fine, 
> it's always the sending endpoint).  Maybe we should just change "remote" 
> to "receiving" in the destination one?

That's an easy fix; good idea.

> 
> >
> >     receive window
> >             This represents the sequence numbers the local (receiving)
> >             TCP endpoint is willing to receive.  Thus, the local TCP
> >             endpoint considers that segments overlapping the range
> >             RCV.NXT to RCV.NXT + RCV.WND - 1 carry acceptable data or
> >             control.  Segments containing sequence numbers entirely
> >             outside of this range are considered duplicates and
> >             discarded.
> >
> > Duplicates or injection attacks (when the sequence numbers in the
> > segment are too large).
> 
> ACK, I added "or injection attacks" in the working copy.
> 
> 
> >
> > Section 5
> >
> >     The collection of applicable RFC Errata that have been reported and
> >     either accepted or held for an update to RFC 793 were incorporated
> >     (Errata IDs: 573, 574, 700, 701, 1283, 1561, 1562, 1564, 1565, 1571,
> >     1572, 2296, 2297, 2298, 2748, 2749, 2934, 3213, 3300, 3301, 6222).
> >     Some errata were not applicable due to other changes (Errata IDs:
> >     572, 575, 1569, 3305, 3602).
> >
> > I think that EID 1565 belongs in the "not applicable due to other
> > changes" list, since the text it attempts to modify involves the
> > now-removed discussion of the IP "precedence" field.
> >
> > Similarly, EID 2296 also affected text about precedence and security
> > that is no longer present in a recognizable form.
> 
> Good points, I've moved those in the working copy.  Originally they were 
> fixed as described, but then other edits made them OBE.

I figured as much, yeah.

> 
> >
> >     The more secure Initial Sequence Number generation algorithm from RFC
> >     6528 was incorporated.  See RFC 6528 for discussion of the attacks
> >     that this mitigates, as well as advice on selecting PRF algorithms
> >     and managing secret key data.
> >
> > (As I mentioned up in §3.4, that guidance is no longer current.)
> 
> Is updating 6528 agreed to be possible future work?

Yes.  (Is this something where I should be expecting to take the action
item to write a draft, or do you think someone else might take up the pen?)

> 
> >
> > Section 9.1
> >
> > It's not clear to me that RFC 2675 ([5]) needs to be classified as
> > normative.
> 
> Yeah, I don't have any concern with making it informative, if you think 
> that's right?

>From looking at where we reference it (it's [25] now in the -27), I think
it looks informative.  I don't remember any other places where we talked
about jumbograms without a reference, which would be the only real concern.

> 
> >
> > The guidance at
> > https://www.ietf.org/about/groups/iesg/statements/normative-informative-references/
> > would suggest that RFC 5961 ([37]) should be classified as normative,
> > since we replicate its MUST-level requirements with the condition that
> > "TCP stacks that implement RFC 5961 MUST [...]", which would appear to
> > make that behavior an "optional feature".
> 
> Ok; I've made it so in the working copy.
> 
> 
> > Appendix A.1.2
> >
> >     The IP security option (IPSO) and compartment defined in [1] was
> >     refined in RFC 1038 that was later obsoleted by RFC 1108.  The
> >     Commercial IP Security Option (CIPSO) is defined in FIPS-188, and is
> >     supported by some vendors and operating systems.  RFC 1108 is now
> >
> > Should we mention that FIPS-188 is archived and withdrawn by NIST?
> > (I also didn't find much to define the actual IP option in the PDF I
> > found,
> > https://csrc.nist.gov/csrc/media/publications/fips/188/archive/1994-09-06/documents/fips188.pdf,
> > but I didn't look very hard.)
> 
> In the working copy, I've added "(withdrawn by NIST in 2015)" after the 
> mention of FIPS-188.
> 
> 
> > Appendix A.3
> >
> > It's fascinating to me that the preferred reference for this modified
> > Nagle algorithm is an Internet-Draft from 1999, vs something more
> > recent.
> 
> Me too!
> 
> 
> > Appendix B
> >
> >       Every 2nd full-sized segment or 2*RMSS ACK'd | SHLD-19|x| | | | |
> >
> > This 'x' seems to be in the "MUST" column, not the "SHOULD" column.
> 
> Thank you!  Fixed in working copy.
> 
> 
> >
> >     Time Stamp support                             | MAY-10 | | |x| | |
> >
> > How do we square timestamp support being a "MAY" with SHLD-4,
> > SHOULD-level guidance to use timestamps to reduce TIME-WAIT?
> 
> MAY-10 is the ***IP*** Timestamp option, and SHLD-4 is about the 
> ***TCP*** Timestamp option.

Oops, sorry for missing that distinction.

> 
> >
> >     Time Exceeded => tell ALP, don't abort         | MUST-56| | | | |x|
> >     Param Problem => tell ALP, don't abort         | MUST-56| | | | |x|
> >
> > Is there a double negative between "don't abort" and the 'x' being in
> > the "MUST NOT" column?
> 
> Agreed, that is confusing.  In the working copy, I'll change both of 
> those titles to "Abort on ..."
> 
> 
> > NITS
> >
> > I made essentially no attempt to de-duplicate the nit-level remarks
> > against the ballot positions from other ADs (in contrast to the other
> > comments, where I made some modest effort to de-duplicate).  My
> > apologies for the extra work to ignore the already-fixed items.
> >
> > Section 1
> >
> >     For several decades, RFC 793 plus a number of other documents have
> >     combined to serve as the core specification for TCP [48].  Over time,
> >     a number of errata have been filed against RFC 793, as well as
> >     deficiencies in security, performance, and many other aspects.  The
> >
> > A naive parse would say that this means a number of errata have been
> > filed against deficiencies.  I suspect the transition between errata and
> > deficiencies should refer to deficiencies having been "discovered" or
> > similar.
> 
> Good point.  In the working copy, I've broke that into 2 sentences, in 
> case you think this is more clear:
> 
>     Over time, a number of errata have been filed against RFC 793. There
>     have also been deficiencies found and resolved in security,
>     performance, and many other aspects.
> 
> 
> >     The purpose of this document is to bring together all of the IETF
> >     Standards Track changes that have been made to the base TCP
> >     functional specification and unify them into an update of RFC 793.
> >
> > It's a little surprising to see this described as an "update of RFC 793"
> > (vs. a "replacement of" or "updated version of") since the relationship
> > is Obsoletes, not Updates.  I might even consider "into a single
> > consolidated specification".
> 
> Ok, I changed to "updated version of" in the working draft.
> 
> 
> > Section 3.1
> >
> >     Options: [TCP Option]; Options#Size == (DOffset-5)*32; present
> >     only when DOffset > 5.
> >
> > The "Options#Size" notation seems confusing and is not using any
> > convention I'm aware of.  It does not appear in RFC 793 or any other RFC
> > that I can find, either.
> 
> This is changed now, as discussed in other threads.
> 
> 
> > Section 3.3.1
> >
> >     maintenance of a TCP connection requires the remembering of several
> >     variables.  We conceive of these variables being stored in a
> >
> > "the remembering of" is a fairly awkward phrase, where something like
> > just "remembering" or "maintaining state for" would flow more naturally.
> 
> Ok, changed to "maintaining state for" in the working copy.
> 
> 
> > Section 3.4
> >
> >     It is essential to remember that the actual sequence number space is
> >     finite, though very large.  This space ranges from 0 to 2**32 - 1.
> >
> > The sense of scale in the broader ecosystem may have evolved out from
> > under us; QUIC's 62-bit sequence space might be more along the lines of
> > "very large" these days, with a 32-bit space being merely "large".
> 
> Ok, removed "very" in the working copy.
> 
> 
> >
> >     A connection is defined by a pair of sockets.  Connections can be
> >
> > This is the first instance of the word "socket" in this document.
> > RFC 793 used the term much more prevalently, but this update has
> > (beneficially, IMO) moved away from that approach in favor of discussing
> > IP addresses and port numbers.  Might such a change be appropriate here
> > as well?  Regardless, we should probably have some introduction to what
> > we mean by "socket" if we are to retain any uses of the term, IMO, more
> > than just the glossary entry.
> 
> Good point.  I think it will help to borrow the sentences below from 793 
> and put them into the section 2.2 on "Key TCP Concepts":
> 
>      To allow for many processes within a single host to use TCP
>      communication facilities simultaneously, TCP provides a set of
>      addresses or ports within each host.  Concatenated with the network
>      and host addresses from the internet communication layer, this forms
>      a socket.

Thanks!

> >
> >     verify this SYN.  The three way handshake and the advantages of a
> >     clock-driven scheme are discussed in [68].
> >
> > I don't have access to the reference, but it's not clear from just it's
> > abstract whether "advantages of" or "advantages over" a clock-driven
> > scheme is the intended meaning.
> 
> We could clarify as "advantages of a clock-driven scheme for ISN 
> selection ...".  (I checked the paper and that's what it's referring to.)

Thanks for checking the reference.

> 
> >     explanation for this specification is given.  TCP implementors may
> >     violate the "quiet time" restriction, but only at the risk of causing
> >     some old data to be accepted as new or new data rejected as old
> >     duplicated by some receivers in the internet system.
> >
> > Maybe "old duplicated data"?  The current phrasing feels like it's
> > missing a word.
> 
> ACK, fixed in my working copy.
> 
> 
> >                                  Hosts that prefer to avoid waiting are
> >     willing to risk possible confusion of old and new packets at a given
> >     destination may choose not to wait for the "quiet time".
> >
> > I think this needs an "and", for "prefer to avoid waiting and are
> > willing to risk".
> 
> ACK, fixed in my working copy.
> 
> 
> >
> >     To summarize: every segment emitted occupies one or more sequence
> >     numbers in the sequence space, the numbers occupied by a segment are
> >     "busy" or "in use" until MSL seconds have passed, upon rebooting a
> >     block of space-time is occupied by the octets and SYN or FIN flags of
> >     the last emitted segment, if a new connection is started too soon and
> >     uses any of the sequence numbers in the space-time footprint of the
> >     last segment of the previous connection incarnation, there is a
> >     potential sequence number overlap area that could cause confusion at
> >     the receiver.
> >
> > This list seems to be missing an "and".
> > (Also, is it really only the last emitted segment that could cause
> > problems?)
> 
> ACK, fixed in my working copy.  You're right that it's not just the last 
> segment, but any that could potentially still be in-flight.
> 
> 
> >
> > Section 3.5
> >
> >                                                               It is the
> >     implementation of a trade-off between memory and messages to provide
> >     information for this checking.
> >
> > I'm not sure this reads well; is "the implementation of" needed?
> 
> Ok, removed in working copy.
> 
> 
> >
> >        If an incoming segment has a security level, or compartment that
> >        does not exactly match the level and compartment requested for the
> >        connection, a reset is sent and the connection goes to the CLOSED
> >        state.  The reset takes its sequence number from the ACK field of
> >
> > The comma in the first line is no longer needed (it was part of the list
> > when precedence was still part of the list).
> 
> ACK, fixed in working copy.
> 
> 
> > Section 3.6
> >
> >        In this case, a FIN segment can be constructed and placed on the
> >        outgoing segment queue.  No further SENDs from the user will be
> >        accepted by the TCP implementation, and it enters the FIN-WAIT-1
> >        state.  RECEIVEs are allowed in this state.  All segments
> >        preceding and including FIN will be retransmitted until
> >        acknowledged.  When the other TCP peer has both acknowledged the
> >        FIN and sent a FIN of its own, the first TCP peer can ACK this
> >        FIN.  Note that a TCP endpoint receiving a FIN will ACK but not
> >        send its own FIN until its user has CLOSED the connection also.
> >
> > Naming the two peers (e.g., A and B) can help avoid awkward grammatical
> > constructions like "can ACK this FIN" and improve clarity.
> 
> Unless you really think it's necessary, I prefer not to change this 
> original 793 text, since it doesn't seem so bad to me.
> 
> 
> > Section 3.8
> >
> >     segments may arrive due to network or TCP retransmission.  As
> >     discussed in the section on sequence numbers the TCP implementation
> >     performs certain tests on the sequence and acknowledgment numbers in
> >     the segments to verify their acceptability.
> >
> > comma after "sequence number".
> 
> ACK, fixed in working copy.
> 
> 
> >
> > Section 3.8.6.2.2
> >
> >     Note that the general effect of this algorithm is to advance RCV.WND
> >     in increments of Eff.snd.MSS (for realistic receive buffers:
> >     Eff.snd.MSS < RCV.BUFF/2).  Note also that the receiver must use its
> >     own Eff.snd.MSS, assuming it is the same as the sender's.
> >
> > I think the last sentence would be more clear if it was something like
> > "making the assumption that is the same" or "on the assumption that it
> > is the same".
> 
> ACK, changed to your suggested "making the assumption" in my working copy.
> 
> 
> > Section 3.8.6.3
> >
> >     Note that there are several current practices that further lead to a
> >     reduced number of ACKs, including generic receive offload (GRO), ACK
> >     compression, and ACK decimation [26].
> >
> > Reference [26] seems reasonable for ACK decimation and ACK compression,
> > but doesn't seem to cover GRO at all.
> 
> True; do you think we should try to find a proper GRO reference? I'm not 
> aware of there being a real standard or academic type of reference.

I wouldn't put much effort into looking.
It's a bit outside my normal field, so I don't have any immediate insight.
Google finds the DPDK docs, which might be stable and well-known, and
wikipedia's https://en.wikipedia.org/wiki/TCP_offload_engine entry has a
brief mention, with links to an LWN article and
https://books.google.com/books?id=C3wQBwAAQBAJ .  If we know someone with
a copy of the latter, it might be worth checking if it would be a useful
reference.

> 
> >
> > Section 3.9.1
> >
> >           If the PUSH flag is set, the application intends the data to be
> >           transmitted promptly to the receiver, and the PUSH bit will be
> >           set in the last TCP segment created from the buffer.  When an
> >           application issues a series of SEND calls without setting the
> >           PUSH flag, the TCP implementation MAY aggregate the data
> >           internally without sending it (MAY-16).
> >
> > There's a dedicated paragraph a few paragraphs later for when the PUSH
> > flag is not set; the last sentence might flow better there.
> 
> ACK.  Done in my working copy.
> 
> 
> >           Some TCP implementations have included a FLUSH call, which will
> >           empty the TCP send queue of any data that the user has issued
> >           SEND calls but is still to the right of the current send
> >           window.  That is, it flushes as much queued send data as
> >
> > I think "has issued SEND calls for" (add "for").
> 
> ACK, fixed in my working copy.
> 
> 
> > Section 3.9.2
> >
> >     When received options are passed up to TCP from the IP layer, TCP
> >     implementations MUST ignore options that it does not understand
> >     (MUST-50).
> >
> > singular/plural mismatch (it/implementations)
> 
> ACK, fixed in working copy.
> 
> 
> >
> > Section 3.9.2.2
> >
> >     Soft Errors
> >       For ICMP these include: Destination Unreachable -- codes 0, 1, 5,
> >       Time Exceeded -- codes 0, 1, and Parameter Problem.
> >
> >       For ICMPv6 these include: Destination Unreachable -- codes 0 and 3,
> >       Time Exceeded -- codes 0, 1, and Parameter Problem -- codes 0, 1,
> >       2.
> >
> >       Since these Unreachable messages indicate soft error conditions,
> >
> > I'm not entirely sure that I'd classify "parameter problem" as an
> > "unreachable" message per se.
> 
> I'm not sure I understand this comment.  This is the list from RFC 5461.

It looks like I was misparsing on first read, since there were commas
serving different roles.  ("Parameter Problem" is an element in the outer
list, matched with "Destination Unreachable" and "Time Exceeded", not one
of the codes for Time Exceeded.)  Using semicolons as delimiters for the
"outer" list, with commas for the lists of type codes, would help
distinguish them.

> 
> > Section 3.10
> >
> >     Please note in the following that all arithmetic on sequence numbers,
> >     acknowledgment numbers, windows, et cetera, is modulo 2**32 the size
> >     of the sequence number space.  Also note that "=<" means less than or
> >
> > Some punctuation around "the size of the sequence number space" seems in
> > order.
> 
> ACK, added parenthesis in working copy.
> 
> 
> >     equal to (modulo 2**32).
> >
> > [In formal mathematics this "less than or equal to, modulo N" operator
> > is not defined.  But it's probably okay in this context.]
> 
> Ok.
> 
> 
> > Section 3.10.1
> >
> >           the parameters of the incoming SYN segment.  Verify the
> >           security and DiffServ value requested are allowed for this
> >           user, if not return "error: precedence not allowed" or "error:
> >           security/compartment not allowed."  If passive enter the LISTEN
> >
> > It's surprising for the error string to mention "precedence" when the
> > predicate is DiffServ value.
> 
> This is a good point.  When the definition of those bits was changed, it 
> doesn't seem like the error message indicated here was correspondingly 
> changed.

Is this something we need to leave alone for backwards compatibility, or do
we have leeway to change the error message here?  (It looks like this text
is unchanged in the -27.)

> 
> >
> >           with "error: insufficient resources".  If Foreign socket was
> >           not specified, then return "error: remote socket unspecified".
> >
> > I suspect s/Foreign/remote/ was intended.  (Also occurs later, but I
> > will just note it once here.)
> 
> ACK, changed in my working copy.  Based on other comments, "foreign" in 
> general was changed to "remote".
> 
> 
> > Section 3.10.3
> >
> >        -  Since the remote side has already sent FIN, RECEIVEs must be
> >           satisfied by data already on hand, but not yet delivered to the
> >           user.  If no text is awaiting delivery, the RECEIVE will get a
> >           "error: connection closing" response.  Otherwise, any remaining
> >           text can be used to satisfy the RECEIVE.
> >
> > I think s/text/data/ should be applied on the last line (since it was
> > already applied on the second line).
> 
> ACK, changed in my working copy.
> 
> 
> >
> > Section 3.10.7.4
> >
> >     o  Segments are processed in sequence.  Initial tests on
> >        arrival are used to discard old duplicates, but further
> >        processing is done in SEG.SEQ order.  If a segment's
> >        contents straddle the boundary between old and new, only the
> >        new parts should be processed.
> >
> > Maybe s/should be/are/?  There's not really optionality about it...
> 
> ACK, changed in my working copy.
> 
> 
> >              *  If this connection was initiated with a passive OPEN
> >                 (i.e., came from the LISTEN state), then return this
> >                 connection to LISTEN state and return.  The user need
> >                 not be informed.  If this connection was initiated
> >                 with an active OPEN (i.e., came from SYN-SENT state)
> >                 then the connection was refused, signal the user
> >                 "connection refused".  In either case, all segments on
> >                 the retransmission queue should be removed.  And in
> >
> > IIUC, what's described here as "removed" is described elsewhere as
> > "flushed"; it would be good to use consistent terminology when possible.
> 
> Do you have a strong preference?

No strong preference, just a general desire to have consistent terminology
unless there is a difference in meaning intended.

> 
> >           +  Once in the ESTABLISHED state, it is possible to deliver
> >              segment text to user RECEIVE buffers.  Text from segments
> >              can be moved into buffers until either the buffer is full
> >              or the segment is empty.  If the segment empties and
> >              [...]
> >
> > As above, it seems like (case-insensitive) s/test/data/ would improve
> > consistency.
> 
> Changed to "data" in working copy.
> 
> 
> >
> > Section 4
> >
> >     internet datagram
> >             The unit of data exchanged between an internet module and the
> >             higher level protocol together with the internet header.
> >
> > "exchanged between an internet module and the higher level protocol"
> > sounds like a local operation; I would have expected the definition of
> > an *internet* datagram to involve transfer over the (inter)network.
> 
> Ok, can you suggest an alternative definition?  I agree this one from 
> the original 793 isn't terrific.

