
From georg.hampel@alcatel-lucent.com  Wed Sep  7 08:01:11 2011
Return-Path: <georg.hampel@alcatel-lucent.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1D1EA21F84D8 for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 08:01:11 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.563
X-Spam-Level: 
X-Spam-Status: No, score=-6.563 tagged_above=-999 required=5 tests=[AWL=0.035,  BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id cc06SSl8cVjd for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 08:01:10 -0700 (PDT)
Received: from ihemail2.lucent.com (ihemail2.lucent.com [135.245.0.35]) by ietfa.amsl.com (Postfix) with ESMTP id 201C621F850E for <multipathtcp@ietf.org>; Wed,  7 Sep 2011 08:01:10 -0700 (PDT)
Received: from usnavsmail1.ndc.alcatel-lucent.com (usnavsmail1.ndc.alcatel-lucent.com [135.3.39.9]) by ihemail2.lucent.com (8.13.8/IER-o) with ESMTP id p87F2xEa003220 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for <multipathtcp@ietf.org>; Wed, 7 Sep 2011 10:02:59 -0500 (CDT)
Received: from USNAVSXCHHUB03.ndc.alcatel-lucent.com (usnavsxchhub03.ndc.alcatel-lucent.com [135.3.39.112]) by usnavsmail1.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id p87F2wce005818 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT) for <multipathtcp@ietf.org>; Wed, 7 Sep 2011 10:02:59 -0500
Received: from USNAVSXCHMBSA2.ndc.alcatel-lucent.com ([135.3.39.124]) by USNAVSXCHHUB03.ndc.alcatel-lucent.com ([135.3.39.112]) with mapi; Wed, 7 Sep 2011 10:02:59 -0500
From: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
To: MultiPath TCP - IETF WG <multipathtcp@ietf.org>
Date: Wed, 7 Sep 2011 10:02:58 -0500
Thread-Topic: Modified MP_CAPABLE handshake
Thread-Index: Acxtbzq4wv7vfsOzRKy1f+FpZ1T3Hw==
Message-ID: <154773479ED2314980CB638A48FC443486DE8EC0@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: multipart/alternative; boundary="_000_154773479ED2314980CB638A48FC443486DE8EC0USNAVSXCHMBSA2n_"
MIME-Version: 1.0
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.35
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.9
Subject: [multipathtcp] Modified MP_CAPABLE handshake
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 07 Sep 2011 15:01:11 -0000

--_000_154773479ED2314980CB638A48FC443486DE8EC0USNAVSXCHMBSA2n_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Hi all,

I'd need clarification on the modified MP_CAPABLE handshake introduced in d=
raft-ietf-mptcp-multiaddressed-04.txt:

Since key A is sent for the first time in the third message of the SYN/ACK =
handshake, it needs to be acknowledged with an additional ACK (fourth messa=
ge). In case the connection-initiating host is the one who sends the first =
data (typical client-server session), does it have to wait until this addit=
ional ACK has been received?

The present WG draft seems to indicate that the host has to hold back on da=
ta transmission since the connection is only in PRE_ESTABLISHED state until=
 this fourth message has been received.

Are we aware that this would prolong the connection setup procedure from 1R=
TT to 2RTT for all client-server sessions?

Thanks,

Georg







--_000_154773479ED2314980CB638A48FC443486DE8EC0USNAVSXCHMBSA2n_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns=3D"http://www.w3.org/TR/REC-html40">

<head>
<meta http-equiv=3DContent-Type content=3D"text/html; charset=3Dus-ascii">
<meta name=3DGenerator content=3D"Microsoft Word 11 (filtered medium)">
<style>
<!--
 /* Font Definitions */
 @font-face
	{font-family:"FuturaA Bk BT";
	panose-1:2 11 5 2 2 2 4 2 3 3;}
 /* Style Definitions */
 p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:11.0pt;
	font-family:"FuturaA Bk BT";}
a:link, span.MsoHyperlink
	{color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{color:#606420;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-compose;
	font-family:Arial;
	color:windowtext;}
@page Section1
	{size:595.3pt 841.9pt;
	margin:1.0in 1.25in 1.0in 1.25in;}
div.Section1
	{page:Section1;}
-->
</style>
<!--[if gte mso 9]><xml>
 <o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
 <o:shapelayout v:ext=3D"edit">
  <o:idmap v:ext=3D"edit" data=3D"1" />
 </o:shapelayout></xml><![endif]-->
</head>