I'm somewhat reluctant to concoct a new definition from scratch at this
point in the process.  I see that RFC 1983 has "IP datagram" refer to just
"datagram", on which it says:

   datagram
      A self-contained, independent entity of data carrying sufficient
      information to be routed from the source to the destination
      computer without reliance on earlier exchanges between this source
      and destination computer and the transporting network.  See also:
      frame, packet.
      [Source: J. Postel]

If I was to overcome my reluctance and try to revise the existing text here
into something that makes more sense, it would be:

internet datagram
  A unit of data [from some higher layer protocol] exchanged between
  internet hosts, together with the internet header that allows the
  datagram to be routed from source to destination.

(I'm not sure whether I'd include the bit in square brackets.)
Feel free to use mine if you like, or go with Postel's (per RFC 1983) or
something else.

Many thanks again for all your tiresome work on this document.  The end is
in sight!

-Ben

> 
> >
> >     segment length
> >             The amount of sequence number space occupied by a segment,
> >             including any controls that occupy sequence space.
> >
> > Should we say that this is a field in the segment header?
> 
> I don't think so (it's not in the TCP header).
> 
> 
> >
> >     URG
> >             A control bit (urgent), occupying no sequence space, used to
> >             indicate that the receiving user should be notified to do
> >             urgent processing as long as there is data to be consumed
> >             with sequence numbers less than the value indicated in the
> >             urgent pointer.
> >
> > To me, "value indicated in" is synonymous with "value contained in",
> > which is problematic here since the urgent field is only 16 bits and
> > sequence numbers 32 bits.  "indicated by" would be an improvement,
> > though of course if we're willing to spend more words we can increase
> > clarity further.
> ACK, changed to "indicated by" in my working copy.
> 
> >
> > Appendix A.1
> >
> >     RFC 793 requires checking the IP security compartment and precedence
> >     on incoming TCP segments for consistency within a connection, and
> >
> > I think the past tense "required" would be more appropriate upon
> > publication of this document as an RFC obsoleting RFC 793.
> >
> ACK, fixed in my working copy.


From nobody Fri Feb 25 14:40:56 2022
Return-Path: <touch@strayalpha.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 07B253A0825; Fri, 25 Feb 2022 14:40:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.319
X-Spam-Level: 
X-Spam-Status: No, score=-1.319 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, SPF_HELO_NONE=0.001, SPF_NEUTRAL=0.779, URIBL_BLOCKED=0.001] autolearn=no autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=strayalpha.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 11ra1At_QYrJ; Fri, 25 Feb 2022 14:40:20 -0800 (PST)
Received: from server217-1.web-hosting.com (server217-1.web-hosting.com [198.54.114.226]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 988AD3A081F; Fri, 25 Feb 2022 14:40:20 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=strayalpha.com; s=default; h=To:References:Message-Id: Content-Transfer-Encoding:Cc:Date:In-Reply-To:From:Subject:Mime-Version: Content-Type:Sender:Reply-To:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id: List-Help:List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=ear0Tk8f9t0D8iQeGLZAhtERYxDpunIFWThnKKT/3jI=; b=ZG6sUGOwiQLeEb9wpKDslSV6Ee awVrZbiKOP62zr+pcNCmQWFqtXUBLcfy7dDCOTRKjGMQ82Z6VE3QRi9O7Wlp8AM8BBuyA0KrOGMdl t7itMzGq0uu86SxiAX/Ceo+T3EsKD7ElMjnTq3ZusJSoTtzsAmHsYGXG0B2cw4GDRXp+TaVcyfk/x OF+VxLUTN3vrX7I3GE9wcnsRZs/0+M8X4E0DOrVvsGMeUP8ve/c7OYGZivxJuo5cD56WqYyKxQXs9 7xCudGhqZ7hDz62RP5v/kEpvVJUBhqqOGQzeb5bvYVRE71OjixXrPzXhH6pvlFyzhFy2jxXehwdDy iFcSp0jQ==;
Received: from cpe-172-114-237-88.socal.res.rr.com ([172.114.237.88]:64334 helo=smtpclient.apple) by server217.web-hosting.com with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from <touch@strayalpha.com>) id 1nNjG3-008aq7-A2; Fri, 25 Feb 2022 17:40:19 -0500
Content-Type: text/plain; charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 15.0 \(3693.60.0.1.1\))
From: "touch@strayalpha.com" <touch@strayalpha.com>
In-Reply-To: <20220225210031.GV12881@kduck.mit.edu>
Date: Fri, 25 Feb 2022 14:40:14 -0800
Cc: Wes Eddy <wes@mti-systems.com>, draft-ietf-tcpm-rfc793bis@ietf.org, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, The IESG <iesg@ietf.org>, tcpm-chairs@ietf.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <5F0B3950-90A5-4C8E-9C2D-7C70B2108C1B@strayalpha.com>
References: <163236803976.28405.5643771942452620510@ietfa.amsl.com> <6f6e7b90-081a-74b4-b329-8879addcb8c4@mti-systems.com> <20220225210031.GV12881@kduck.mit.edu>
To: Benjamin Kaduk <kaduk@mit.edu>
X-Mailer: Apple Mail (2.3693.60.0.1.1)
X-OutGoing-Spam-Status: No, score=-0.2
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - server217.web-hosting.com
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - strayalpha.com
X-Get-Message-Sender-Via: server217.web-hosting.com: authenticated_id: touch@strayalpha.com
X-Authenticated-Sender: server217.web-hosting.com: touch@strayalpha.com
X-Source: 
X-Source-Args: 
X-Source-Dir: 
X-From-Rewrite: unmodified, already matched
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/of1_aPBE0MYPIgUGDg2ESbhMj28>
Subject: Re: [tcpm] Benjamin Kaduk's Discuss on draft-ietf-tcpm-rfc793bis-25: (with DISCUSS and COMMENT)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 25 Feb 2022 22:40:26 -0000

Hi, Ben,

Notes below=E2=80=A6

Joe

> On Feb 25, 2022, at 1:00 PM, Benjamin Kaduk <kaduk@mit.edu> wrote:
>=20
> Hi Wes,
>=20
> I'm also sorry for the slow response -- your note arrived during a
> particularly busy time, and since I was going to have to swap in a =
bunch of
> state on the document to reply, I needed a bigger chunk of time in =
order to
> produce something useful.
>=20
> On Sun, Jan 09, 2022 at 11:02:12PM -0500, Wesley Eddy wrote:
>> Hello, I'm looping back on this thread, since we need to close the =
loop=20
>> on the DISCUSS points, and address the other COMMENT contents.  Sorry=20=

>> for the rather slow response, but I've finally been able to process=20=

>> everything from your review:
>>=20
>>=20
>> On 9/22/2021 11:34 PM, Benjamin Kaduk via Datatracker wrote:
>>> Benjamin Kaduk has entered the following ballot position for
>>> draft-ietf-tcpm-rfc793bis-25: Discuss
>>>=20
>>> ...
>>> =
----------------------------------------------------------------------
>>> DISCUSS:
>>> =
----------------------------------------------------------------------
>>>=20
>>> Many thanks for taking on the task of producing a roll-up update for =
the
>>> core TCP specification!  I am sure it was a lot of work, but I am =
happy
>>> to see it done.
>>>=20
>>> That said, I do have a few points that I would like to have a bit =
more
>>> discussion on before the document is published; I'm happy to see =
that
>>> Warren already linked to
>>> https://www.ietf.org/blog/handling-iesg-ballot-positions/  on the =
topic
>>> of what a DISCUSS position can (and cannot) mean.
>>>=20
>>> (1) We incorporate some long-standing enhancements that improve the
>>> security and robustness of TCP (in particular, random ISN and =
protection
>>> against off-path in-window attacks come to mind), but only at SHOULD =
or
>>> MAY requirements level.
>>>=20
>>> For example, we currently say:
>>>=20
>>>    A TCP implementation MUST use the above type of "clock" for =
clock-
>>>    driven selection of initial sequence numbers (MUST-8), and SHOULD
>>>    generate its Initial Sequence Numbers with the expression:
>>>=20
>>>    ISN =3D M + F(localip, localport, remoteip, remoteport, =
secretkey)
>>>=20
>>> and:
>>>=20
>>>          +  RFC 5961 [37] section 5 describes a potential blind data
>>>             injection attack, and mitigation that implementations =
MAY
>>>             choose to include (MAY-12).  TCP stacks that implement
>>>             RFC 5961 MUST add an input check that the ACK value is
>>>             [...]
>>>=20
>>> What prevents us from making a MUST-level requirement for randomized
>>> ISNs?  Is it just the fact that it was only a SHOULD in RFC 6528 and =
a
>>> perception that promoting to a MUST would be incompatible with =
retaining
>>> Internet Standard status?
>>>=20
>>> Likewise, what prevents using stronger normative language (e.g., =
MUST)
>>> for the RFC 5961 protections?
>>>=20
>>> It seems to me that these mechanisms are of general applicability =
and
>>> provide significant value for use of TCP on the internet, even =
though
>>> they are not fully robust and do not use cryptographic mechanisms.  =
If
>>> there are scenarios where their use is harmful or even just not
>>> applicable, that seems like an exceptional case that should get
>>> documented so as to strengthen the general recommendation for the
>>> non-exception cases.
>>=20
>> On this question, I don't know if you (Ben) agreed, but the final=20
>> message I saw was from Joe:
>>=20
>> =
https://mailarchive.ietf.org/arch/msg/tcpm/vDCi0xAf1Iayzsb4BS6DwlW3nhY/
>=20
> That is the final message I see as well, where Joe included the =
statement
> "Remember that one axiom of TCP is =E2=80=9Cthink twice before =
changing TCP, then
> don=E2=80=99t=E2=80=9D, as a rule", which didn't really seem like a =
great sign that
> continuing the conversation was likely to be productive, so I wanted =
to
> wait and hear your thoughts as well before sending further replies.
>=20
>> I think your question would be a good one to bring up in TCPM for =
future=20
>> work, but the working group was trying to avoid such changes in this=20=

>> document.
>=20
> I would like to see some stronger justification for pushing this =
change to
> future work rather than incorporating it now.

The primary issue is IOT devices. Requiring a clock or strong =
randomization simply isn=E2=80=99t always warranted.

> In essence, I think that we require a fairly strong justification to
> publish an Internet Standard in 2022 that says it's okay to adopt a =
data
> model where a host has a global piece of state that it freely sends to
> anyone who asks, where that piece of state can be used to =
attack/disrupt
> all new connections that host makes, as opposed to just connections on =
the
> 5-tuple that asked.

That=E2=80=99s not quite what=E2=80=99s happening. Just because the ISN =
doesn=E2=80=99t use either a clock or strong randomization doesn=E2=80=99t=
 mean it can be guessed for new connections.

> The actual scope of utility of an ISN is local to an individual =
5-tuple,
> not global to a host, and false sharing of ISNs across connections =
adds
> risk. =20

You need to consider the overall risk, which is:

a) that you open a connection and learn an ISN
AND
b) that you can attack other connections to that host (i.e., as an =
off-path attacker), notably by knowing the entire 4-tuple

(On-path attackers can simply watch whatever ISN you PRF anyway).

The latter still needs to be =E2=80=9Cin window=E2=80=9D at least, so =
the vulnerability is already somewhat mitigated.

Finally, the vulnerability is one that can be mitigated by the end =
system, e.g., by using a clock. So at best, it shoots itself in the =
foot.

I see no good reason to raise the bar for IoT implementations simply to =
prevent them from doing something with such local impact.

> My stance as SEC AD is that the IETF should produce protocols that
> are as secure as possible *subject to any constraints on them*.

Let=E2=80=99s please be clear that this is NOT *security*. Security =
would be TCP-AO or IPsec (as noted in RFC6528, sec 4).

> Here, we
> have the constraint of retaining compatibility with the existing =
deployment
> base (which is a very compelling constraint!), so you do not see me
> advocating for mandatory tcpcrypt (for example).  But I want to see =
some
> explanation of what harm or risk there is in saying that the ISN is =
always
> produced with the PRF of the 5-tuple (okay, 4-tuple since the protocol =
is
> always TCP), to motivate a divergence from the more-secure behavior =
and
> thereby justify retaining the behavior currently in the draft.  What
> constraint are we subject to that prevents doing the right =
(security-wise)
> thing?

We already have RFC6528 that specifies this as a SHOULD.

My question to you is =E2=80=9Cwhat has changed since 2012 that makes =
you think this is now appropriate for a MUST=E2=80=9D.

AFAICT, that=E2=80=99s where the burden of proof is, not on us to defend =
keeping a SHOULD.

Joe



From nobody Fri Feb 25 17:33:07 2022
Return-Path: <agenda@ietf.org>
X-Original-To: tcpm@ietf.org
Delivered-To: tcpm@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 975423A0DEF; Fri, 25 Feb 2022 17:29:02 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: "\"IETF Secretariat\"" <agenda@ietf.org>
To: <nsd.ietf@gmail.com>, <tcpm-chairs@ietf.org>
Cc: martin.h.duke@gmail.com, tcpm@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 7.46.0
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <164583894259.24617.10302401566665139971@ietfa.amsl.com>
Date: Fri, 25 Feb 2022 17:29:02 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/jRB7rSTvMLqvKdifmTqICV3TF2k>
Subject: [tcpm] tcpm - Requested session has been scheduled for IETF 113
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 26 Feb 2022 01:29:03 -0000

Dear Yoshifumi Nishida,

The session(s) that you have requested have been scheduled.
Below is the scheduled session information followed by
the original request. 


    tcpm Session 1 (2:00 requested)
    Wednesday, 23 March 2022, Afternoon Session II 1430-1630
    Room Name: Grand Klimt Hall 2 size: 165
    ---------------------------------------------


iCalendar: https://datatracker.ietf.org/meeting/113/sessions/tcpm.ics

Request Information:


---------------------------------------------------------
Working Group Name: TCP Maintenance and Minor Extensions
Area Name: Transport Area
Session Requester: Yoshifumi Nishida


Number of Sessions: 1
Length of Session(s): 
Number of Attendees: 60
Conflicts to Avoid: 

       


People who must be present:
  Martin Duke
  Michael Scharf
  Michael Tuexen
  Yoshifumi Nishida

Resources Requested:

Special Requests:
  
---------------------------------------------------------



From nobody Sat Feb 26 23:45:53 2022
Return-Path: <do_not_reply@mnot.net>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B7FA23A0B77 for <tcpm@ietfa.amsl.com>; Sat, 26 Feb 2022 23:45:50 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.708
X-Spam-Level: 
X-Spam-Status: No, score=-1.708 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_INVALID=0.1, DKIM_SIGNED=0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001] autolearn=no autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=fail (2048-bit key) reason="fail (message has been altered)" header.d=mnot.net header.b=kzxH10HZ; dkim=fail (2048-bit key) reason="fail (message has been altered)" header.d=messagingengine.com header.b=C+L6dBMf
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id K1x-qtz0KGdj for <tcpm@ietfa.amsl.com>; Sat, 26 Feb 2022 23:45:46 -0800 (PST)
Received: from out3-smtp.messagingengine.com (out3-smtp.messagingengine.com [66.111.4.27]) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0F7BF3A0B8F for <tcpm@ietf.org>; Sat, 26 Feb 2022 23:45:45 -0800 (PST)
Received: from compute1.internal (compute1.nyi.internal [10.202.2.41]) by mailout.nyi.internal (Postfix) with ESMTP id 474455C012F for <tcpm@ietf.org>; Sun, 27 Feb 2022 02:38:48 -0500 (EST)
Received: from mailfrontend2 ([10.202.2.163]) by compute1.internal (MEProxy); Sun, 27 Feb 2022 02:38:48 -0500
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mnot.net; h=cc :content-type:date:from:from:in-reply-to:mime-version:reply-to :sender:subject:subject:to:to; s=fm2; bh=r74xxpAL6CWC0Ls2iEPJC6v uqjJ7h33F1DacNHzT908=; b=kzxH10HZlL2RKZn5mQA++V9Wb0aOQFcDGOdZ8tb FQWtOBvEMQedwamurj+E2UTK++ruRqmr0w3dRCyJZg+BiYd7wShMLnsP9UIS1u3+ qohD8BAIZSYOltXVIoHk5xqlMfXT3aI6KXRIRpofJCfwRL8LJl0c35TfSxnmO/ZT 5jyX/3y5qFMYAcGRavV9G68lX9zr0Ij8ncMrDU5ndGTMdYXkJrOvn6ugxmanunMl DIcqXTg7OrijcoxeOqqh0LrUaqadehzMtz5Tgz2AGxjwWaiJo37V8FpZsxT4IYm/ qCzIKcIVn+bbzfoG5GAfWLrJlYCtwKOzh/O4C+kaPx0eerQ==
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=cc:content-type:date:from:from :in-reply-to:mime-version:reply-to:sender:subject:subject:to:to :x-me-proxy:x-me-proxy:x-me-sender:x-me-sender:x-sasl-enc; s= fm2; bh=r74xxpAL6CWC0Ls2iEPJC6vuqjJ7h33F1DacNHzT908=; b=C+L6dBMf hr6LHCrw+ZpgtrO0EINK9VuzHve1uW/TCDa+RGIvcWWYNMm24DNVqxtR40U9Penj 5Fivhvn60CCe0bmwyd0dkPL3DzwyDNo7MNiVolIN6hIHYXjp5QW4Tlh1mInvbJt3 sxsiiNvuq0kV34GY6NxYKBJGK70EFJxwDEYqBVejOT/cmq589p7coUIpZs84VY3O 3VixXR/r0U1wRNLlBHeupWPf/aOuBku13CiWxD+kXe+BZ7lmgJIqM7VfQnDE8DPe nk14X7lb3lte1OOScYFMj1zP2zZA2h9rO4scAJaEJL3wO9bas0pTctHv52O/S20t 3mp0eo3hh50RnQ==
X-ME-Sender: <xms:iCobYo8EFjvtFXvDhGKx9c0zW_1Te7IZoRqVp2pgskAaDOmXBRy81A> <xme:iCobYguHQm5olfOZL77urj6XhMCnpNFv2Ja5c6iFXBhdK7kXLldiXMz3-AX3Rgvzy sfDJrKlHXrPk31S3A>
X-ME-Received: <xmr:iCobYuC5K7dtJKCY1uC7xV_knc4M5eyirG-qZObVrvPjOknLASEMwViw2f6OpVwxmEl5KetW_XbeXK7-zdP1pukNRCCD6IkvU20_ePWmWm_8-l5GeKMg9J5CNxnlHtJx>
X-ME-Proxy-Cause: gggruggvucftvghtrhhoucdtuddrgedvvddrleejgddutdehucetufdoteggodetrfdotf fvucfrrhhofhhilhgvmecuhfgrshhtofgrihhlpdfqfgfvpdfurfetoffkrfgpnffqhgen uceurghilhhouhhtmecufedttdenucfpohcuuggrthgvuchfihgvlhguucdlgeelmdenuc fjughrpegtggfhvffusegrtddtredttdejnecuhfhrohhmpeftvghpohhsihhtohhrhicu tegtthhivhhithihucfuuhhmmhgrrhihuceuohhtuceoughopghnohhtpghrvghplhihse hmnhhothdrnhgvtheqnecuggftrfgrthhtvghrnhepkeefvdduteejvdefkeehieevuefg fefhteetveegffekffefteffvdelheduieetnecuffhomhgrihhnpehgihhthhhusgdrtg homhenucevlhhushhtvghrufhiiigvpedtnecurfgrrhgrmhepmhgrihhlfhhrohhmpegu ohgpnhhothgprhgvphhlhiesmhhnohhtrdhnvght
X-ME-Proxy: <xmx:iCobYofQACa9SShXgHZrrIjpYDgYbeIoPnkd6hOtp5YzAhkkrr53nQ> <xmx:iCobYtPXaXTZ2x8xvTIHqZDNBnktIw26BsY9Bb4PJYP3j4Iyb28KKw> <xmx:iCobYikTYxpTqufXKz_quQQzswzS1nl7WIicFlYw3h-DXQmoD50rPg> <xmx:iCobYmbs-rmNsxhxf1GXYnAOBkO3HBaITOnhehFGdGWlFSY8guoBRQ>
Received: by mail.messagingengine.com (Postfix) with ESMTPA for <tcpm@ietf.org>; Sun, 27 Feb 2022 02:38:48 -0500 (EST)
Content-Type: multipart/alternative; boundary="===============8391552910773655416=="
MIME-Version: 1.0
From: Repository Activity Summary Bot <do_not_reply@mnot.net>
To: tcpm@ietf.org
Message-Id: <20220227074546.0F7BF3A0B8F@ietfa.amsl.com>
Date: Sat, 26 Feb 2022 23:45:45 -0800 (PST)
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/YRNPBt_6Y8OyIAJBigRrH-v0u_M>
Subject: [tcpm] Weekly github digest (RFC8312bis Activity Summary)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 27 Feb 2022 07:45:51 -0000