<body lang=3DEN-US link=3Dblue vlink=3D"#606420">

<div class=3DSection1>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'>Hi all,<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'>I&#8217;d need clarification on the modified MP_CAPABLE
handshake introduced in draft-ietf-mptcp-multiaddressed-04.txt:<o:p></o:p><=
/span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'>Since key A is sent for the first time in the third mess=
age of
the SYN/ACK handshake, it needs to be acknowledged with an additional ACK
(fourth message). In case the connection-initiating host is the one who sen=
ds
the first data (typical client-server session), does it have to wait until =
this
additional ACK has been received?<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'>The present WG draft seems to indicate that the host has=
 to hold
back on data transmission since the connection is only in PRE_ESTABLISHED s=
tate
until this fourth message has been received.<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'>Are we aware that this would prolong the connection setu=
p
procedure from 1RTT to 2RTT for all client-server sessions?<o:p></o:p></spa=
n></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'>Thanks,<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><br>
Georg<o:p></o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

<p class=3DMsoNormal><font size=3D2 face=3DArial><span style=3D'font-size:1=
0.0pt;
font-family:Arial'><o:p>&nbsp;</o:p></span></font></p>

</div>

</body>

</html>

--_000_154773479ED2314980CB638A48FC443486DE8EC0USNAVSXCHMBSA2n_--

From christoph.paasch@uclouvain.be  Wed Sep  7 08:24:31 2011
Return-Path: <christoph.paasch@uclouvain.be>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 19B5721F8AF6 for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 08:24:31 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.599
X-Spam-Level: 
X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gfeCqb5co7Jt for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 08:24:26 -0700 (PDT)
Received: from smtp6.sgsi.ucl.ac.be (smtp.sgsi.ucl.ac.be [130.104.5.67]) by ietfa.amsl.com (Postfix) with ESMTP id 65EF221F8AF1 for <multipathtcp@ietf.org>; Wed,  7 Sep 2011 08:24:26 -0700 (PDT)
Received: from cpaasch-mac.localnet (a84-230-244-21.elisa-laajakaista.fi [84.230.244.21]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: cpaasch@smtp6.sgsi.ucl.ac.be) by smtp6.sgsi.ucl.ac.be (Postfix) with ESMTPSA id 6B35F1C56F4; Wed,  7 Sep 2011 17:26:11 +0200 (CEST)
X-DKIM: Sendmail DKIM Filter v2.8.3 smtp6.sgsi.ucl.ac.be 6B35F1C56F4
DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=uclouvain.be; s=selucl; t=1315409171; bh=eYVx0z0+ovAwaqnWHRrSOqnX8NB5+NKfTGZP+DEGy0A=; h=From:Reply-To:To:Subject:Date:Cc:References:In-Reply-To: MIME-Version:Content-Type:Content-Transfer-Encoding:Message-Id; b=juzJD7/Rxk6Rp0SfOERrg1l87CEt5a20iiuFmWnDl9DXar0qcuEU7Bj6TFQ2775yg 3/EZgeZ6tqeEEAQtKKFg6R8T+kV60R6E2qkL/XTLDTzQaHnF2drzya3TLMm2emYSVH b6N7SYep915vCS2I+79KPRIDMYLDhIl+3poMBevI=
From: Christoph Paasch <christoph.paasch@uclouvain.be>
Organization: =?iso-8859-1?q?Universit=E9_Catholique_de?= Louvain
To: multipathtcp@ietf.org
Date: Wed, 7 Sep 2011 18:26:10 +0300
User-Agent: KMail/1.13.6 (Linux/2.6.38-11-generic; KDE/4.6.2; x86_64; ; )
References: <154773479ED2314980CB638A48FC443486DE8EC0@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
In-Reply-To: <154773479ED2314980CB638A48FC443486DE8EC0@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
MIME-Version: 1.0
Content-Type: Text/Plain; charset="iso-8859-15"
Content-Transfer-Encoding: quoted-printable
Message-Id: <201109071826.10767.christoph.paasch@uclouvain.be>
X-Virus-Scanned: clamav-milter 0.97-exp at smtp-6.sipr-dc.ucl.ac.be
X-Virus-Status: Clean
X-Sgsi-Spamcheck: SASL authenticated, 
X-SGSI-MailScanner-ID: 6B35F1C56F4.AF32B
X-SGSI-MailScanner: Found to be clean
X-SGSI-From: christoph.paasch@uclouvain.be
X-SGSI-Spam-Status: No
Subject: Re: [multipathtcp] Modified MP_CAPABLE handshake
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: christoph.paasch@uclouvain.be
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 07 Sep 2011 15:24:31 -0000

Hi,

On Wednesday 07 September 2011 wrote Hampel, K Georg (K Georg):
> Since key A is sent for the first time in the third message of the SYN/ACK
> handshake, it needs to be acknowledged with an additional ACK (fourth
> message). In case the connection-initiating host is the one who sends the
> first data (typical client-server session), does it have to wait until
> this additional ACK has been received?

Yes, that's the intention.
I must admit, that we haven't implemented it yet either.

> The present WG draft seems to indicate that the host has to hold back on
> data transmission since the connection is only in PRE_ESTABLISHED state
> until this fourth message has been received.
>=20
> Are we aware that this would prolong the connection setup procedure from
> 1RTT to 2RTT for all client-server sessions?

Mark Handley proposed another solution on this list (http://www.ietf.org/ma=
il-
archive/web/multipathtcp/current/msg01529.html)
I personally prefer this solution to the one in the draft, because adding a=
=20
retransmission-timer for a non-data packet (the third ack) is not that easy=
=20
and clean from an implementation point-of-view.

Another solution might be to just not handle the loss of the third ack=20
(assuming that this is a relatively rare event). If the third ack gets lost=
,=20
and the client is the first one to send data, the server will reply to this=
=20
data  without an mptcp-option (because from his point-of-view we are in non-
mptcp-mode) and thus the client will fallback to regular TCP as specified i=
n=20
section 3.5 of the proto-draft.
This would be much cleaner for an implementation and avoid the additional R=
TT=20
as in the proto-draft, or the reduced option-space (even timestamps don't f=
it=20
anymore in the first data-packet) in Mark's solution.

Cheers,
Christoph

=2D-
Christoph Paasch
PhD Student

IP Networking Lab --- http://inl.info.ucl.ac.be
MultiPath TCP in the Linux Kernel --- http://inl.info.ucl.ac.be/mptcp
Universit=E9 Catholique de Louvain

www.rollerbulls.be
=2D-

From alan.ford@gmail.com  Wed Sep  7 08:30:20 2011
Return-Path: <alan.ford@gmail.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 43D4E21F8B2B for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 08:30:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.598
X-Spam-Level: 
X-Spam-Status: No, score=-3.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4EQj5Riqj9oo for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 08:30:19 -0700 (PDT)
Received: from mail-qy0-f179.google.com (mail-qy0-f179.google.com [209.85.216.179]) by ietfa.amsl.com (Postfix) with ESMTP id 60EFE21F8B27 for <multipathtcp@ietf.org>; Wed,  7 Sep 2011 08:30:19 -0700 (PDT)
Received: by qyk35 with SMTP id 35so4677163qyk.10 for <multipathtcp@ietf.org>; Wed, 07 Sep 2011 08:32:07 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=M0hzpqNMLSw0ngC/QVU6ToDKsb+KO3xLHbAMfsCRaKQ=; b=UerLmHksrDSJUR+UeRn63YmbuKwcDnhi2j+n+9UeeTKY1llpA2rNBrukEC22JWORax ELrLgdUx6TsbsHcuSkZzOj4QHl/PtjMLtin5sr1396f2/9iXg7U9M4K6x3gIa/KIxaaR XOeMqoTecD6LK3NbR6XAjI7tmuJ3lRVMXWzHw=
MIME-Version: 1.0
Received: by 10.229.48.70 with SMTP id q6mr5121271qcf.159.1315409527263; Wed, 07 Sep 2011 08:32:07 -0700 (PDT)
Received: by 10.229.250.6 with HTTP; Wed, 7 Sep 2011 08:32:07 -0700 (PDT)
In-Reply-To: <201109071826.10767.christoph.paasch@uclouvain.be>
References: <154773479ED2314980CB638A48FC443486DE8EC0@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <201109071826.10767.christoph.paasch@uclouvain.be>
Date: Wed, 7 Sep 2011 16:32:07 +0100
Message-ID: <CAOs_kTZ=BAhvsofwLxqk59Gm8xWZ-1X=CDOoKsBuCHFixCFvVg@mail.gmail.com>
From: Alan Ford <alan.ford@gmail.com>
To: christoph.paasch@uclouvain.be
Content-Type: multipart/alternative; boundary=0016368323fa9b51a204ac5ba578
Cc: multipathtcp@ietf.org
Subject: Re: [multipathtcp] Modified MP_CAPABLE handshake
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 07 Sep 2011 15:30:20 -0000