--===============8391552910773655416==
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="utf-8"; format="flowed"




Events=20

Issues
------
* NTAP/rfc8312bis (+0/-1/=F0=9F=92=AC0)
  1 issues closed:
  - Clarification about cwnd and flight size when referring to RFC 7661 htt=
ps://github.com/NTAP/rfc8312bis/issues/143=20



Pull requests
-------------
* NTAP/rfc8312bis (+0/-1/=F0=9F=92=AC0)
  1 pull requests merged:
  - add text to explain why 7661 is safe to use even when cwnd > rwnd
    https://github.com/NTAP/rfc8312bis/pull/144=20


Repositories tracked by this digest:
-----------------------------------
* https://github.com/NTAP/rfc8312bis

--===============8391552910773655416==
Content-Type: text/html; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable

<!doctype html>
<html lang=3D"en">
<head>
<meta charset=3D"utf-8">
<title>Weekly github digest (RFC8312bis Activity Summary)</title>
<style>
body { font-family: Gotham, "Helvetica Neue", Helvetica, Arial, sans-serif;=
 font-size: 14px; }
h2 { margin-top: 3em; color: #A52A2A; font-style: italic; font-weight: norm=
al; }
h3 { margin-bottom:0; margin-top: 2em; font-size: 1.2em; }
h1+h2 { margin-top: 1em; }
a { color: #bb6219; text-decoration: none; }
li { margin-bottom: .35em; }
.repos { margin-bottom: 0; margin-top:0; line-height: 1.2; }
.new { color: red; }
.label { display: inline;
	padding: .2em .6em .3em;
	font-size: 75%;
	font-weight: 700;
	line-height: 1;
	color: #fff;
	text-align: center;
	white-space: nowrap;
	vertical-align: baseline;
	border-radius: .25em;
}
details {
	margin-top: 8em;
	}
summary {
	margin-bottom: 1em;
	cursor: pointer;
}
</style>
</head>

<body>
<h1>Sunday February 27, 2022</h1>

<p>Events </p>

<h2>Issues</h2>

<h3>NTAP/rfc8312bis (+0/-1/=F0=9F=92=AC0)</h3>


  <p>1 issues closed:</p>
  <ul>
  <li>#143 <a href=3D"https://github.com/NTAP/rfc8312bis/issues/143">Clarif=
ication about cwnd and flight size when referring to RFC 7661</a> </li>
  </ul>



<h2>Pull requests</h2>
<h3>NTAP/rfc8312bis (+0/-1/=F0=9F=92=AC0)</h3>


  <p>1 pull requests merged:</p>
  <ul>
  <li>#144 <a href=3D"https://github.com/NTAP/rfc8312bis/pull/144">add text=
 to explain why 7661 is safe to use even when cwnd &gt; rwnd</a> </li>
  </ul>


  <details>
    <summary>Repositories tracked by this digest:</summary>
<ul class=3D"repos">
  <li><a href=3D"https://github.com/NTAP/rfc8312bis">https://github.com/NTA=
P/rfc8312bis</a></li>
</ul>
</details>
</body>
</html>

--===============8391552910773655416==--


From nobody Mon Feb 28 02:00:52 2022
Return-Path: <nsd.ietf@gmail.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 39B6C3A0DFE; Mon, 28 Feb 2022 02:00:50 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.106
X-Spam-Level: 
X-Spam-Status: No, score=-2.106 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8JFY1Bd95QJW; Mon, 28 Feb 2022 02:00:45 -0800 (PST)
Received: from mail-qv1-xf2b.google.com (mail-qv1-xf2b.google.com [IPv6:2607:f8b0:4864:20::f2b]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B709D3A0E01; Mon, 28 Feb 2022 02:00:45 -0800 (PST)
Received: by mail-qv1-xf2b.google.com with SMTP id x3so11842011qvd.8; Mon, 28 Feb 2022 02:00:45 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112;  h=mime-version:from:date:message-id:subject:to:cc; bh=8+KFyjrXcVTuDUOZNcC8AoShXQWVnCa/YPPRqOMPOdU=; b=fKLL98YeyJQPO5SERf6bMql43upxDZhYe1Dw6om8Rr8JE1jqzJXey+w1wvsSzTnR2M QzI5KCW0pQCBwHzTzN3W9frjRGKS4FPJPpPBYyAHaon4H/xe/O0a68/NoqAsUguqCFOF xC9AgiP8rB6T3WJEZCMp7pxv6y9THS4nw+j+vmDKKEdfXsQ0BpS78mTDg4z6T0fvP7GD WVlZt0ykGgmXzoP2M0MlPdeegeVdYv66uxcTEs5nOZ9j833qyAECoBgW7Xu+8hV284jp MX+X8blvApJw/CYaBLz7+zrIQWCec9i1WAN9F5Qp9MNrjtEGv3vH2N4w5SZtCSynf01c mucw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:from:date:message-id:subject:to:cc; bh=8+KFyjrXcVTuDUOZNcC8AoShXQWVnCa/YPPRqOMPOdU=; b=LTk/o0FRRnmdTlwvSaYa/TXLtlEhk4j5egaXAue07Wj0hwsweV/GJtX97OfkMATCi0 qSXSYwaMCDzGXeOf1vB5Hfrwt8ChDd6jlsw+I0CBk2bD0l256JNGWEK9Ieoe8HnZNNkJ zVPfyK/HhS1qNFWDXINM7Mx8j0jycduhBFJ0+sUBFX/VoZ70xRpb69MD2jaS8etxoQiu 15BGeqWcWqW0c2EDsboWgJxurvqdMXIsqFWTFA7Gr+UUq0Xxq8kxiyAIUp7Iec7wjO/n /JXys4fgqKYgUX/K6HiB8METN/DccLUL4N2/n0VDzu3zWJjzp7eQ/oEcXsy47NXDIS4/ jm9w==
X-Gm-Message-State: AOAM530FlUAohrEhanmiYPbEO9y4vrHRcWJF7to1FVunasrZlPnoyNib mrwLDES1IUUxGoBG+5Bn830YEvN/HTFOyjATfXhwRmDmRh0=
X-Google-Smtp-Source: ABdhPJxwzsCcFRCN97D5Vl1JdhxXNgLJtKzb513tENlTXQ0g4K6ovpfV64KNWE06ohK8u4d8YFAZ+XoD5xWlWfueuwk=
X-Received: by 2002:ad4:5e85:0:b0:433:2cf2:2552 with SMTP id jl5-20020ad45e85000000b004332cf22552mr1341622qvb.111.1646042444157; Mon, 28 Feb 2022 02:00:44 -0800 (PST)
MIME-Version: 1.0
From: Yoshifumi Nishida <nsd.ietf@gmail.com>
Date: Mon, 28 Feb 2022 02:00:33 -0800
Message-ID: <CAAK044RU_nXaqcj-QOF7upQOzQB=KdnpOXKHujbUo22WrM+3Bw@mail.gmail.com>
To: "tcpm@ietf.org Extensions" <tcpm@ietf.org>
Cc: tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: multipart/alternative; boundary="0000000000002b574d05d9111fe6"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/epJTfDs_YX40UPbOQKImydLm9eU>
Subject: [tcpm] Agenda for TCPM WG meeting at IETF 113
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 28 Feb 2022 10:00:50 -0000

--0000000000002b574d05d9111fe6
Content-Type: text/plain; charset="UTF-8"

Hello,

Our WG meeting is scheduled on Wednesday (3/23) 13:30-15:30 UTC (please
check https://datatracker.ietf.org/meeting/113/agenda/ )
If you are planning to present something, please let the chairs know the
following information.

* Title / draft name
* Presenter's name
* Total time (including Q/A)
* on-site or remote (just in case)

Thanks,
--
tcpm co-chairs

--0000000000002b574d05d9111fe6
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hello,<div><br></div><div>Our WG meeting is scheduled on W=
ednesday (3/23) 13:30-15:30 UTC (please check=C2=A0<a href=3D"https://datat=
racker.ietf.org/meeting/113/agenda/" target=3D"_blank">https://datatracker.=
ietf.org/meeting/113/agenda/</a>=C2=A0)</div><div>If you are planning to pr=
esent something, please let the chairs know the following information.<div>=
<br>* Title / draft name<br>* Presenter&#39;s name<br>* Total time (includi=
ng Q/A)</div></div><div>* on-site or remote (just in case)</div><div><br></=
div><div>Thanks,</div><div>--</div><div>tcpm co-chairs</div></div>

--0000000000002b574d05d9111fe6--


From nobody Mon Feb 28 02:07:12 2022
Return-Path: <nsd.ietf@gmail.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B602E3A0BDD; Mon, 28 Feb 2022 02:07:09 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.107
X-Spam-Level: 
X-Spam-Status: No, score=-2.107 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id B-m4Is8fWqJQ; Mon, 28 Feb 2022 02:07:05 -0800 (PST)
Received: from mail-qk1-x730.google.com (mail-qk1-x730.google.com [IPv6:2607:f8b0:4864:20::730]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CE32E3A0B8B; Mon, 28 Feb 2022 02:07:04 -0800 (PST)
Received: by mail-qk1-x730.google.com with SMTP id c7so9890655qka.7; Mon, 28 Feb 2022 02:07:04 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112;  h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=CQK+z5iWHPVygVF/vZkjmpMcIUDos0bMg8PJnnQEaps=; b=PSm4rUA5iyYeXE9ouymulgIXibGo/moRa5Kb1bvYBALE+zsrgril9t2lJuqB4mdvIS Jk1b30HAVhGbkKmc5TNqGjTSeaQwtOMZEjPqMqy+7Ei4/79KhYsjKgXhprVkPaseY39x Irxb13IXcsXs6pOkiJYOwyR/QMpabV9CpWJm/BtZusDPmpqJIs+8+ru4+bBhy3wb1udv awAoXT0hC8dXAWJvES0JAeCBuVo7VlqmBBp+TsSiT5KEFuYSDmG6MdKgR9Ns33iTNBHk 8lVVxktIE7T9WobvApiBm6CIrxdEfD8xk9L6icwntcFUaHdX++BtkXgC6jYDumWFiAT2 cwcg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=CQK+z5iWHPVygVF/vZkjmpMcIUDos0bMg8PJnnQEaps=; b=0LiIH/rtyvTeij1tTH6DH6QdCMe4z88pa/bc6DXmmrpkMOKQdBKSWuGZ9JXN3kUsCx epU38XWAoeq1SDeT35Jb/KUHsKZy+s7BkiTLYLa1buV7E+v9GhXFgnMDiHVu2FD7pzSJ zjsq9aM0+i4uBK8oKnT6/7kt9c+2FfRQCNSRrD6Gn4lKdhmPU79KnCKnhpXHnSVP5UV4 DD6ClbccFZkHM32TYQoJnLSVhud/+JZ9QeAQ2LbXMvSCu4/nfBMrzQ4PhEHMgBFuJdxJ EsemwEqEaql3xwnP00yVZ/XBvxqp//QQ9Ypr06suwSnDKQ1wlu9VAiVnXssvDsCIAryz 41lA==
X-Gm-Message-State: AOAM532Gw0kqZDby5vZ5lCeeQTBdn8tczO/yPvzZhknc+Z9yEWNkrDz8 ytx8N+FdWyCcxSf5ZL++ttZSGGjf7vVSQ1VYpv3+dQby
X-Google-Smtp-Source: ABdhPJx0aoCgaZGOECTYyYljbY0hh5e1vVwlYKHnItPIcCRap8zRtDDWW0qZG8oyROsiSutSJe3xrzuemQO9OiIUPxo=
X-Received: by 2002:a05:620a:111b:b0:648:fd03:c029 with SMTP id o27-20020a05620a111b00b00648fd03c029mr10584651qkk.314.1646042823423; Mon, 28 Feb 2022 02:07:03 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com>
In-Reply-To: <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com>
From: Yoshifumi Nishida <nsd.ietf@gmail.com>
Date: Mon, 28 Feb 2022 02:06:52 -0800
Message-ID: <CAAK044Q5Lv175T6RcVp+XKEfqZ=Z_+TfjJ5Q9NzzA6wVerMkSw@mail.gmail.com>
To: "tcpm@ietf.org Extensions" <tcpm@ietf.org>
Cc: tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: multipart/alternative; boundary="000000000000c67d4705d911351d"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/C8_y9Hte5RlzmlbH-I5ar5Htc-Y>
Subject: [tcpm] Proceeding 8312bis draft (Re: 2nd WGLC for draft-ietf-tcpm-rfc8312bis)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 28 Feb 2022 10:07:10 -0000

--000000000000c67d4705d911351d
Content-Type: text/plain; charset="UTF-8"

Hi,

We've been discussing how to proceed 8312bis draft, but I think we still
haven't settled yet.
In my understanding, there seems to be the following options so far.
if you have some thoughts on them or you have other options for this,
please share.
BTW, please note that this discussion might affect our future drafts
because we want to apply the same bar for all standard docs as much as
possible.

1: Publish the draft as a non-standard doc
    a: publish as an informational doc

    b: publish as an experimental doc
        * we might need to describe what experiments are expected.

2: Publish the draft as a PS doc
    a: do additional experiments to make sure there's no concern as a PS
doc.
        * we will need to decide what kinds of experiments are required,
what would be the expected results before the experiments.
        * we need to delay the publication of the doc until the validation
of the experiments has finished

    b: publish an additional doc that describes the checking points on the
doc (and other related docs) for long term analysis.

Thanks,
--
Yoshi


On Mon, Jan 31, 2022 at 11:43 PM Yoshifumi Nishida <nsd.ietf@gmail.com>
wrote:

> Hello,
>
> After some discussions among chairs, we decided to run the 2nd WGLC on
> draft-ietf-tcpm-rfc8312bis in consideration of the importance of the draft.
> We'll be grateful if you could send your feedback to the ML. The WGLC runs
> until *Feb 11*.
>
> If interested, you can check in-depth past discussions in the following
> URL.
> https://github.com/NTAP/rfc8312bis/
>
> Thank you so much!
> --
> tcpm co-chairs
>
>
> On Wed, Jan 26, 2022 at 2:50 AM Lars Eggert <lars@eggert.org> wrote:
>
>> Hi,
>>
>> this -06 version rolls in all the changes requested during (and after)
>> WGLC ended.
>>
>> I'll leave it up to the chairs to decide if another WGLC is warranted or
>> the document can progress as-is.
>>
>> Thanks,
>> Lars
>>
>>
>> > On 2022-1-26, at 11:12, 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 TCP Maintenance and Minor Extensions
>> WG of the IETF.
>> >
>> >        Title           : CUBIC for Fast and Long-Distance Networks
>> >        Authors         : Lisong Xu
>> >                          Sangtae Ha
>> >                          Injong Rhee
>> >                          Vidhi Goel
>> >                          Lars Eggert
>> >       Filename        : draft-ietf-tcpm-rfc8312bis-06.txt
>> >       Pages           : 35
>> >       Date            : 2022-01-26
>> >
>> > Abstract:
>> >   CUBIC is a standard TCP congestion control algorithm that uses a
>> >   cubic function instead of a linear congestion window increase
>> >   function to improve scalability and stability over fast and long-
>> >   distance networks.  CUBIC has been adopted as the default TCP
>> >   congestion control algorithm by the Linux, Windows, and Apple stacks.
>> >
>> >   This document updates the specification of CUBIC to include
>> >   algorithmic improvements based on these implementations and recent
>> >   academic work.  Based on the extensive deployment experience with
>> >   CUBIC, it also moves the specification to the Standards Track,
>> >   obsoleting RFC 8312.  This also requires updating RFC 5681, to allow
>> >   for CUBIC's occasionally more aggressive sending behavior.
>> >
>> >
>> > The IETF datatracker status page for this draft is:
>> > https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis/
>> >
>> > There is also an HTML version available at:
>> > https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-06.html
>> >
>> > A diff from the previous version is available at:
>> > https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rfc8312bis-06
>> >
>> >
>> > Internet-Drafts are also available by rsync at rsync.ietf.org:
>> :internet-drafts
>> >
>> >
>> > _______________________________________________
>> > tcpm mailing list
>> > tcpm@ietf.org
>> > https://www.ietf.org/mailman/listinfo/tcpm
>>
>> _______________________________________________
>> tcpm mailing list
>> tcpm@ietf.org
>> https://www.ietf.org/mailman/listinfo/tcpm
>>
>

--000000000000c67d4705d911351d
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr">Hi,<br></div><div dir=3D"ltr"><br></div><=
div>We&#39;ve been discussing how to proceed 8312bis draft, but I think we =
still haven&#39;t settled yet.</div><div>In my understanding, there seems t=
o be the following options so far.</div><div>if you have some thoughts on t=
hem or you have other options for this, please share.=C2=A0=C2=A0</div><div=
>BTW, please note that this discussion might affect our future drafts becau=
se we want to apply the same bar for all standard docs as much as possible.=
=C2=A0<br></div><div><br></div><div>1: Publish the draft as a non-standard =
doc</div><div>=C2=A0 =C2=A0 a: publish as an informational doc</div><div>=
=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0</div><div>=C2=A0 =C2=A0 b: publish as an =
experimental doc</div><div>=C2=A0 =C2=A0 =C2=A0 =C2=A0 * we might need to d=
escribe what experiments are expected.</div><div><br></div><div>2: Publish =
the draft as a PS doc</div><div>=C2=A0 =C2=A0 a: do additional=C2=A0experim=
ents to make sure there&#39;s no concern as a PS doc.=C2=A0</div><div>=C2=
=A0 =C2=A0 =C2=A0 =C2=A0 * we will need to decide what kinds of experiments=
 are required, what would be the expected results before the experiments.</=
div><div>=C2=A0 =C2=A0 =C2=A0 =C2=A0 * we need to delay the publication of =
the doc until the validation of the experiments has finished</div><div>=C2=
=A0 =C2=A0 =C2=A0 =C2=A0</div><div>=C2=A0 =C2=A0 b: publish an additional=
=C2=A0doc that describes the checking points on the doc (and other related =
docs) for long term analysis.=C2=A0=C2=A0</div><div><br></div><div>Thanks,<=
/div><div>--</div><div>Yoshi</div><div><br></div><div><br></div><div class=
=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Mon, Jan 31, 2022=
 at 11:43 PM Yoshifumi Nishida &lt;<a href=3D"mailto:nsd.ietf@gmail.com">ns=
d.ietf@gmail.com</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" =
style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);pa=
dding-left:1ex"><div dir=3D"ltr"><div>Hello,</div><div><br></div><div>After=
 some discussions among chairs, we decided to run the 2nd WGLC on draft-iet=
f-tcpm-rfc8312bis in consideration of the importance of the draft.=C2=A0</d=
iv><div>We&#39;ll be grateful if you could send your feedback to the ML. Th=
e WGLC runs until *Feb 11*.</div><div><br></div><div>If interested, you can=
 check in-depth past discussions in the following URL.</div><div><a href=3D=
"https://github.com/NTAP/rfc8312bis/" target=3D"_blank">https://github.com/=
NTAP/rfc8312bis/</a><br></div><div><br></div><div>Thank you so much!</div><=
div>--</div><div>tcpm co-chairs</div><div><br></div><br><div class=3D"gmail=
_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Wed, Jan 26, 2022 at 2:50 =
AM Lars Eggert &lt;<a href=3D"mailto:lars@eggert.org" target=3D"_blank">lar=
s@eggert.org</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" styl=
e=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);paddin=
g-left:1ex">Hi,<br>
<br>
this -06 version rolls in all the changes requested during (and after) WGLC=
 ended.<br>
<br>
I&#39;ll leave it up to the chairs to decide if another WGLC is warranted o=
r the document can progress as-is.<br>
<br>
Thanks,<br>
Lars<br>
<br>
<br>
&gt; On 2022-1-26, at 11:12, <a href=3D"mailto:internet-drafts@ietf.org" ta=
rget=3D"_blank">internet-drafts@ietf.org</a> wrote:<br>
&gt; <br>
&gt; <br>
&gt; A New Internet-Draft is available from the on-line Internet-Drafts dir=
ectories.<br>
&gt; This draft is a work item of the TCP Maintenance and Minor Extensions =
WG of the IETF.<br>
&gt; <br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Title=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0: CUBIC for Fast and Long-Distance Networks<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 Authors=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0: =
Lisong Xu<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Sangtae Ha<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Injong Rhee<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Vidhi Goel<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0 =C2=A0 Lars Eggert<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Filename=C2=A0 =C2=A0 =C2=A0 =C2=A0 : draft-=
ietf-tcpm-rfc8312bis-06.txt<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Pages=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0: 35<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0Date=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 : 2022-01-26<br>
&gt; <br>
&gt; Abstract:<br>
&gt;=C2=A0 =C2=A0CUBIC is a standard TCP congestion control algorithm that =
uses a<br>
&gt;=C2=A0 =C2=A0cubic function instead of a linear congestion window incre=
ase<br>
&gt;=C2=A0 =C2=A0function to improve scalability and stability over fast an=
d long-<br>
&gt;=C2=A0 =C2=A0distance networks.=C2=A0 CUBIC has been adopted as the def=
ault TCP<br>
&gt;=C2=A0 =C2=A0congestion control algorithm by the Linux, Windows, and Ap=
ple stacks.<br>
&gt; <br>
&gt;=C2=A0 =C2=A0This document updates the specification of CUBIC to includ=
e<br>
&gt;=C2=A0 =C2=A0algorithmic improvements based on these implementations an=
d recent<br>
&gt;=C2=A0 =C2=A0academic work.=C2=A0 Based on the extensive deployment exp=
erience with<br>
&gt;=C2=A0 =C2=A0CUBIC, it also moves the specification to the Standards Tr=
ack,<br>
&gt;=C2=A0 =C2=A0obsoleting RFC 8312.=C2=A0 This also requires updating RFC=
 5681, to allow<br>
&gt;=C2=A0 =C2=A0for CUBIC&#39;s occasionally more aggressive sending behav=
ior.<br>
&gt; <br>
&gt; <br>
&gt; The IETF datatracker status page for this draft is:<br>
&gt; <a href=3D"https://datatracker.ietf.org/doc/draft-ietf-tcpm-rfc8312bis=
/" rel=3D"noreferrer" target=3D"_blank">https://datatracker.ietf.org/doc/dr=
aft-ietf-tcpm-rfc8312bis/</a><br>
&gt; <br>
&gt; There is also an HTML version available at:<br>
&gt; <a href=3D"https://www.ietf.org/archive/id/draft-ietf-tcpm-rfc8312bis-=
06.html" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/archive/=
id/draft-ietf-tcpm-rfc8312bis-06.html</a><br>
&gt; <br>
&gt; A diff from the previous version is available at:<br>
&gt; <a href=3D"https://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-rfc8312=
bis-06" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org/rfcdiff?u=
rl2=3Ddraft-ietf-tcpm-rfc8312bis-06</a><br>
&gt; <br>
&gt; <br>
&gt; Internet-Drafts are also available by rsync at rsync.ietf.org::interne=
t-drafts<br>
&gt; <br>
&gt; <br>
&gt; _______________________________________________<br>
&gt; tcpm mailing list<br>
&gt; <a href=3D"mailto:tcpm@ietf.org" target=3D"_blank">tcpm@ietf.org</a><b=
r>
&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/tcpm" rel=3D"noreferr=
er" target=3D"_blank">https://www.ietf.org/mailman/listinfo/tcpm</a><br>
<br>
_______________________________________________<br>
tcpm mailing list<br>
<a href=3D"mailto:tcpm@ietf.org" target=3D"_blank">tcpm@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/tcpm" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/listinfo/tcpm</a><br>
</blockquote></div></div>
</blockquote></div></div>

--000000000000c67d4705d911351d--


From nobody Mon Feb 28 02:41:23 2022
Return-Path: <noreply@ietf.org>
X-Original-To: tcpm@ietf.org
Delivered-To: tcpm@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id D74B83A1092; Mon, 28 Feb 2022 02:41:13 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: Gorry Fairhurst via Datatracker <noreply@ietf.org>
To: <tsv-art@ietf.org>
Cc: draft-ietf-tcpm-yang-tcp.all@ietf.org, last-call@ietf.org, tcpm@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 7.46.0
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <164604487381.13348.5521968806633530984@ietfa.amsl.com>
Reply-To: Gorry Fairhurst <gorry@erg.abdn.ac.uk>
Date: Mon, 28 Feb 2022 02:41:13 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/1SpPpI6LEdy2ClE5zBhbkkRNBFg>
Subject: [tcpm] Tsvart last call review of draft-ietf-tcpm-yang-tcp-06
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 28 Feb 2022 10:41:22 -0000

Reviewer: Gorry Fairhurst
Review result: Ready with Nits

This document has been reviewed as part of the transport area review team's
ongoing effort to review key IETF documents. These comments were written
primarily for the transport area directors, but are copied to the document's
authors and WG to allow them to address any issues raised and also to the IETF
discussion list for information.

When done at the time of IETF Last Call, the authors should consider this
review as part of the last-call comments they receive. Please always CC
tsv-art@ietf.org if you reply to or forward this review.

>From a transport perspective, the document appears ready with minor NiTs on the
use of wording and references.  It specifies a YANG model for TCP.  It is
well-written and does not raise any congestion control or other
transport-related protocol issues. There are two normative dependencies, both
adopted WG items in/completing WGLC.

The following NiTs are listed below:

/as it allows enabling of TCP-AO/
- are the words /as it/ best here, or would it be better to say /therefore this
allows/ or /and this allows/ or something similar?

/This issue is further discussed below./
- I did not really see an /issue/ discussed. Is it possible just to say the
support for MD5 is discussed or something similar?

Is there a suitable RFC ref for TCP_NODELAY (the TCP base spec)?:
/algorithm by TCP_NODELAY/.

Suggest: /traditionally different TCP implementation/traditionally different
TCP implementations/ - i.e.e add /s/?

/As the TCP MD5 signature option is obsoleted by TCP-AO, it is strongly
RECOMMENDED to use TCP-AO./ - It seems odd to declare this in the scoping
section of a document without a REF!  Also, what is a "strong" recommendation.
Is this simply restating the intent of RFC5925, in which case this could be
rephrased a little, perhaps something like: /The TCP MD5 signature option was
obsoleted by TCP-AO in 2010. Current implementations are therefore RECOMMENDED
to use TCP-AO [RFCxxxx]./

/This version of the module does not cover Multipath TCP [RFC8684]./
- To me, it seems OK that this YANG model does not "specify a method for
Multipath TCP", but the implications of the word /cover/ is a little unclear,
does that mean other specs might provide additional functions to support
multipath? or that that this spec prevents the use of multipath? or can only be
used when multipath is not used? There seems a wide variety of options...

Section 4 appears to require RFC-Ed action to replace ID labels within the yang
models with the published RFC Refs and a note to the editor here could be
helpful also at the start of this section. e.g., I-D.ietf-tcpm-rfc793bis.

The meaning of  the word "segment" is very clear for TCP, but would perhaps be
worth prefixing by TCP, within the model since the word /segment/ is also used
in other contexts: /The total number of segments received in error/. This might
be better as: /The total number of TCP segments received in error/. /The total
number of segments received,/ This might be better as: /The total number of TCP
segments received,/ and similarly for segments sent and segments retransmitted.




From nobody Mon Feb 28 03:55:15 2022
Return-Path: <Michael.Scharf@hs-esslingen.de>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 52DE43A1067; Mon, 28 Feb 2022 03:55:03 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -7.107
X-Spam-Level: 
X-Spam-Status: No, score=-7.107 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_HI=-5, SPF_HELO_NONE=0.001, SPF_NONE=0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=hs-esslingen.de
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id dOtFa7nNY1jG; Mon, 28 Feb 2022 03:54:58 -0800 (PST)
Received: from mail.hs-esslingen.de (mail.hs-esslingen.de [134.108.32.78]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id F15DA3A1008; Mon, 28 Feb 2022 03:54:57 -0800 (PST)
Received: from localhost (localhost.localdomain [127.0.0.1]) by mail.hs-esslingen.de (Postfix) with ESMTP id A596925A15; Mon, 28 Feb 2022 12:54:55 +0100 (CET)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=hs-esslingen.de; s=mail; t=1646049295; bh=QMI6yfb3W39QP9jtiQfdQsPPnE5Q7fPMhWWRsuOO7VM=; h=From:To:CC:Subject:Date:References:In-Reply-To:From; b=azYsVlJ8s/FLrIb8Jo/AIhTm1klEtqs/iLK22j9Uc1xSyvSo9ARvMbWBGaZgTMJX3 eeIh979qAqDSYPnWyDsWuW8Ds4r7pEM8z061D+oPbLTmiPBoaQFOhhlBsAc9A7Oq+4 9sPddclbebAbdXPmE5mXjeTyWC9sr2HKT/qrqARk=
X-Virus-Scanned: by amavisd-new-2.7.1 (20120429) (Debian) at hs-esslingen.de
Received: from mail.hs-esslingen.de ([127.0.0.1]) by localhost (hs-esslingen.de [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0wuLcSio5-zV; Mon, 28 Feb 2022 12:54:54 +0100 (CET)
Received: from rznt8202.rznt.rzdir.fht-esslingen.de (rznt8202.hs-esslingen.de [134.108.48.165]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.hs-esslingen.de (Postfix) with ESMTPS; Mon, 28 Feb 2022 12:54:54 +0100 (CET)
Received: from rznt8202.rznt.rzdir.fht-esslingen.de (134.108.48.165) by rznt8202.rznt.rzdir.fht-esslingen.de (134.108.48.165) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.18; Mon, 28 Feb 2022 12:54:48 +0100
Received: from rznt8202.rznt.rzdir.fht-esslingen.de ([fe80::aca4:171a:3ee1:57e0]) by rznt8202.rznt.rzdir.fht-esslingen.de ([fe80::aca4:171a:3ee1:57e0%3]) with mapi id 15.01.2375.018; Mon, 28 Feb 2022 12:54:48 +0100
From: "Scharf, Michael" <Michael.Scharf@hs-esslingen.de>
To: "touch@strayalpha.com" <touch@strayalpha.com>, Benjamin Kaduk <kaduk@mit.edu>
CC: Wes Eddy <wes@mti-systems.com>, "draft-ietf-tcpm-rfc793bis@ietf.org" <draft-ietf-tcpm-rfc793bis@ietf.org>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, The IESG <iesg@ietf.org>, "tcpm-chairs@ietf.org" <tcpm-chairs@ietf.org>
Thread-Topic: [tcpm] Benjamin Kaduk's Discuss on draft-ietf-tcpm-rfc793bis-25: (with DISCUSS and COMMENT)
Thread-Index: AQHXsCvcnc/noM1n0UyxQ6Ty5CuigaxcPJUAgElnwoCAABvcAIAEDXqA
Date: Mon, 28 Feb 2022 11:54:48 +0000
Message-ID: <927f0d019fee43c2a385b04c7909c0e9@hs-esslingen.de>
References: <163236803976.28405.5643771942452620510@ietfa.amsl.com> <6f6e7b90-081a-74b4-b329-8879addcb8c4@mti-systems.com> <20220225210031.GV12881@kduck.mit.edu> <5F0B3950-90A5-4C8E-9C2D-7C70B2108C1B@strayalpha.com>
In-Reply-To: <5F0B3950-90A5-4C8E-9C2D-7C70B2108C1B@strayalpha.com>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [134.108.140.249]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/hXulKDOF8w2tDrio_hiCZCDaUNY>
Subject: Re: [tcpm] Benjamin Kaduk's Discuss on draft-ietf-tcpm-rfc793bis-25: (with DISCUSS and COMMENT)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 28 Feb 2022 11:55:04 -0000

Q2hpbWluZyBpbiBhcyBUQ1BNIGNvLWNoYWlyIGFuZCBkb2N1bWVudCBzaGVwaGVyZC4uLg0KDQo+
ID4gSGVyZSwgd2UNCj4gPiBoYXZlIHRoZSBjb25zdHJhaW50IG9mIHJldGFpbmluZyBjb21wYXRp
YmlsaXR5IHdpdGggdGhlIGV4aXN0aW5nIGRlcGxveW1lbnQNCj4gPiBiYXNlICh3aGljaCBpcyBh
IHZlcnkgY29tcGVsbGluZyBjb25zdHJhaW50ISksIHNvIHlvdSBkbyBub3Qgc2VlIG1lDQo+ID4g
YWR2b2NhdGluZyBmb3IgbWFuZGF0b3J5IHRjcGNyeXB0IChmb3IgZXhhbXBsZSkuICBCdXQgSSB3
YW50IHRvIHNlZSBzb21lDQo+ID4gZXhwbGFuYXRpb24gb2Ygd2hhdCBoYXJtIG9yIHJpc2sgdGhl
cmUgaXMgaW4gc2F5aW5nIHRoYXQgdGhlIElTTiBpcyBhbHdheXMNCj4gPiBwcm9kdWNlZCB3aXRo
IHRoZSBQUkYgb2YgdGhlIDUtdHVwbGUgKG9rYXksIDQtdHVwbGUgc2luY2UgdGhlIHByb3RvY29s
IGlzDQo+ID4gYWx3YXlzIFRDUCksIHRvIG1vdGl2YXRlIGEgZGl2ZXJnZW5jZSBmcm9tIHRoZSBt
b3JlLXNlY3VyZSBiZWhhdmlvciBhbmQNCj4gPiB0aGVyZWJ5IGp1c3RpZnkgcmV0YWluaW5nIHRo
ZSBiZWhhdmlvciBjdXJyZW50bHkgaW4gdGhlIGRyYWZ0LiAgV2hhdA0KPiA+IGNvbnN0cmFpbnQg
YXJlIHdlIHN1YmplY3QgdG8gdGhhdCBwcmV2ZW50cyBkb2luZyB0aGUgcmlnaHQgKHNlY3VyaXR5
LXdpc2UpDQo+ID4gdGhpbmc/DQo+IA0KPiBXZSBhbHJlYWR5IGhhdmUgUkZDNjUyOCB0aGF0IHNw
ZWNpZmllcyB0aGlzIGFzIGEgU0hPVUxELg0KPg0KPiBNeSBxdWVzdGlvbiB0byB5b3UgaXMg4oCc
d2hhdCBoYXMgY2hhbmdlZCBzaW5jZSAyMDEyIHRoYXQgbWFrZXMgeW91IHRoaW5rDQo+IHRoaXMg
aXMgbm93IGFwcHJvcHJpYXRlIGZvciBhIE1VU1TigJ0uDQoNCkl0IHdhcyBjbGVhciBjb25zZW5z
dXMgaW4gVENQTSAqbm90KiB0byByZWFsaXplIFNIT1VMRC9NVVNUIG1vZGlmaWNhdGlvbnMgb2Yg
dGhlIFRDUCBwcm90b2NvbCBpbiA3OTNiaXMsIGJ1dCBpbnN0ZWFkIHRvIGxpbWl0IHRoZSBkb2N1
bWVudCB0byBwcm90b2NvbCBtZWNoYW5pc21zIHRoYXQgYWxyZWFkeSBoYXZlIElFVEYgY29uc2Vu
c3VzLg0KDQpJdCB3YXMgYWxzbyBUQ1BNIGNvbnNlbnN1cyB0aGF0IHN0YWNrcyB0aGF0IGNvcnJl
Y3RseSBpbXBsZW1lbnQgZXhpc3RpbmcgVENQIHN0YW5kYXJkcyBwcmlvciB0byA3OTNiaXMgc2hv
dWxkIG5vdCBiZWNvbWUgaW5jb21wYXRpYmxlIHRvIHRoZSBUQ1Agc3RhbmRhcmRzIG9uY2UgNzkz
YmlzIGlzIHB1Ymxpc2hlZC4NCiANCj4gQUZBSUNULCB0aGF04oCZcyB3aGVyZSB0aGUgYnVyZGVu
IG9mIHByb29mIGlzLCBub3Qgb24gdXMgdG8gZGVmZW5kIGtlZXBpbmcgYQ0KPiBTSE9VTEQuDQoN
CklmIGEgU0hPVUxEIHNoYWxsIGJlIHJlcGxhY2VkIGJ5IGEgTVVTVCwgYW4gYWx0ZXJuYXRpdmUg
YXBwcm9hY2ggd291bGQgYmUgdG8gc3RhcnQgYSBuZXcgc3RhbmRhcmRzIHRyYWNrIGRvY3VtZW50
IGluIFRDUE0sIGdldCBmZWVkYmFjayBmcm9tIGltcGxlbWVudGVycywgZW5zdXJlIHRoYXQgdGhl
IGNoYW5nZSBpbmRlZWQgaGFzIFRDUE0gd29ya2luZyBncm91cCBhbmQgSUVURiBjb25zZW5zdXMs
IGFuZCB0aGVuIHVwZGF0ZSB0aGUgVENQIHN0YW5kYXJkcyBhY2NvcmRpbmdseS4NCg0KSW4gdGhp
cyBzcGVjaWZpYyBjYXNlLCB0aGF0IHNwZWMgd291bGQgYmUgcHJvYmFibHkgYSBzaG9ydCBkb2N1
bWVudCB0aGF0IGNhbiBiZSByZXZpZXdlZCByZWxhdGl2ZWx5IGVhc2lseSBhbmQgdGh1cyBwdWJs
aXNoZWQgZmFzdCwgaWYgdGhlIFRDUE0gd29ya2luZyBncm91cCBhZ3JlZXMgdGhhdCBpdCBpcyB0
aGUgcmlnaHQtdGhpbmctdG8tZG8uDQoNCk1pY2hhZWwNCg0K