--0016368323fa9b51a204ac5ba578
Content-Type: text/plain; charset=ISO-8859-1

Hi,

2011/9/7 Christoph Paasch <christoph.paasch@uclouvain.be>

> Hi,
>
> On Wednesday 07 September 2011 wrote Hampel, K Georg (K Georg):
> > Since key A is sent for the first time in the third message of the
> SYN/ACK
> > handshake, it needs to be acknowledged with an additional ACK (fourth
> > message). In case the connection-initiating host is the one who sends the
> > first data (typical client-server session), does it have to wait until
> > this additional ACK has been received?
>
> Yes, that's the intention.
> I must admit, that we haven't implemented it yet either.
>

Indeed. The intention was to save option space, however it looks as if it
may be causing more problems than it is worth.

> The present WG draft seems to indicate that the host has to hold back on
> > data transmission since the connection is only in PRE_ESTABLISHED state
> > until this fourth message has been received.
> >
> > Are we aware that this would prolong the connection setup procedure from
> > 1RTT to 2RTT for all client-server sessions?
>
> Mark Handley proposed another solution on this list (
> http://www.ietf.org/mail-
> archive/web/multipathtcp/current/msg01529.html)
> I personally prefer this solution to the one in the draft, because adding a
> retransmission-timer for a non-data packet (the third ack) is not that easy
> and clean from an implementation point-of-view.


Mark's solution is unfortunately not feasible since we do not have enough
option space to fit both a DSS and a MP_CAPABLE in the same packet.


> Another solution might be to just not handle the loss of the third ack
> (assuming that this is a relatively rare event). If the third ack gets
> lost,
> and the client is the first one to send data, the server will reply to this
> data  without an mptcp-option (because from his point-of-view we are in
> non-
> mptcp-mode) and thus the client will fallback to regular TCP as specified
> in
> section 3.5 of the proto-draft.
> This would be much cleaner for an implementation and avoid the additional
> RTT
> as in the proto-draft, or the reduced option-space (even timestamps don't
> fit
> anymore in the first data-packet) in Mark's solution.
>

Indeed, that would work.

However, I think it may well be safest, and easiest, to go back to the -03
version where the keys are in both the SYN and SYN/ACK, and only echoed in
the ACK for reliability.

If a server wants to act statelessly and thus rely on the third ACK for its
keys, then fallback would be the only sensible option in the event of the
third packet loss.