From nobody Mon Feb 28 06:29:50 2022
Return-Path: <ncardwell@google.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AE67B3A112A for <tcpm@ietfa.amsl.com>; Mon, 28 Feb 2022 06:29:48 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -17.608
X-Spam-Level: 
X-Spam-Status: No, score=-17.608 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_MED=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, ENV_AND_HDR_SPF_MATCH=-0.5, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001, USER_IN_DEF_DKIM_WL=-7.5, USER_IN_DEF_SPF_WL=-7.5] autolearn=unavailable autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=google.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id N3DtAXdBy2BJ for <tcpm@ietfa.amsl.com>; Mon, 28 Feb 2022 06:29:45 -0800 (PST)
Received: from mail-qv1-xf2f.google.com (mail-qv1-xf2f.google.com [IPv6:2607:f8b0:4864:20::f2f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 944313A1129 for <tcpm@ietf.org>; Mon, 28 Feb 2022 06:29:45 -0800 (PST)
Received: by mail-qv1-xf2f.google.com with SMTP id e22so12668195qvf.9 for <tcpm@ietf.org>; Mon, 28 Feb 2022 06:29:45 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=OpWN07ZAXMwLgFgGfMs+VQOdcgJYmAv6AAvzwillocs=; b=jQsfZ3WNdzWBTVT/NDxB8GjmCcMqmv+/cSEboo3KOW4J0qe3lBu2MED006SuUlx+b/ NKVGVHrmlJY18G6qxbcsmFbed09f3qQF8sMHjFhnZLWqF+1RbEdSGWWNH2EhpYMfwMWY hzEuz77+w8QtuHQZwNAWOKTJp+urPmAh3J1iYMI+fsMYNZY1vp3tjWwMCXefS+CcjuE/ DpVpaSOxqQBOPMML51FXgqFPWwFxfg6V2/nwOFQoPKUeAHKv+KzW0XEpjzy0MnyluZUM JLdoA1Jc/zmB+jcsB0ie6hgbLfeDFpBHhsHwQ9tNy8/XVWc2wVRTV/o78M8sg3wot55n Vjfg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=OpWN07ZAXMwLgFgGfMs+VQOdcgJYmAv6AAvzwillocs=; b=jLHmVMICBF08RenHZ2U0POZMdTV7ercw8xmhw2wbzuaSgKJEa0sh9BUCjTWj5KeCz0 sdx8e2N/sVBFVFTIQbEoJEC3i2JCDYVKx4B9Z3sU0N7jO2uWC2OPLQXdTRC05WzE6jnv LpiE9u0HnBdA8//al2RnEi3C321G5FhXMA0BRtm4kKKWEqbqm2e75x+u0BsSZe3HVPXT Zz+03qdZLvnFwEKU1b5rxJ155fM2C4NUtJ/JJaQDUKinXM8C1/Q1nF7roSm0Zimja/a+ 6LqkwxzXf8JQ4EzbbDbNMJsqjb1dGYTOLwY8tWlfzAA/FQIzptfl6uwoNjC9WDuGHv7a BDHA==
X-Gm-Message-State: AOAM533mkULSo0YpZYL0xcs5W0b4uAWVA+0eXKigNTNG1t/VYHCPEEA2 9WPDIYOQbn/YUf5sE3NSK1J7IraGoWcX5IIlVrXf4g==
X-Google-Smtp-Source: ABdhPJyIqR8woeQv2cjMNSr7VdEuvXCveykIhiyo4KGSfviBn897H1gcxDvVFiY9OVzGIrPEM+ZiVj/M1A6CoS4pzrM=
X-Received: by 2002:a05:622a:5c8:b0:2df:f3d4:e214 with SMTP id d8-20020a05622a05c800b002dff3d4e214mr9277499qtb.183.1646058583956; Mon, 28 Feb 2022 06:29:43 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <CAAK044Q5Lv175T6RcVp+XKEfqZ=Z_+TfjJ5Q9NzzA6wVerMkSw@mail.gmail.com>
In-Reply-To: <CAAK044Q5Lv175T6RcVp+XKEfqZ=Z_+TfjJ5Q9NzzA6wVerMkSw@mail.gmail.com>
From: Neal Cardwell <ncardwell@google.com>
Date: Mon, 28 Feb 2022 09:29:28 -0500
Message-ID: <CADVnQykqf5zU-or7VbjcaU8ARTv1AOt0zR+xVquBG8ck5SqyRA@mail.gmail.com>
To: Yoshifumi Nishida <nsd.ietf@gmail.com>
Cc: "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: multipart/alternative; boundary="0000000000002db15305d914e1b9"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/_dNl_K5HxyL0X5UkgpKeuagXv5E>
Subject: Re: [tcpm] Proceeding 8312bis draft (Re: 2nd WGLC for draft-ietf-tcpm-rfc8312bis)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 28 Feb 2022 14:29:49 -0000

--0000000000002db15305d914e1b9
Content-Type: text/plain; charset="UTF-8"

On Mon, Feb 28, 2022 at 5:07 AM Yoshifumi Nishida <nsd.ietf@gmail.com>
wrote:

> Hi,
>
> We've been discussing how to proceed 8312bis draft, but I think we still
> haven't settled yet.
> In my understanding, there seems to be the following options so far.
> if you have some thoughts on them or you have other options for this,
> please share.
> BTW, please note that this discussion might affect our future drafts
> because we want to apply the same bar for all standard docs as much as
> possible.
>
> 1: Publish the draft as a non-standard doc
>     a: publish as an informational doc
>
>     b: publish as an experimental doc
>         * we might need to describe what experiments are expected.
>
> 2: Publish the draft as a PS doc
>     a: do additional experiments to make sure there's no concern as a PS
> doc.
>         * we will need to decide what kinds of experiments are required,
> what would be the expected results before the experiments.
>         * we need to delay the publication of the doc until the validation
> of the experiments has finished
>
>     b: publish an additional doc that describes the checking points on the
> doc (and other related docs) for long term analysis.
>

I would argue for:

  (2) publish the 8312bis draft as a PS doc, and

  (2b) publish an additional doc that describes any aspects of CUBIC where
the community would like to see further experiments or improvements.

IMHO Michael Scharf made an excellent argument for PS status for CUBIC on
Feb 22 (
https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5pQ6LrDE/ )
by simply quoting a definition of PS from RFC 7127:

   A Proposed Standard specification is stable, has resolved known
   design choices, has received significant community review, and
   appears to enjoy enough community interest to be considered valuable.

   Usually, neither implementation nor operational experience is
   required for the designation of a specification as a Proposed
   Standard.  However, such experience is highly desirable and will
   usually represent a strong argument in favor of a Proposed Standard
   designation.

   The IESG may require implementation and/or operational experience
   prior to granting Proposed Standard status to a specification that
   materially affects the core Internet protocols or that specifies
   behavior that may have significant operational impact on the
   Internet.

   A Proposed Standard will have no known technical omissions with
   respect to the requirements placed upon it.  Proposed Standards are
   of such quality that implementations can be deployed in the Internet.
   However, as with all technical specifications, Proposed Standards may
   be revised if problems are found or better solutions are identified,
   when experiences with deploying implementations of such technologies
   at scale is gathered.


It seems clear, IMHO, that CUBIC meets those criteria for being named a PS.

best regards,
neal

--0000000000002db15305d914e1b9
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote">=
<div dir=3D"ltr" class=3D"gmail_attr">On Mon, Feb 28, 2022 at 5:07 AM Yoshi=
fumi Nishida &lt;<a href=3D"mailto:nsd.ietf@gmail.com">nsd.ietf@gmail.com</=
a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0p=
x 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><d=
iv dir=3D"ltr"><div dir=3D"ltr">Hi,<br></div><div dir=3D"ltr"><br></div><di=
v>We&#39;ve been discussing how to proceed 8312bis draft, but I think we st=
ill haven&#39;t settled yet.</div><div>In my understanding, there seems to =
be the following options so far.</div><div>if you have some thoughts on the=
m or you have other options for this, please share.=C2=A0=C2=A0</div><div>B=
TW, please note that this discussion might affect our future drafts because=
 we want to apply the same bar for all standard docs as much as possible.=
=C2=A0<br></div><div><br></div><div>1: Publish the draft as a non-standard =
doc</div><div>=C2=A0 =C2=A0 a: publish as an informational doc</div><div>=
=C2=A0 =C2=A0 =C2=A0 =C2=A0=C2=A0</div><div>=C2=A0 =C2=A0 b: publish as an =
experimental doc</div><div>=C2=A0 =C2=A0 =C2=A0 =C2=A0 * we might need to d=
escribe what experiments are expected.</div><div><br></div><div>2: Publish =
the draft as a PS doc</div><div>=C2=A0 =C2=A0 a: do additional=C2=A0experim=
ents to make sure there&#39;s no concern as a PS doc.=C2=A0</div><div>=C2=
=A0 =C2=A0 =C2=A0 =C2=A0 * we will need to decide what kinds of experiments=
 are required, what would be the expected results before the experiments.</=
div><div>=C2=A0 =C2=A0 =C2=A0 =C2=A0 * we need to delay the publication of =
the doc until the validation of the experiments has finished</div><div>=C2=
=A0 =C2=A0 =C2=A0 =C2=A0</div><div>=C2=A0 =C2=A0 b: publish an additional=
=C2=A0doc that describes the checking points on the doc (and other related =
docs) for long term analysis.=C2=A0=C2=A0</div></div></blockquote><div><br>=
</div><div>I would argue for:</div><div><br></div><div>=C2=A0 (2) publish t=
he 8312bis draft as a PS doc, and=C2=A0</div><div><br></div><div>=C2=A0 (2b=
) publish an additional doc that describes any aspects of CUBIC where the c=
ommunity would like to see further experiments or improvements.</div><div><=
br></div>IMHO Michael Scharf made an excellent argument for PS status for C=
UBIC on Feb 22 ( <a href=3D"https://mailarchive.ietf.org/arch/msg/tcpm/k47h=
lSsx1lRupy5LRTJ5pQ6LrDE/">https://mailarchive.ietf.org/arch/msg/tcpm/k47hlS=
sx1lRupy5LRTJ5pQ6LrDE/</a> ) by simply quoting a definition of PS from RFC =
7127:</div><div class=3D"gmail_quote"><br><div><pre class=3D"gmail-newpage"=
 style=3D"font-size:13.3333px;margin-top:0px;margin-bottom:0px;break-before=
:page;color:rgb(0,0,0)">   A Proposed Standard specification is stable, has=
 resolved known
   design choices, has received significant community review, and
   appears to enjoy enough community interest to be considered valuable.

   Usually, neither implementation nor operational experience is
   required for the designation of a specification as a Proposed
   Standard.  However, such experience is highly desirable and will
   usually represent a strong argument in favor of a Proposed Standard
   designation.

   The IESG may require implementation and/or operational experience
   prior to granting Proposed Standard status to a specification that
   materially affects the core Internet protocols or that specifies
   behavior that may have significant operational impact on the
   Internet.

   A Proposed Standard will have no known technical omissions with
   respect to the requirements placed upon it.  Proposed Standards are
   of such quality that implementations can be deployed in the Internet.
   However, as with all technical specifications, Proposed Standards may
   be revised if problems are found or better solutions are identified,
   when experiences with deploying implementations of such technologies
   at scale is gathered.</pre></div><div>=C2=A0</div><div>It seems clear, I=
MHO, that CUBIC meets those criteria for being named a PS.</div><div><br></=
div><div>best regards,</div><div>neal</div><div><br></div><div><br></div></=
div></div>

--0000000000002db15305d914e1b9--


From nobody Mon Feb 28 11:01:14 2022
Return-Path: <iana-shared@icann.org>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 673943A130E for <tcpm@ietfa.amsl.com>; Mon, 28 Feb 2022 11:01:12 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.64
X-Spam-Level: 
X-Spam-Status: No, score=-0.64 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HEADER_FROM_DIFFERENT_DOMAINS=0.248, MISSING_HEADERS=1.021, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001] autolearn=no autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id TO2OssuDh8is for <tcpm@ietfa.amsl.com>; Mon, 28 Feb 2022 11:01:08 -0800 (PST)
Received: from smtp.lax.icann.org (smtp.lax.icann.org [IPv6:2620:0:2d0:201::1:81]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 63FF83A12EB for <tcpm@ietf.org>; Mon, 28 Feb 2022 11:01:08 -0800 (PST)
Received: from request4.lax.icann.org (request1.lax.icann.org [10.32.11.221]) by smtp.lax.icann.org (Postfix) with ESMTP id BCE25E04D8; Mon, 28 Feb 2022 19:01:06 +0000 (UTC)
Received: by request4.lax.icann.org (Postfix, from userid 48) id BA340206CB; Mon, 28 Feb 2022 19:01:06 +0000 (UTC)
RT-Owner: sabrina.tanamal
From: "Sabrina Tanamal via RT" <drafts-expert-review@iana.org>
Reply-To: drafts-expert-review@iana.org
In-Reply-To: <rt-4.4.3-31282-1646074716-1018.1226135-37-0@icann.org>
References: <RT-Ticket-1226135@icann.org> <rt-4.4.3-31282-1646074716-1018.1226135-37-0@icann.org>
Message-ID: <rt-4.4.3-31280-1646074866-959.1226135-37-0@icann.org>
X-RT-Loop-Prevention: IANA
X-RT-Ticket: IANA #1226135
X-Managed-BY: RT 4.4.3 (http://www.bestpractical.com/rt/)
X-RT-Originator: sabrina.tanamal@icann.org
CC: tbray@textuality.com, mt@lowentropy.net, tcpm@ietf.org
Content-Type: text/plain; charset="utf-8"
X-RT-Original-Encoding: utf-8
Precedence: bulk
Date: Mon, 28 Feb 2022 19:01:06 +0000
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/NaXmV8Pf6zkhFsqVelDZTF-6JPI>
Subject: [tcpm] [IANA #1226135] expert review for draft-ietf-tcpm-yang-tcp (xml-registry)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 28 Feb 2022 19:01:12 -0000

Tim and Martin (cc: tcpm WG), 

As the designated experts for the IETF XML ns registry, can you review the proposed registration in draft-ietf-tcpm-yang-tcp for us? Please see

https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/

If this is OK, when the IESG approves the document for publication, we'll make the registration at

https://www.iana.org/assignments/xml-registry

The IESG has asked us to set a two-week deadline for registration reviews. The due date for this request is March 14th. 

thanks,

Sabrina Tanamal
Lead IANA Services Specialist


From nobody Mon Feb 28 11:08:30 2022
Return-Path: <tbray@textuality.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C2EE73A13F6 for <tcpm@ietfa.amsl.com>; Mon, 28 Feb 2022 11:08:28 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.908
X-Spam-Level: 
X-Spam-Status: No, score=-1.908 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, HTML_MESSAGE=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=textuality-com.20210112.gappssmtp.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8CW0tbU_y747 for <tcpm@ietfa.amsl.com>; Mon, 28 Feb 2022 11:08:24 -0800 (PST)
Received: from mail-lf1-x12a.google.com (mail-lf1-x12a.google.com [IPv6:2a00:1450:4864:20::12a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1583B3A0A1D for <tcpm@ietf.org>; Mon, 28 Feb 2022 11:08:23 -0800 (PST)
Received: by mail-lf1-x12a.google.com with SMTP id j7so23054889lfu.6 for <tcpm@ietf.org>; Mon, 28 Feb 2022 11:08:23 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=textuality-com.20210112.gappssmtp.com; s=20210112; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=d59Vbq9SJOoqC+B7ja/7RXKYohfwO2YsnLtwfE9baHc=; b=Hgue6YtnVo4tiVcYfwbCiQg9s4e0WPf9oPqghgEQpGAlJQ0zLy2JQ9ts+IDt4VTin1 Tt8UVfT+aTcBMQsqNN+lrCcfXvqnbO42abelA4Jk0tQPs1YSG3GlN71OgKpTC6Z85SSw 1KCMUUWIgbHd44jlBJOtc9DF60ZoClctb23NKOZFlZrjuZFZNrJpx9pQdipWcmtW6xvv aCDbiB0rFQG7CGX4AhXw9onDgeQUubvmnX5q6NksvEL4Rtwvl5EsE+GSpurecPstwUEz 1W2pTMr5a8mp8ym1cGWEC12bsTD3e+7CohtCDe35ioWgPeeswVb1AmU6D3UT3CPpdwMh M3lw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=d59Vbq9SJOoqC+B7ja/7RXKYohfwO2YsnLtwfE9baHc=; b=TdpaFl48Ckda+kFwcCNwrydzyvdjdKTUpEFQJ0Lpenc0oQGGQizgADpVFXJwQy9ZPj bXthD6usj/ttyLFa2O49iKE/hkxw8R8dRK/hj+KLXjFhHFbcgTTsYeW3jjyp/iDnEAso tX13DoC6Z2yOw1nJt3pGZRU8p/arzPCe49n1012wFRFfAdIg8RhKuMLQ07QlQ/LFwPlt lQ6PtpgkeZdv3MZLJK5aLqoXOGPPgk9cwWh4joPInyxmxOZFKX88f4A9APZbLywfEfPz 7CxSbN4sXgsm8KcyZWD59yhZHbvZHE2/NF9xwGu/3fbcDvJaK8olSRcBug4EQ4gDaj7I 8AJA==
X-Gm-Message-State: AOAM532mezcpl/5T7LDl4unRAJlXIUawEn2zWNBJoHs2DJNKhhpe8H2S FzPI8SCs7CcZw2yenv5uORa1oPEFlpaD9qHRgb2r+hm9qGU=
X-Google-Smtp-Source: ABdhPJyZ5DkeSo9mrVriB8+O0TTqLUsXWSV/ISr62KtQboy2jCHl5NzFHgvIYgnH42L5VUN5WMJI5grMGI2UG8NwvEQ=
X-Received: by 2002:a05:6512:3b0f:b0:42d:1ebb:dcbf with SMTP id f15-20020a0565123b0f00b0042d1ebbdcbfmr13527737lfv.275.1646075301630; Mon, 28 Feb 2022 11:08:21 -0800 (PST)
MIME-Version: 1.0
References: <RT-Ticket-1226135@icann.org> <rt-4.4.3-31282-1646074716-1018.1226135-37-0@icann.org> <rt-4.4.3-31280-1646074866-959.1226135-37-0@icann.org>
In-Reply-To: <rt-4.4.3-31280-1646074866-959.1226135-37-0@icann.org>
From: Tim Bray <tbray@textuality.com>
Date: Mon, 28 Feb 2022 11:08:10 -0800
Message-ID: <CAHBU6iuCH_=ZuBfGH2oVtM+WAKGWYHMhZZ_qXby-j0qrEX0c1Q@mail.gmail.com>
To: Sabrina Tanamal via RT <drafts-expert-review@iana.org>
Cc: Martin Thomson <mt@lowentropy.net>, tcpm@ietf.org
Content-Type: multipart/alternative; boundary="000000000000a0b07b05d918c53c"
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/ukiVOXIn3JANqQvtkZ1j6BhvNbY>
Subject: Re: [tcpm] [IANA #1226135] expert review for draft-ietf-tcpm-yang-tcp (xml-registry)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 28 Feb 2022 19:08:29 -0000

--000000000000a0b07b05d918c53c
Content-Type: text/plain; charset="UTF-8"

Looks OK to me.

On Mon, Feb 28, 2022 at 11:01 AM Sabrina Tanamal via RT <
drafts-expert-review@iana.org> wrote:

> Tim and Martin (cc: tcpm WG),
>
> As the designated experts for the IETF XML ns registry, can you review the
> proposed registration in draft-ietf-tcpm-yang-tcp for us? Please see
>
> https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/
>
> If this is OK, when the IESG approves the document for publication, we'll
> make the registration at
>
> https://www.iana.org/assignments/xml-registry
>
> The IESG has asked us to set a two-week deadline for registration reviews.
> The due date for this request is March 14th.
>
> thanks,
>
> Sabrina Tanamal
> Lead IANA Services Specialist
>
>

--000000000000a0b07b05d918c53c
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_default" style=3D"font-size:small">Loo=
ks OK to me.=C2=A0</div></div><br><div class=3D"gmail_quote"><div dir=3D"lt=
r" class=3D"gmail_attr">On Mon, Feb 28, 2022 at 11:01 AM Sabrina Tanamal vi=
a RT &lt;<a href=3D"mailto:drafts-expert-review@iana.org">drafts-expert-rev=
iew@iana.org</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" styl=
e=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid=
;border-left-color:rgb(204,204,204);padding-left:1ex">Tim and Martin (cc: t=
cpm WG), <br>
<br>
As the designated experts for the IETF XML ns registry, can you review the =
proposed registration in draft-ietf-tcpm-yang-tcp for us? Please see<br>
<br>
<a href=3D"https://datatracker.ietf.org/doc/draft-ietf-tcpm-yang-tcp/" rel=
=3D"noreferrer" target=3D"_blank">https://datatracker.ietf.org/doc/draft-ie=
tf-tcpm-yang-tcp/</a><br>
<br>
If this is OK, when the IESG approves the document for publication, we&#39;=
ll make the registration at<br>
<br>
<a href=3D"https://www.iana.org/assignments/xml-registry" rel=3D"noreferrer=
" target=3D"_blank">https://www.iana.org/assignments/xml-registry</a><br>
<br>
The IESG has asked us to set a two-week deadline for registration reviews. =
The due date for this request is March 14th. <br>
<br>
thanks,<br>
<br>
Sabrina Tanamal<br>
Lead IANA Services Specialist<br>
<br>
</blockquote></div>

--000000000000a0b07b05d918c53c--


From nobody Mon Feb 28 11:23:35 2022
Return-Path: <vidhi_goel@apple.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 202BF3A143F; Mon, 28 Feb 2022 11:23:33 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.107
X-Spam-Level: 
X-Spam-Status: No, score=-2.107 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001] autolearn=unavailable autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=apple.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PvrF3VdmA1fc; Mon, 28 Feb 2022 11:23:31 -0800 (PST)
Received: from rn-mailsvcp-ppex-lapp15.apple.com (rn-mailsvcp-ppex-lapp15.rno.apple.com [17.179.253.34]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 060D53A1427; Mon, 28 Feb 2022 11:22:42 -0800 (PST)
Received: from pps.filterd (rn-mailsvcp-ppex-lapp15.rno.apple.com [127.0.0.1]) by rn-mailsvcp-ppex-lapp15.rno.apple.com (8.16.1.2/8.16.1.2) with SMTP id 21SJG5Lf014351; Mon, 28 Feb 2022 11:22:41 -0800
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=apple.com; h=from : message-id : content-type : mime-version : subject : date : in-reply-to : cc : to : references; s=20180706; bh=F/wiHOp5hfWcIbGUXhRwvPMG912yQbMTW2JUxnJP4iE=; b=ekkWBCsjVyTscEU7Vh46KNKFVDbdO9V/VAYijevrdN0/E7Zq5RIvFHFxpBHaaYKqLQin qTYZc0JRkDBCiefOBRN/pUl7PopuUTlnlyRVQVxegVUDs5UE2a7PVMIw4VEt7Y8sImvM gLBpRw8SPBiHEdYwbkHtuG+DscuUZAEn7BEeLpenG6+Bl+CnFM1VyAP2zNwGBcuqlqwN 3GfPXhKsa8iSrz2+IpAeJr89AciAcCeImkkhy8gX4Th/obm3Bv3mzzxKzmrmlfjerAUS JhuObJ2gzkL89LD7NliR+a4DAhONIXPAyYNJzn8YKdmNZSwY3MqkidzMuCJ3y97E/QO3 jg== 
Received: from rn-mailsvcp-mta-lapp03.rno.apple.com (rn-mailsvcp-mta-lapp03.rno.apple.com [10.225.203.151]) by rn-mailsvcp-ppex-lapp15.rno.apple.com with ESMTP id 3efjnfj4gf-2 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NO); Mon, 28 Feb 2022 11:22:41 -0800
Received: from rn-mailsvcp-mmp-lapp01.rno.apple.com (rn-mailsvcp-mmp-lapp01.rno.apple.com [17.179.253.14]) by rn-mailsvcp-mta-lapp03.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPS id <0R8100H5I4HR4WN0@rn-mailsvcp-mta-lapp03.rno.apple.com>;  Mon, 28 Feb 2022 11:22:39 -0800 (PST)
Received: from process_milters-daemon.rn-mailsvcp-mmp-lapp01.rno.apple.com by rn-mailsvcp-mmp-lapp01.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) id <0R81006004C2DL00@rn-mailsvcp-mmp-lapp01.rno.apple.com>; Mon, 28 Feb 2022 11:22:39 -0800 (PST)
X-Va-A: 
X-Va-T-CD: 49f532c9d10c5e683882ea4a58508f1c
X-Va-E-CD: b1f2128cd86a564e6a73a6f9cbd73762
X-Va-R-CD: 4463cdae966025dad18731a8d3192484
X-Va-CD: 0
X-Va-ID: 23c2d0d7-5b77-4a31-a6fa-ea418352328e
X-V-A: 
X-V-T-CD: 49f532c9d10c5e683882ea4a58508f1c
X-V-E-CD: b1f2128cd86a564e6a73a6f9cbd73762
X-V-R-CD: 4463cdae966025dad18731a8d3192484
X-V-CD: 0
X-V-ID: b1391145-73c3-40c6-bf98-6ed62fc1b863
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-28_08:2022-02-26, 2022-02-28 signatures=0
Received: from smtpclient.apple (unknown [17.234.120.69]) by rn-mailsvcp-mmp-lapp01.rno.apple.com (Oracle Communications Messaging Server 8.1.0.12.20210903 64bit (built Sep 3 2021)) with ESMTPSA id <0R8100JJF4HR4B00@rn-mailsvcp-mmp-lapp01.rno.apple.com>; Mon, 28 Feb 2022 11:22:39 -0800 (PST)
From: Vidhi Goel <vidhi_goel@apple.com>
Message-id: <C0D01F76-5519-49AF-8AEE-93A04E723F86@apple.com>
Content-type: multipart/signed; boundary="Apple-Mail=_B607C71F-A562-43AD-A583-8D608D203CF8"; protocol="application/pkcs7-signature"; micalg=sha-256
MIME-version: 1.0 (Mac OS X Mail 16.0 \(3696.80.82.1.1\))
Date: Mon, 28 Feb 2022 11:22:38 -0800
In-reply-to: <CADVnQykqf5zU-or7VbjcaU8ARTv1AOt0zR+xVquBG8ck5SqyRA@mail.gmail.com>
Cc: Yoshifumi Nishida <nsd.ietf@gmail.com>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
To: Neal Cardwell <ncardwell=40google.com@dmarc.ietf.org>
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <CAAK044Q5Lv175T6RcVp+XKEfqZ=Z_+TfjJ5Q9NzzA6wVerMkSw@mail.gmail.com> <CADVnQykqf5zU-or7VbjcaU8ARTv1AOt0zR+xVquBG8ck5SqyRA@mail.gmail.com>
X-Mailer: Apple Mail (2.3696.80.82.1.1)
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.425, 18.0.816 definitions=2022-02-28_08:2022-02-26, 2022-02-28 signatures=0
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/n6lIj5X7XYHvT6cnpvgBr8plun8>
Subject: Re: [tcpm] Proceeding 8312bis draft (Re: 2nd WGLC for draft-ietf-tcpm-rfc8312bis)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 28 Feb 2022 19:23:33 -0000

--Apple-Mail=_B607C71F-A562-43AD-A583-8D608D203CF8
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_479257F8-7F02-4F7D-8653-99171B75BBA8"


--Apple-Mail=_479257F8-7F02-4F7D-8653-99171B75BBA8
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

> I would argue for:
>=20
>   (2) publish the 8312bis draft as a PS doc, and=20
>=20
>   (2b) publish an additional doc that describes any aspects of CUBIC =
where the community would like to see further experiments or =
improvements.
>=20
> IMHO Michael Scharf made an excellent argument for PS status for CUBIC =
on Feb 22 ( =
https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5pQ6LrDE/ =
<https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5pQ6LrDE/> =
) by simply quoting a definition of PS from RFC 7127:
>=20

I agree with Neal=E2=80=99s suggestion based on PS status definition =
from the above RFC.


Thanks,
Vidhi

> On Feb 28, 2022, at 6:29 AM, Neal Cardwell =
<ncardwell=3D40google.com@dmarc.ietf.org> wrote:
>=20
>=20
>=20
> On Mon, Feb 28, 2022 at 5:07 AM Yoshifumi Nishida <nsd.ietf@gmail.com =
<mailto:nsd.ietf@gmail.com>> wrote:
> Hi,
>=20
> We've been discussing how to proceed 8312bis draft, but I think we =
still haven't settled yet.
> In my understanding, there seems to be the following options so far.
> if you have some thoughts on them or you have other options for this, =
please share. =20
> BTW, please note that this discussion might affect our future drafts =
because we want to apply the same bar for all standard docs as much as =
possible.=20
>=20
> 1: Publish the draft as a non-standard doc
>     a: publish as an informational doc
>        =20
>     b: publish as an experimental doc
>         * we might need to describe what experiments are expected.
>=20
> 2: Publish the draft as a PS doc
>     a: do additional experiments to make sure there's no concern as a =
PS doc.=20
>         * we will need to decide what kinds of experiments are =
required, what would be the expected results before the experiments.
>         * we need to delay the publication of the doc until the =
validation of the experiments has finished
>       =20
>     b: publish an additional doc that describes the checking points on =
the doc (and other related docs) for long term analysis. =20
>=20
> I would argue for:
>=20
>   (2) publish the 8312bis draft as a PS doc, and=20
>=20
>   (2b) publish an additional doc that describes any aspects of CUBIC =
where the community would like to see further experiments or =
improvements.
>=20
> IMHO Michael Scharf made an excellent argument for PS status for CUBIC =
on Feb 22 ( =
https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5pQ6LrDE/ =
<https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5pQ6LrDE/> =
) by simply quoting a definition of PS from RFC 7127:
>=20
>    A Proposed Standard specification is stable, has resolved known
>    design choices, has received significant community review, and
>    appears to enjoy enough community interest to be considered =
valuable.
>=20
>    Usually, neither implementation nor operational experience is
>    required for the designation of a specification as a Proposed
>    Standard.  However, such experience is highly desirable and will
>    usually represent a strong argument in favor of a Proposed Standard
>    designation.
>=20
>    The IESG may require implementation and/or operational experience
>    prior to granting Proposed Standard status to a specification that
>    materially affects the core Internet protocols or that specifies
>    behavior that may have significant operational impact on the
>    Internet.
>=20
>    A Proposed Standard will have no known technical omissions with
>    respect to the requirements placed upon it.  Proposed Standards are
>    of such quality that implementations can be deployed in the =
Internet.
>    However, as with all technical specifications, Proposed Standards =
may
>    be revised if problems are found or better solutions are =
identified,
>    when experiences with deploying implementations of such =
technologies
>    at scale is gathered.
> =20
> It seems clear, IMHO, that CUBIC meets those criteria for being named =
a PS.
>=20
> best regards,
> neal
>=20
>=20
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org <mailto:tcpm@ietf.org>
> https://www.ietf.org/mailman/listinfo/tcpm =
<https://www.ietf.org/mailman/listinfo/tcpm>

--Apple-Mail=_479257F8-7F02-4F7D-8653-99171B75BBA8
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=utf-8

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; =
charset=3Dutf-8"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; line-break: after-white-space;" =
class=3D""><blockquote type=3D"cite" class=3D""><div dir=3D"ltr" =
class=3D""><div class=3D"gmail_quote"><div class=3D"">I would argue =
for:</div></div></div></blockquote><blockquote type=3D"cite" =
class=3D""><div dir=3D"ltr" class=3D""><div class=3D"gmail_quote"><div =
class=3D""><br class=3D""></div><div class=3D"">&nbsp; (2) publish the =
8312bis draft as a PS doc, and&nbsp;</div><div class=3D""><br =
class=3D""></div><div class=3D"">&nbsp; (2b) publish an additional doc =
that describes any aspects of CUBIC where the community would like to =
see further experiments or improvements.</div><div class=3D""><br =
class=3D""></div>IMHO Michael Scharf made an excellent argument for PS =
status for CUBIC on Feb 22 (&nbsp;<a =
href=3D"https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5pQ6=
LrDE/" =
class=3D"">https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5=
pQ6LrDE/</a>&nbsp;) by simply quoting a definition of PS from RFC =
7127:</div><div class=3D""><br class=3D""></div></div></blockquote><div =
class=3D""><br class=3D""></div>I agree with Neal=E2=80=99s suggestion =
based on PS status definition from the above RFC.<div class=3D""><br =
class=3D""></div><div class=3D""><br class=3D""></div><div =
class=3D"">Thanks,</div><div class=3D"">Vidhi<br class=3D""><div><br =
class=3D""><blockquote type=3D"cite" class=3D""><div class=3D"">On Feb =
28, 2022, at 6:29 AM, Neal Cardwell &lt;<a =
href=3D"mailto:ncardwell=3D40google.com@dmarc.ietf.org" =
class=3D"">ncardwell=3D40google.com@dmarc.ietf.org</a>&gt; =
wrote:</div><br class=3D"Apple-interchange-newline"><div class=3D""><meta =
charset=3D"UTF-8" class=3D""><div dir=3D"ltr" style=3D"caret-color: =
rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: =
normal; font-variant-caps: normal; font-weight: 400; letter-spacing: =
normal; text-align: start; text-indent: 0px; text-transform: none; =
white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; =
text-decoration: none;" class=3D""><br =
class=3D"Apple-interchange-newline"><br class=3D""><div =
class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Mon, Feb =
28, 2022 at 5:07 AM Yoshifumi Nishida &lt;<a =
href=3D"mailto:nsd.ietf@gmail.com" class=3D"">nsd.ietf@gmail.com</a>&gt; =
wrote:<br class=3D""></div><blockquote class=3D"gmail_quote" =
style=3D"margin: 0px 0px 0px 0.8ex; border-left-width: 1px; =
border-left-style: solid; border-left-color: rgb(204, 204, 204); =
padding-left: 1ex;"><div dir=3D"ltr" class=3D""><div dir=3D"ltr" =
class=3D"">Hi,<br class=3D""></div><div dir=3D"ltr" class=3D""><br =
class=3D""></div><div class=3D"">We've been discussing how to proceed =
8312bis draft, but I think we still haven't settled yet.</div><div =
class=3D"">In my understanding, there seems to be the following options =
so far.</div><div class=3D"">if you have some thoughts on them or you =
have other options for this, please share.&nbsp;&nbsp;</div><div =
class=3D"">BTW, please note that this discussion might affect our future =
drafts because we want to apply the same bar for all standard docs as =
much as possible.&nbsp;<br class=3D""></div><div class=3D""><br =
class=3D""></div><div class=3D"">1: Publish the draft as a non-standard =
doc</div><div class=3D"">&nbsp; &nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>a: publish as an =
informational doc</div><div class=3D"">&nbsp; &nbsp; &nbsp; =
&nbsp;&nbsp;</div><div class=3D"">&nbsp; &nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>b: publish as an =
experimental doc</div><div class=3D"">&nbsp; &nbsp; &nbsp; &nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>* we might need to describe =
what experiments are expected.</div><div class=3D""><br =
class=3D""></div><div class=3D"">2: Publish the draft as a PS =
doc</div><div class=3D"">&nbsp; &nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>a: do =
additional&nbsp;experiments to make sure there's no concern as a PS =
doc.&nbsp;</div><div class=3D"">&nbsp; &nbsp; &nbsp; &nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>* we will need to decide =
what kinds of experiments are required, what would be the expected =
results before the experiments.</div><div class=3D"">&nbsp; &nbsp; =
&nbsp; &nbsp;<span class=3D"Apple-converted-space">&nbsp;</span>* we =
need to delay the publication of the doc until the validation of the =
experiments has finished</div><div class=3D"">&nbsp; &nbsp; &nbsp; =
&nbsp;</div><div class=3D"">&nbsp; &nbsp;<span =
class=3D"Apple-converted-space">&nbsp;</span>b: publish an =
additional&nbsp;doc that describes the checking points on the doc (and =
other related docs) for long term =
analysis.&nbsp;&nbsp;</div></div></blockquote><div class=3D""><br =
class=3D""></div><div class=3D"">I would argue for:</div><div =
class=3D""><br class=3D""></div><div class=3D"">&nbsp; (2) publish the =
8312bis draft as a PS doc, and&nbsp;</div><div class=3D""><br =
class=3D""></div><div class=3D"">&nbsp; (2b) publish an additional doc =
that describes any aspects of CUBIC where the community would like to =
see further experiments or improvements.</div><div class=3D""><br =
class=3D""></div>IMHO Michael Scharf made an excellent argument for PS =
status for CUBIC on Feb 22 (<span =
class=3D"Apple-converted-space">&nbsp;</span><a =
href=3D"https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5pQ6=
LrDE/" =
class=3D"">https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5=
pQ6LrDE/</a><span class=3D"Apple-converted-space">&nbsp;</span>) by =
simply quoting a definition of PS from RFC 7127:</div><div =
class=3D"gmail_quote"><br class=3D""><div class=3D""><pre =
class=3D"gmail-newpage" style=3D"font-size: 13.3333px; margin-top: 0px; =
margin-bottom: 0px; break-before: page;">   A Proposed Standard =
specification is stable, has resolved known
   design choices, has received significant community review, and
   appears to enjoy enough community interest to be considered valuable.

   Usually, neither implementation nor operational experience is
   required for the designation of a specification as a Proposed
   Standard.  However, such experience is highly desirable and will
   usually represent a strong argument in favor of a Proposed Standard
   designation.

   The IESG may require implementation and/or operational experience
   prior to granting Proposed Standard status to a specification that
   materially affects the core Internet protocols or that specifies
   behavior that may have significant operational impact on the
   Internet.

   A Proposed Standard will have no known technical omissions with
   respect to the requirements placed upon it.  Proposed Standards are
   of such quality that implementations can be deployed in the Internet.
   However, as with all technical specifications, Proposed Standards may
   be revised if problems are found or better solutions are identified,
   when experiences with deploying implementations of such technologies
   at scale is gathered.</pre></div><div class=3D"">&nbsp;</div><div =
class=3D"">It seems clear, IMHO, that CUBIC meets those criteria for =
being named a PS.</div><div class=3D""><br class=3D""></div><div =
class=3D"">best regards,</div><div class=3D"">neal</div><div =
class=3D""><br class=3D""></div><div class=3D""><br =
class=3D""></div></div></div><span style=3D"caret-color: rgb(0, 0, 0); =
font-family: Helvetica; font-size: 12px; font-style: normal; =
font-variant-caps: normal; font-weight: 400; letter-spacing: normal; =
text-align: start; text-indent: 0px; text-transform: none; white-space: =
normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; =
text-decoration: none; float: none; display: inline !important;" =
class=3D"">_______________________________________________</span><br =
style=3D"caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: =
12px; font-style: normal; font-variant-caps: normal; font-weight: 400; =
letter-spacing: normal; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; word-spacing: 0px; =
-webkit-text-stroke-width: 0px; text-decoration: none;" class=3D""><span =
style=3D"caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: =
12px; font-style: normal; font-variant-caps: normal; font-weight: 400; =
letter-spacing: normal; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; word-spacing: 0px; =
-webkit-text-stroke-width: 0px; text-decoration: none; float: none; =
display: inline !important;" class=3D"">tcpm mailing list</span><br =
style=3D"caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: =
12px; font-style: normal; font-variant-caps: normal; font-weight: 400; =
letter-spacing: normal; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; word-spacing: 0px; =
-webkit-text-stroke-width: 0px; text-decoration: none;" class=3D""><a =
href=3D"mailto:tcpm@ietf.org" style=3D"font-family: Helvetica; =
font-size: 12px; font-style: normal; font-variant-caps: normal; =
font-weight: 400; letter-spacing: normal; orphans: auto; text-align: =
start; text-indent: 0px; text-transform: none; white-space: normal; =
widows: auto; word-spacing: 0px; -webkit-text-size-adjust: auto; =
-webkit-text-stroke-width: 0px;" class=3D"">tcpm@ietf.org</a><br =
style=3D"caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: =
12px; font-style: normal; font-variant-caps: normal; font-weight: 400; =
letter-spacing: normal; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; word-spacing: 0px; =
-webkit-text-stroke-width: 0px; text-decoration: none;" class=3D""><a =
href=3D"https://www.ietf.org/mailman/listinfo/tcpm" style=3D"font-family: =
Helvetica; font-size: 12px; font-style: normal; font-variant-caps: =
normal; font-weight: 400; letter-spacing: normal; orphans: auto; =
text-align: start; text-indent: 0px; text-transform: none; white-space: =
normal; widows: auto; word-spacing: 0px; -webkit-text-size-adjust: auto; =
-webkit-text-stroke-width: 0px;" =
class=3D"">https://www.ietf.org/mailman/listinfo/tcpm</a></div></blockquot=
e></div><br class=3D""></div></body></html>=

--Apple-Mail=_479257F8-7F02-4F7D-8653-99171B75BBA8--

--Apple-Mail=_B607C71F-A562-43AD-A583-8D608D203CF8
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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=
--Apple-Mail=_B607C71F-A562-43AD-A583-8D608D203CF8--


From nobody Mon Feb 28 12:07:58 2022
Return-Path: <ycheng@google.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BECE93A1308 for <tcpm@ietfa.amsl.com>; Mon, 28 Feb 2022 12:07:56 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -17.61
X-Spam-Level: 
X-Spam-Status: No, score=-17.61 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_MED=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, ENV_AND_HDR_SPF_MATCH=-0.5, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001, USER_IN_DEF_DKIM_WL=-7.5, USER_IN_DEF_SPF_WL=-7.5] autolearn=unavailable autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=google.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XTTCE8jxBxVd for <tcpm@ietfa.amsl.com>; Mon, 28 Feb 2022 12:07:53 -0800 (PST)
Received: from mail-wr1-x42a.google.com (mail-wr1-x42a.google.com [IPv6:2a00:1450:4864:20::42a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9256A3A1304 for <tcpm@ietf.org>; Mon, 28 Feb 2022 12:07:53 -0800 (PST)
Received: by mail-wr1-x42a.google.com with SMTP id d17so17158247wrc.9 for <tcpm@ietf.org>; Mon, 28 Feb 2022 12:07:53 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc:content-transfer-encoding; bh=r7a9p1iH7r672DzK9pbHKecNgcCJcqBdhP1X9FV3q0A=; b=qhdEJcILmy4QekaHhFBa5te4QxB2rE86LNTXIp6sW5kELXLm41Uszrk03xiIQkqbWr n2N3+5vWhzme9swyjZFOJ4QvrmFvWL1W3cT97/JsHdH98OcznkPgXVTU9Bz+EiDozk7P R/GfTLKyhhz7z950YXaStaCSAxfy3xTNRk6uyDsVWQR2FmrlRb6IMdjsDOKixYvh/mu3 SsikL+pjCti4l/PIKY4rcIMi2Lub8ruefFTNOdje/l4+APURr87uZklKS/ObfCA5aLEx sNmbMENF3krDUbtX95sILPQcEmQRbnXvI/2X/lIKOE369VtdYPJ2YISQWSL7LVTTZnfO dqYw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc:content-transfer-encoding; bh=r7a9p1iH7r672DzK9pbHKecNgcCJcqBdhP1X9FV3q0A=; b=1Mk95P9aRafY5w75pnVEVXUaE2sfZKC5rJaz5lYjuaR6mUO84CMP4qlZ9SzlNRnUiT nPQT0HRlygbudbJeZECJKcExsbHFqsep89Hax4cea7fn6gShUf63Bo3r+PUXBUvAHs2t YdqsU5xAKRkGD0Q8Bgd4GsVyJpesOm5ldCUHeSU8xwGsXBe8dxr5Hj/q8l//LRoEdTeR t6v2sX/1WN9wMRUD7wOqmHvLVltZ8oTR2b3R/XXZ2IkK9GW0rCmihaeO+m6sYL5oECXw 3rnnrQ44ZN6mnHQYczowql6L06WgV08fhs8aITlVJLnCFdgI4ct2cI4Gb0SVtrXc489J P3sA==
X-Gm-Message-State: AOAM530vw7UKqpK/y0l5we2HkD1UGOvgW/mcTcUW19pC+C2YOdUXbrXo bNjwG8TuysMC2gRUFyzHsYFpyk1g+R+01kY/VfOGYeKvtDuGsQ==
X-Google-Smtp-Source: ABdhPJzU6562HdNu3PJXgn9SmNxG3BVKiAFlcsOqEr86GHdIYknAWMVDRBfT/GAMTt9Ym/G2u9T2ZSjv0HkHimvHwKM=
X-Received: by 2002:adf:d212:0:b0:1ef:6e7a:b04c with SMTP id j18-20020adfd212000000b001ef6e7ab04cmr12620858wrh.412.1646078871445; Mon, 28 Feb 2022 12:07:51 -0800 (PST)
MIME-Version: 1.0
References: <164318837039.21788.17451980682651967578@ietfa.amsl.com> <EEA435EC-AAAC-4899-8E94-2D54EDE5F72E@eggert.org> <CAAK044S9HQXvfvgM6mBuvOWJPHtCaa6xo6CoP2r8Vq61tKaY5g@mail.gmail.com> <CAAK044Q5Lv175T6RcVp+XKEfqZ=Z_+TfjJ5Q9NzzA6wVerMkSw@mail.gmail.com> <CADVnQykqf5zU-or7VbjcaU8ARTv1AOt0zR+xVquBG8ck5SqyRA@mail.gmail.com> <C0D01F76-5519-49AF-8AEE-93A04E723F86@apple.com>
In-Reply-To: <C0D01F76-5519-49AF-8AEE-93A04E723F86@apple.com>
From: Yuchung Cheng <ycheng@google.com>
Date: Mon, 28 Feb 2022 12:07:14 -0800
Message-ID: <CAK6E8=dkJrw0Vv6du-CMASjydSTXxzDXM4z=O191R4t+NeE7dg@mail.gmail.com>
To: Vidhi Goel <vidhi_goel=40apple.com@dmarc.ietf.org>
Cc: Neal Cardwell <ncardwell=40google.com@dmarc.ietf.org>,  "tcpm@ietf.org Extensions" <tcpm@ietf.org>, tcpm-chairs <tcpm-chairs@ietf.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/GJdxuMlogsEwB2HaYPpHob9Q7b0>
Subject: Re: [tcpm] Proceeding 8312bis draft (Re: 2nd WGLC for draft-ietf-tcpm-rfc8312bis)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 28 Feb 2022 20:07:57 -0000

+1 to Neal and Vidhi's comments.

On Mon, Feb 28, 2022 at 11:23 AM Vidhi Goel
<vidhi_goel=3D40apple.com@dmarc.ietf.org> wrote:
>
> I would argue for:
>
>
>   (2) publish the 8312bis draft as a PS doc, and
>
>   (2b) publish an additional doc that describes any aspects of CUBIC wher=
e the community would like to see further experiments or improvements.
>
> IMHO Michael Scharf made an excellent argument for PS status for CUBIC on=
 Feb 22 ( https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5pQ=
6LrDE/ ) by simply quoting a definition of PS from RFC 7127:
>
>
> I agree with Neal=E2=80=99s suggestion based on PS status definition from=
 the above RFC.
>
>
> Thanks,
> Vidhi
>
> On Feb 28, 2022, at 6:29 AM, Neal Cardwell <ncardwell=3D40google.com@dmar=
c.ietf.org> wrote:
>
>
>
> On Mon, Feb 28, 2022 at 5:07 AM Yoshifumi Nishida <nsd.ietf@gmail.com> wr=
ote:
>>
>> Hi,
>>
>> We've been discussing how to proceed 8312bis draft, but I think we still=
 haven't settled yet.
>> In my understanding, there seems to be the following options so far.
>> if you have some thoughts on them or you have other options for this, pl=
ease share.
>> BTW, please note that this discussion might affect our future drafts bec=
ause we want to apply the same bar for all standard docs as much as possibl=
e.
>>
>> 1: Publish the draft as a non-standard doc
>>     a: publish as an informational doc
>>
>>     b: publish as an experimental doc
>>         * we might need to describe what experiments are expected.
>>
>> 2: Publish the draft as a PS doc
>>     a: do additional experiments to make sure there's no concern as a PS=
 doc.
>>         * we will need to decide what kinds of experiments are required,=
 what would be the expected results before the experiments.
>>         * we need to delay the publication of the doc until the validati=
on of the experiments has finished
>>
>>     b: publish an additional doc that describes the checking points on t=
he doc (and other related docs) for long term analysis.
>
>
> I would argue for:
>
>   (2) publish the 8312bis draft as a PS doc, and
>
>   (2b) publish an additional doc that describes any aspects of CUBIC wher=
e the community would like to see further experiments or improvements.
>
> IMHO Michael Scharf made an excellent argument for PS status for CUBIC on=
 Feb 22 ( https://mailarchive.ietf.org/arch/msg/tcpm/k47hlSsx1lRupy5LRTJ5pQ=
6LrDE/ ) by simply quoting a definition of PS from RFC 7127:
>
>    A Proposed Standard specification is stable, has resolved known
>    design choices, has received significant community review, and
>    appears to enjoy enough community interest to be considered valuable.
>
>    Usually, neither implementation nor operational experience is
>    required for the designation of a specification as a Proposed
>    Standard.  However, such experience is highly desirable and will
>    usually represent a strong argument in favor of a Proposed Standard
>    designation.
>
>    The IESG may require implementation and/or operational experience
>    prior to granting Proposed Standard status to a specification that
>    materially affects the core Internet protocols or that specifies
>    behavior that may have significant operational impact on the
>    Internet.
>
>    A Proposed Standard will have no known technical omissions with
>    respect to the requirements placed upon it.  Proposed Standards are
>    of such quality that implementations can be deployed in the Internet.
>    However, as with all technical specifications, Proposed Standards may
>    be revised if problems are found or better solutions are identified,
>    when experiences with deploying implementations of such technologies
>    at scale is gathered.
>
>
> It seems clear, IMHO, that CUBIC meets those criteria for being named a P=
S.
>
> best regards,
> neal
>
>
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm
>
>
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm


From nobody Mon Feb 28 18:52:50 2022
Return-Path: <wes@mti-systems.com>
X-Original-To: tcpm@ietfa.amsl.com
Delivered-To: tcpm@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C3D233A00E4 for <tcpm@ietfa.amsl.com>; Mon, 28 Feb 2022 18:52:41 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.906
X-Spam-Level: 
X-Spam-Status: No, score=-1.906 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, HTML_MESSAGE=0.001, NICE_REPLY_A=-0.001, RCVD_IN_DNSWL_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_NONE=0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001] autolearn=unavailable autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=mti-systems-com.20210112.gappssmtp.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zc_sonbn4ayP for <tcpm@ietfa.amsl.com>; Mon, 28 Feb 2022 18:52:36 -0800 (PST)
Received: from mail-qv1-xf36.google.com (mail-qv1-xf36.google.com [IPv6:2607:f8b0:4864:20::f36]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 25C833A0E03 for <tcpm@ietf.org>; Mon, 28 Feb 2022 18:52:35 -0800 (PST)
Received: by mail-qv1-xf36.google.com with SMTP id fc19so15693282qvb.7 for <tcpm@ietf.org>; Mon, 28 Feb 2022 18:52:35 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mti-systems-com.20210112.gappssmtp.com; s=20210112; h=message-id:date:mime-version:user-agent:subject:content-language:to :cc:references:from:in-reply-to; bh=oFyHolV+ejP6K32wNHrkyf+rFNGdTfoM1xkVJ11B+W4=; b=oZTlpTGd4fuvb9UBflHQPwlB4of97mB6jI7VSZ0K4F3QRMMvbKx6n5j3puQrVmsSDT ulfdOReMw5Vc6QTCs2qYY5JE2fD1yATHgJ4wk59j+XMBMF/7WLUuQ5dStn0pRx2r4UxD jpqEUT+iENuTZ0Y5g+yDX2NNLXl5N5xPny4wgVzrDEYpFLFjIgudMoYkTedo5uVdBQOa D3RrADGDRua4qcXHjn/uOSrWSmUe8W6oTUKoupZvP2aJrjma6ffx8/mOc2GiadbDnH5c cmOnTOXJTa1twTphdRbpSyIpLW1IHJtdm7ssTzx0o2uo94DWRLknGN+LJWGk4b8jz9RC miyg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:message-id:date:mime-version:user-agent:subject :content-language:to:cc:references:from:in-reply-to; bh=oFyHolV+ejP6K32wNHrkyf+rFNGdTfoM1xkVJ11B+W4=; b=RgBNjjkUiRHfTPw18XsDRhuMwBLWXyDTmXvYZ21FS090epSZMY0gAcW3iULLmAVqlx 00DAUUDDXjCLxdrRSkV4+jQJwUq1EyBLnS+mPtMMzGjUzfKLOUMhPIhnQZKlZ+RdP2T0 dkxRozM7kXSMwfu3n4dZKGKy415O1cpzKavy8QwjOXdNdg15VRYTh6asnt+t7Ijbvp3x JI2QfXszw79jMse5oqUHFwf6C3WPMRbxxwxAE9D8KMMXN8vmUmEl372sR1lJBu1gCZiB fQo+gO305qvTwIP6bqjqrFQJQlNJB7eWBRbv+f964zhs55dL1cde9qGnYBXBqhjo7MLv ULMw==
X-Gm-Message-State: AOAM533jn0MyVkbDiFP2bdy2+y+z0ASWHpa0Xkzpyvg8MYuMnmjnQPPy ZC2nNsmh00AlI9ckGYQQuiOvvlo5aKko8Q==
X-Google-Smtp-Source: ABdhPJx+pjRoYs9cQHC0qJ09rNxUiUfe2IANerv3wrYXXIa0k0cGbdK609N2/9cObsIkRppOisBcAA==
X-Received: by 2002:a05:6214:c82:b0:433:7bd:9702 with SMTP id r2-20020a0562140c8200b0043307bd9702mr7347925qvr.8.1646103154575;  Mon, 28 Feb 2022 18:52:34 -0800 (PST)
Received: from [192.168.1.18] (cpe-66-61-72-87.neo.res.rr.com. [66.61.72.87]) by smtp.gmail.com with ESMTPSA id u3-20020a05622a010300b002dd22803f20sm8067538qtw.46.2022.02.28.18.52.32 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Mon, 28 Feb 2022 18:52:33 -0800 (PST)
Content-Type: multipart/alternative; boundary="------------GpwO2dZKownFYRaflUxcDxxb"
Message-ID: <96c51cc2-512e-f126-1022-b84ce896d08b@mti-systems.com>
Date: Mon, 28 Feb 2022 21:52:29 -0500
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Thunderbird/91.6.1
Content-Language: en-US
To: Benjamin Kaduk <kaduk@mit.edu>
Cc: The IESG <iesg@ietf.org>, draft-ietf-tcpm-rfc793bis@ietf.org, tcpm-chairs@ietf.org, tcpm@ietf.org, Michael Scharf <michael.scharf@hs-esslingen.de>
References: <163236803976.28405.5643771942452620510@ietfa.amsl.com> <6f6e7b90-081a-74b4-b329-8879addcb8c4@mti-systems.com> <20220225210031.GV12881@kduck.mit.edu>
From: Wesley Eddy <wes@mti-systems.com>
In-Reply-To: <20220225210031.GV12881@kduck.mit.edu>
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/mGqVLElRS3g56gjSsVyJZiNF-I8>
Subject: Re: [tcpm] Benjamin Kaduk's Discuss on draft-ietf-tcpm-rfc793bis-25: (with DISCUSS and COMMENT)
X-BeenThere: tcpm@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpm>, <mailto:tcpm-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm/>
List-Post: <mailto:tcpm@ietf.org>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpm>, <mailto:tcpm-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 01 Mar 2022 02:52:42 -0000

This is a multi-part message in MIME format.
--------------GpwO2dZKownFYRaflUxcDxxb
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

First, on the one DISCUSS point remaining:

On 2/25/2022 4:00 PM, Benjamin Kaduk wrote:

> In essence, I think that we require a fairly strong justification to
> publish an Internet Standard in 2022 that says it's okay to adopt a data
> model where a host has a global piece of state that it freely sends to
> anyone who asks, where that piece of state can be used to attack/disrupt
> all new connections that host makes, as opposed to just connections on the
> 5-tuple that asked.

I agree with Joe's explanation on the applicability of the concern being 
a bit more nuanced, since it's important for some things (like BGP 
sessions, which 5961 was written in response to), but less so for 
shorter-lived connections, hosts that aren't servers, etc.


> The actual scope of utility of an ISN is local to an individual 5-tuple,
> not global to a host, and false sharing of ISNs across connections adds
> risk.  My stance as SEC AD is that the IETF should produce protocols that
> are as secure as possible *subject to any constraints on them*.  Here, we
> have the constraint of retaining compatibility with the existing deployment
> base (which is a very compelling constraint!), so you do not see me
> advocating for mandatory tcpcrypt (for example).  But I want to see some
> explanation of what harm or risk there is in saying that the ISN is always
> produced with the PRF of the 5-tuple (okay, 4-tuple since the protocol is
> always TCP), to motivate a divergence from the more-secure behavior and
> thereby justify retaining the behavior currently in the draft.  What
> constraint are we subject to that prevents doing the right (security-wise)
> thing?
>
I think that the WG would have to decide on this, since it's more than 
just an editorial matter.

A fair change might be to leave SHOULD, but be more clear about why it 
isn't felt to justify a MUST presently, if that seems to make sense to you?


And now, closing the loop on a few COMMENTs:

>
> Section 3.8.4
>
>      An implementation SHOULD send a keep-alive segment with no data
>      (SHLD-12); however, it MAY be configurable to send a keep-alive
>      segment containing one garbage octet (MAY-6), for compatibility with
>      erroneous TCP implementations.
>
> Such misbehaved TCP impelementations were misbehaved even in 1989 when
> RFC 1122 was published -- do we have a sense for whether they are still
> around to any significant degree?
>> That's a great question; I don't know, but would highly doubt they are
>> around anymore (at least not in production on the Internet).
> Might be an interesting experiment, but it sounds like we don't have enough
> data to justify changing the text here (yet).

As a follow-up action, we can bring this to attention in TCPM and see if 
anyone is interested in helping to gather data on whether any 
implementations still need this.  That would then let us see if this can 
be removed in an update.  But anyways, probably shouldn't block this doc 
(which I think you agree with, since this was just a COMMENT, not 
DISCUSS topic).


>>> Section 4
>>>
>>>      Destination Address
>>>              The network layer address of the remote endpoint.
>>>      [...]
>>>      Source Address
>>>              The network layer address of the sending endpoint.
>>>
>>> These definitions don't seem to work in the context of a receiver
>>> validating the TCP checksum, where the destination address is the local
>>> endpoint's address and the source address is the remote endpoint's
>>> address.  (I note that these definitions are different from what RFC 793
>>> itself used.)
>> Interesting point on the destination (though the source one seems fine,
>> it's always the sending endpoint).  Maybe we should just change "remote"
>> to "receiving" in the destination one?
> That's an easy fix; good idea.
>
In rev -27, I changed this to:

    Destination Address
            The network layer address of the endpoint intended to receive
            a segment.



>>>      The more secure Initial Sequence Number generation algorithm from RFC
>>>      6528 was incorporated.  See RFC 6528 for discussion of the attacks
>>>      that this mitigates, as well as advice on selecting PRF algorithms
>>>      and managing secret key data.
>>>
>>> (As I mentioned up in §3.4, that guidance is no longer current.)
>> Is updating 6528 agreed to be possible future work?
> Yes.  (Is this something where I should be expecting to take the action
> item to write a draft, or do you think someone else might take up the pen?)

It might be worthwhile to succinctly describe the proposed update to 
TCPM, and see what the interest level is.  If you have bandwidth to do 
so, you'd probably be best suited, but otherwise, I'd also be happy to 
bring it up on-list and at the next meeting (per chairs approval), as 
part of closing-out this 793bis work.



>>> Section 3.8.6.3
>>>
>>>      Note that there are several current practices that further lead to a
>>>      reduced number of ACKs, including generic receive offload (GRO), ACK
>>>      compression, and ACK decimation [26].
>>>
>>> Reference [26] seems reasonable for ACK decimation and ACK compression,
>>> but doesn't seem to cover GRO at all.
>> True; do you think we should try to find a proper GRO reference? I'm not
>> aware of there being a real standard or academic type of reference.
> I wouldn't put much effort into looking.
> It's a bit outside my normal field, so I don't have any immediate insight.
> Google finds the DPDK docs, which might be stable and well-known, and
> wikipedia'shttps://en.wikipedia.org/wiki/TCP_offload_engine  entry has a
> brief mention, with links to an LWN article and
> https://books.google.com/books?id=C3wQBwAAQBAJ  .  If we know someone with
> a copy of the latter, it might be worth checking if it would be a useful
> reference.

This one looks good to me, and I can include it in the next revision, if 
you agree:

https://www.kernel.org/doc/html/latest/networking/segmentation-offloads.html



>>> Section 3.9.2.2
>>>
>>>      Soft Errors
>>>        For ICMP these include: Destination Unreachable -- codes 0, 1, 5,
>>>        Time Exceeded -- codes 0, 1, and Parameter Problem.
>>>
>>>        For ICMPv6 these include: Destination Unreachable -- codes 0 and 3,
>>>        Time Exceeded -- codes 0, 1, and Parameter Problem -- codes 0, 1,
>>>        2.
>>>
>>>        Since these Unreachable messages indicate soft error conditions,
>>>
>>> I'm not entirely sure that I'd classify "parameter problem" as an
>>> "unreachable" message per se.
>> I'm not sure I understand this comment.  This is the list from RFC 5461.
> It looks like I was misparsing on first read, since there were commas
> serving different roles.  ("Parameter Problem" is an element in the outer
> list, matched with "Destination Unreachable" and "Time Exceeded", not one
> of the codes for Time Exceeded.)  Using semicolons as delimiters for the
> "outer" list, with commas for the lists of type codes, would help
> distinguish them.
>
I see; semicolons will be in the next revision.



>>> Section 3.10.1
>>>
>>>            the parameters of the incoming SYN segment.  Verify the
>>>            security and DiffServ value requested are allowed for this
>>>            user, if not return "error: precedence not allowed" or "error:
>>>            security/compartment not allowed."  If passive enter the LISTEN
>>>
>>> It's surprising for the error string to mention "precedence" when the
>>> predicate is DiffServ value.
>> This is a good point.  When the definition of those bits was changed, it
>> doesn't seem like the error message indicated here was correspondingly
>> changed.
> Is this something we need to leave alone for backwards compatibility, or do
> we have leeway to change the error message here?  (It looks like this text
> is unchanged in the -27.)

I don't think there's much of a compatibility issue, since the socket 
API works differently than this anyways.  We could change "precedence" 
to "DSCP".


>>>               *  If this connection was initiated with a passive OPEN
>>>                  (i.e., came from the LISTEN state), then return this
>>>                  connection to LISTEN state and return.  The user need
>>>                  not be informed.  If this connection was initiated
>>>                  with an active OPEN (i.e., came from SYN-SENT state)
>>>                  then the connection was refused, signal the user
>>>                  "connection refused".  In either case, all segments on
>>>                  the retransmission queue should be removed.  And in
>>>
>>> IIUC, what's described here as "removed" is described elsewhere as
>>> "flushed"; it would be good to use consistent terminology when possible.
>> Do you have a strong preference?
> No strong preference, just a general desire to have consistent terminology
> unless there is a difference in meaning intended.
>
>
Ok, I can change this to say that the retransmission queue should be 
flushed in the next revision.



>>> Section 4
>>>
>>>      internet datagram
>>>              The unit of data exchanged between an internet module and the
>>>              higher level protocol together with the internet header.
>>>
>>> "exchanged between an internet module and the higher level protocol"
>>> sounds like a local operation; I would have expected the definition of
>>> an *internet* datagram to involve transfer over the (inter)network.
>> Ok, can you suggest an alternative definition?  I agree this one from
>> the original 793 isn't terrific.
> I'm somewhat reluctant to concoct a new definition from scratch at this
> point in the process.  I see that RFC 1983 has "IP datagram" refer to just
> "datagram", on which it says:
>
>     datagram
>        A self-contained, independent entity of data carrying sufficient
>        information to be routed from the source to the destination
>        computer without reliance on earlier exchanges between this source
>        and destination computer and the transporting network.  See also:
>        frame, packet.
>        [Source: J. Postel]
>
> If I was to overcome my reluctance and try to revise the existing text here
> into something that makes more sense, it would be:
>
> internet datagram
>    A unit of data [from some higher layer protocol] exchanged between
>    internet hosts, together with the internet header that allows the
>    datagram to be routed from source to destination.
>
> (I'm not sure whether I'd include the bit in square brackets.)
> Feel free to use mine if you like, or go with Postel's (per RFC 1983) or
> something else.
Yours looks good to me, and I'll queue it for the next revision.

--------------GpwO2dZKownFYRaflUxcDxxb
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <div class="moz-cite-prefix">First, on the one DISCUSS point
      remaining:</div>
    <div class="moz-cite-prefix"><br>
    </div>
    <div class="moz-cite-prefix">On 2/25/2022 4:00 PM, Benjamin Kaduk
      wrote:
    </div>
    <p>
    </p>
    <blockquote type="cite"
      cite="mid:20220225210031.GV12881@kduck.mit.edu">
      <pre class="moz-quote-pre" wrap="">In essence, I think that we require a fairly strong justification to
publish an Internet Standard in 2022 that says it's okay to adopt a data
model where a host has a global piece of state that it freely sends to
anyone who asks, where that piece of state can be used to attack/disrupt
all new connections that host makes, as opposed to just connections on the
5-tuple that asked.</pre>
    </blockquote>
    <p>I agree with Joe's explanation on the applicability of the
      concern being a bit more nuanced, since it's important for some
      things (like BGP sessions, which 5961 was written in response to),
      but less so for shorter-lived connections, hosts that aren't
      servers, etc.</p>
    <p><br>
    </p>
    <p>
    </p>
    <blockquote type="cite"
      cite="mid:20220225210031.GV12881@kduck.mit.edu">
      <pre class="moz-quote-pre" wrap="">
The actual scope of utility of an ISN is local to an individual 5-tuple,
not global to a host, and false sharing of ISNs across connections adds
risk.  My stance as SEC AD is that the IETF should produce protocols that
are as secure as possible *subject to any constraints on them*.  Here, we
have the constraint of retaining compatibility with the existing deployment
base (which is a very compelling constraint!), so you do not see me
advocating for mandatory tcpcrypt (for example).  But I want to see some
explanation of what harm or risk there is in saying that the ISN is always
produced with the PRF of the 5-tuple (okay, 4-tuple since the protocol is
always TCP), to motivate a divergence from the more-secure behavior and
thereby justify retaining the behavior currently in the draft.  What
constraint are we subject to that prevents doing the right (security-wise)
thing?

</pre>
    </blockquote>
    <p>I think that the WG would have to decide on this, since it's more
      than just an editorial matter.</p>
    <p>A fair change might be to leave SHOULD, but be more clear about
      why it isn't felt to justify a MUST presently, if that seems to
      make sense to you?<br>
    </p>
    <p><br>
    </p>
    <p>And now, closing the loop on a few COMMENTs:<br>
    </p>
    <blockquote type="cite"
      cite="mid:20220225210031.GV12881@kduck.mit.edu">
      <pre class="moz-quote-pre" wrap="">

Section 3.8.4

    An implementation SHOULD send a keep-alive segment with no data
    (SHLD-12); however, it MAY be configurable to send a keep-alive
    segment containing one garbage octet (MAY-6), for compatibility with
    erroneous TCP implementations.

Such misbehaved TCP impelementations were misbehaved even in 1989 when
RFC 1122 was published -- do we have a sense for whether they are still
around to any significant degree?
</pre>
      <blockquote type="cite">
        <blockquote type="cite">
          <pre class="moz-quote-pre" wrap=""></pre>
        </blockquote>
        <pre class="moz-quote-pre" wrap="">
That's a great question; I don't know, but would highly doubt they are 
around anymore (at least not in production on the Internet).
</pre>
      </blockquote>
      <pre class="moz-quote-pre" wrap="">
Might be an interesting experiment, but it sounds like we don't have enough
data to justify changing the text here (yet).</pre>
    </blockquote>
    <p>As a follow-up action, we can bring this to attention in TCPM and
      see if anyone is interested in helping to gather data on whether
      any implementations still need this.  That would then let us see
      if this can be removed in an update.  But anyways, probably
      shouldn't block this doc (which I think you agree with, since this
      was just a COMMENT, not DISCUSS topic).<br>
    </p>
    <p><br>
    </p>
    <blockquote type="cite"
      cite="mid:20220225210031.GV12881@kduck.mit.edu">
      <blockquote type="cite">
        <blockquote type="cite">
          <pre class="moz-quote-pre" wrap="">
Section 4

    Destination Address
            The network layer address of the remote endpoint.
    [...]
    Source Address
            The network layer address of the sending endpoint.

These definitions don't seem to work in the context of a receiver
validating the TCP checksum, where the destination address is the local
endpoint's address and the source address is the remote endpoint's
address.  (I note that these definitions are different from what RFC 793
itself used.)
</pre>
        </blockquote>
        <pre class="moz-quote-pre" wrap="">
Interesting point on the destination (though the source one seems fine, 
it's always the sending endpoint).  Maybe we should just change "remote" 
to "receiving" in the destination one?
</pre>
      </blockquote>
      <pre class="moz-quote-pre" wrap="">
That's an easy fix; good idea.

</pre>
    </blockquote>
    <p>In rev -27, I changed this to:</p>
    <pre style="box-sizing: border-box; overflow: auto; font-family: &quot;PT Mono&quot;, Monaco, monospace; font-size: 14px; display: block; padding: 10px; margin: 0px 0px 10.5px; line-height: 1.214; color: rgb(0, 0, 0); word-break: break-all; overflow-wrap: break-word; background-color: rgb(255, 253, 245); border: 1px solid rgb(204, 204, 204); border-radius: 4px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-thickness: initial; text-decoration-style: initial; text-decoration-color: initial;">   Destination Address
           The network layer address of the endpoint intended to receive
           a segment.</pre>
    <p><br>
    </p>
    <p><br>
    </p>
    <blockquote type="cite"
      cite="mid:20220225210031.GV12881@kduck.mit.edu">
      <blockquote type="cite">
        <blockquote type="cite">
          <pre class="moz-quote-pre" wrap="">    The more secure Initial Sequence Number generation algorithm from RFC
    6528 was incorporated.  See RFC 6528 for discussion of the attacks
    that this mitigates, as well as advice on selecting PRF algorithms
    and managing secret key data.

(As I mentioned up in §3.4, that guidance is no longer current.)
</pre>
        </blockquote>
        <pre class="moz-quote-pre" wrap="">
Is updating 6528 agreed to be possible future work?
</pre>
      </blockquote>
      <pre class="moz-quote-pre" wrap="">
Yes.  (Is this something where I should be expecting to take the action
item to write a draft, or do you think someone else might take up the pen?)</pre>
    </blockquote>
    <p>It might be worthwhile to succinctly describe the proposed update
      to TCPM, and see what the interest level is.  If you have
      bandwidth to do so, you'd probably be best suited, but otherwise,
      I'd also be happy to bring it up on-list and at the next meeting
      (per chairs approval), as part of closing-out this 793bis work.</p>
    <p><br>
    </p>
    <p><br>
    </p>
    <p>
    </p>
    <blockquote type="cite"
      cite="mid:20220225210031.GV12881@kduck.mit.edu">
      <blockquote type="cite">
        <blockquote type="cite">
          <pre class="moz-quote-pre" wrap="">Section 3.8.6.3

    Note that there are several current practices that further lead to a
    reduced number of ACKs, including generic receive offload (GRO), ACK
    compression, and ACK decimation [26].

Reference [26] seems reasonable for ACK decimation and ACK compression,
but doesn't seem to cover GRO at all.
</pre>
        </blockquote>
        <pre class="moz-quote-pre" wrap="">
True; do you think we should try to find a proper GRO reference? I'm not 
aware of there being a real standard or academic type of reference.
</pre>
      </blockquote>
      <pre class="moz-quote-pre" wrap="">
I wouldn't put much effort into looking.
It's a bit outside my normal field, so I don't have any immediate insight.
Google finds the DPDK docs, which might be stable and well-known, and
wikipedia's <a class="moz-txt-link-freetext" href="https://en.wikipedia.org/wiki/TCP_offload_engine">https://en.wikipedia.org/wiki/TCP_offload_engine</a> entry has a
brief mention, with links to an LWN article and
<a class="moz-txt-link-freetext" href="https://books.google.com/books?id=C3wQBwAAQBAJ">https://books.google.com/books?id=C3wQBwAAQBAJ</a> .  If we know someone with
a copy of the latter, it might be worth checking if it would be a useful
reference.
</pre>
    </blockquote>
    <br>
    <p>This one looks good to me, and I can include it in the next
      revision, if you agree:</p>
    <p><a class="moz-txt-link-freetext" href="https://www.kernel.org/doc/html/latest/networking/segmentation-offloads.html">https://www.kernel.org/doc/html/latest/networking/segmentation-offloads.html</a></p>
    <br>
    <p><br>
    </p>
    <blockquote type="cite"
      cite="mid:20220225210031.GV12881@kduck.mit.edu">
      <blockquote type="cite">
        <pre class="moz-quote-pre" wrap="">
</pre>
        <blockquote type="cite">
          <pre class="moz-quote-pre" wrap="">
Section 3.9.2.2

    Soft Errors
      For ICMP these include: Destination Unreachable -- codes 0, 1, 5,
      Time Exceeded -- codes 0, 1, and Parameter Problem.

      For ICMPv6 these include: Destination Unreachable -- codes 0 and 3,
      Time Exceeded -- codes 0, 1, and Parameter Problem -- codes 0, 1,
      2.

      Since these Unreachable messages indicate soft error conditions,

I'm not entirely sure that I'd classify "parameter problem" as an
"unreachable" message per se.
</pre>
        </blockquote>
        <pre class="moz-quote-pre" wrap="">
I'm not sure I understand this comment.  This is the list from RFC 5461.
</pre>
      </blockquote>
      <pre class="moz-quote-pre" wrap="">
It looks like I was misparsing on first read, since there were commas
serving different roles.  ("Parameter Problem" is an element in the outer
list, matched with "Destination Unreachable" and "Time Exceeded", not one
of the codes for Time Exceeded.)  Using semicolons as delimiters for the
"outer" list, with commas for the lists of type codes, would help
distinguish them.

</pre>
    </blockquote>
    <p>I see; semicolons will be in the next revision.<br>
    </p>
    <p><br>
    </p>
    <p><br>
    </p>
    <blockquote type="cite"
      cite="mid:20220225210031.GV12881@kduck.mit.edu">
      <blockquote type="cite">
        <pre class="moz-quote-pre" wrap="">
</pre>
        <blockquote type="cite">
          <pre class="moz-quote-pre" wrap="">Section 3.10.1

          the parameters of the incoming SYN segment.  Verify the
          security and DiffServ value requested are allowed for this
          user, if not return "error: precedence not allowed" or "error:
          security/compartment not allowed."  If passive enter the LISTEN

It's surprising for the error string to mention "precedence" when the
predicate is DiffServ value.
</pre>
        </blockquote>
        <pre class="moz-quote-pre" wrap="">
This is a good point.  When the definition of those bits was changed, it 
doesn't seem like the error message indicated here was correspondingly 
changed.
</pre>
      </blockquote>
      <pre class="moz-quote-pre" wrap="">
Is this something we need to leave alone for backwards compatibility, or do
we have leeway to change the error message here?  (It looks like this text
is unchanged in the -27.)</pre>
    </blockquote>
    <p>I don't think there's much of a compatibility issue, since the
      socket API works differently than this anyways.  We could change
      "precedence" to "DSCP".<br>
    </p>
    <p><br>
    </p>
    <blockquote type="cite"
      cite="mid:20220225210031.GV12881@kduck.mit.edu">
      <blockquote type="cite">
        <pre class="moz-quote-pre" wrap="">
</pre>
        <blockquote type="cite">
          <pre class="moz-quote-pre" wrap="">             *  If this connection was initiated with a passive OPEN
                (i.e., came from the LISTEN state), then return this
                connection to LISTEN state and return.  The user need
                not be informed.  If this connection was initiated
                with an active OPEN (i.e., came from SYN-SENT state)
                then the connection was refused, signal the user
                "connection refused".  In either case, all segments on
                the retransmission queue should be removed.  And in

IIUC, what's described here as "removed" is described elsewhere as
"flushed"; it would be good to use consistent terminology when possible.
</pre>
        </blockquote>
        <pre class="moz-quote-pre" wrap="">
Do you have a strong preference?
</pre>
      </blockquote>
      <pre class="moz-quote-pre" wrap="">
No strong preference, just a general desire to have consistent terminology
unless there is a difference in meaning intended.


</pre>
    </blockquote>
    <p>Ok, I can change this to say that the retransmission queue should
      be flushed in the next revision.<br>
    </p>
    <p><br>
    </p>
    <p><br>
    </p>
    <p>
    </p>
    <blockquote type="cite"
      cite="mid:20220225210031.GV12881@kduck.mit.edu">
      <blockquote type="cite">
        <pre class="moz-quote-pre" wrap="">
</pre>
        <blockquote type="cite">
          <pre class="moz-quote-pre" wrap="">
Section 4

    internet datagram
            The unit of data exchanged between an internet module and the
            higher level protocol together with the internet header.

"exchanged between an internet module and the higher level protocol"
sounds like a local operation; I would have expected the definition of
an *internet* datagram to involve transfer over the (inter)network.
</pre>
        </blockquote>
        <pre class="moz-quote-pre" wrap="">
Ok, can you suggest an alternative definition?  I agree this one from 
the original 793 isn't terrific.
</pre>
      </blockquote>
      <pre class="moz-quote-pre" wrap="">
I'm somewhat reluctant to concoct a new definition from scratch at this
point in the process.  I see that RFC 1983 has "IP datagram" refer to just
"datagram", on which it says:

   datagram
      A self-contained, independent entity of data carrying sufficient
      information to be routed from the source to the destination
      computer without reliance on earlier exchanges between this source
      and destination computer and the transporting network.  See also:
      frame, packet.
      [Source: J. Postel]

If I was to overcome my reluctance and try to revise the existing text here
into something that makes more sense, it would be:

internet datagram
  A unit of data [from some higher layer protocol] exchanged between
  internet hosts, together with the internet header that allows the
  datagram to be routed from source to destination.

(I'm not sure whether I'd include the bit in square brackets.)
Feel free to use mine if you like, or go with Postel's (per RFC 1983) or
something else.</pre>
    </blockquote>
    Yours looks good to me, and I'll queue it for the next revision.
    <pre class="moz-quote-pre" wrap="">
</pre>
  </body>
</html>

--------------GpwO2dZKownFYRaflUxcDxxb--