Regards,
Alan

--0016368323fa9b51a204ac5ba578
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Hi,<br><br><div class=3D"gmail_quote">2011/9/7 Christoph Paasch <span dir=
=3D"ltr">&lt;<a href=3D"mailto:christoph.paasch@uclouvain.be">christoph.paa=
sch@uclouvain.be</a>&gt;</span><br><blockquote class=3D"gmail_quote" style=
=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Hi,<br>
<br>
On Wednesday 07 September 2011 wrote Hampel, K Georg (K Georg):<br>
<div class=3D"im">&gt; Since key A is sent for the first time in the third =
message of the SYN/ACK<br>
&gt; handshake, it needs to be acknowledged with an additional ACK (fourth<=
br>
&gt; message). In case the connection-initiating host is the one who sends =
the<br>
&gt; first data (typical client-server session), does it have to wait until=
<br>
&gt; this additional ACK has been received?<br>
<br>
</div>Yes, that&#39;s the intention.<br>
I must admit, that we haven&#39;t implemented it yet either.<br></blockquot=
e><div><br></div><div>Indeed. The intention was to save option space, howev=
er it looks as if it may be causing more problems than it is worth.</div>
<div><br></div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex=
;border-left:1px #ccc solid;padding-left:1ex;"><div class=3D"im">
&gt; The present WG draft seems to indicate that the host has to hold back =
on<br>
&gt; data transmission since the connection is only in PRE_ESTABLISHED stat=
e<br>
&gt; until this fourth message has been received.<br>
&gt;<br>
&gt; Are we aware that this would prolong the connection setup procedure fr=
om<br>
&gt; 1RTT to 2RTT for all client-server sessions?<br>
<br>
</div>Mark Handley proposed another solution on this list (<a href=3D"http:=
//www.ietf.org/mail-
archive/web/multipathtcp/current/msg01529.html" target=3D"_blank">http://ww=
w.ietf.org/mail-<br>
archive/web/multipathtcp/current/msg01529.html</a>)<br>
I personally prefer this solution to the one in the draft, because adding a=
<br>
retransmission-timer for a non-data packet (the third ack) is not that easy=
<br>
and clean from an implementation point-of-view.</blockquote><div><br></div>=
<div>Mark&#39;s solution is unfortunately not feasible since we do not have=
 enough option space to fit both a DSS and a MP_CAPABLE in the same packet.=
</div>
<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex;">
Another solution might be to just not handle the loss of the third ack<br>
(assuming that this is a relatively rare event). If the third ack gets lost=
,<br>
and the client is the first one to send data, the server will reply to this=
<br>
data =A0without an mptcp-option (because from his point-of-view we are in n=
on-<br>
mptcp-mode) and thus the client will fallback to regular TCP as specified i=
n<br>
section 3.5 of the proto-draft.<br>
This would be much cleaner for an implementation and avoid the additional R=
TT<br>
as in the proto-draft, or the reduced option-space (even timestamps don&#39=
;t fit<br>
anymore in the first data-packet) in Mark&#39;s solution.<br></blockquote><=
div><br></div><div>Indeed, that would work.</div><div><br></div><div>Howeve=
r, I think it may well be safest, and easiest, to go back to the -03 versio=
n where the keys are in both the SYN and SYN/ACK, and only echoed in the AC=
K for reliability.</div>
<div><br></div><div>If a server wants to act statelessly and thus rely on t=
he third ACK for its keys, then fallback would be the only sensible option =
in the event of the third packet loss.</div><div><br></div><div>Regards,</d=
iv>
<div>Alan</div><div><br></div></div>

--0016368323fa9b51a204ac5ba578--

From alan.ford@gmail.com  Wed Sep  7 08:38:30 2011
Return-Path: <alan.ford@gmail.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B124F21F8C3A for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 08:38:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -3.598
X-Spam-Level: 
X-Spam-Status: No, score=-3.598 tagged_above=-999 required=5 tests=[AWL=-0.000, BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8bCxKJVkcm+W for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 08:38:30 -0700 (PDT)
Received: from mail-qw0-f52.google.com (mail-qw0-f52.google.com [209.85.216.52]) by ietfa.amsl.com (Postfix) with ESMTP id 17B7621F8C30 for <multipathtcp@ietf.org>; Wed,  7 Sep 2011 08:38:29 -0700 (PDT)
Received: by qwb8 with SMTP id 8so380052qwb.25 for <multipathtcp@ietf.org>; Wed, 07 Sep 2011 08:40:19 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=SPcF3MjBOp+9LYuJyQF3JYzvAIjvCR7we2BwAlEOPIs=; b=dMv5ozTWrS3zi6T8vxNHJtd7Q730oNfk79j6278wSbO6pqZ88r4g2oXjKol/ZHcU4a D/72avVetOBlDMIOw0x4DsShYQAfL3FJie7hszD1Hy6pvtBF6kbd+WbvS5lOR20G4sL2 66AIU1yLglD7FnE6PetwGlncwwbevJim22aw8=
MIME-Version: 1.0
Received: by 10.229.51.66 with SMTP id c2mr3585720qcg.197.1315410019358; Wed, 07 Sep 2011 08:40:19 -0700 (PDT)
Received: by 10.229.250.6 with HTTP; Wed, 7 Sep 2011 08:40:19 -0700 (PDT)
In-Reply-To: <CAOs_kTZ=BAhvsofwLxqk59Gm8xWZ-1X=CDOoKsBuCHFixCFvVg@mail.gmail.com>
References: <154773479ED2314980CB638A48FC443486DE8EC0@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <201109071826.10767.christoph.paasch@uclouvain.be> <CAOs_kTZ=BAhvsofwLxqk59Gm8xWZ-1X=CDOoKsBuCHFixCFvVg@mail.gmail.com>
Date: Wed, 7 Sep 2011 16:40:19 +0100
Message-ID: <CAOs_kTaYSX5nWNQraxT5chsn9E5DUCSWN42qrkw-+NCu70a5Og@mail.gmail.com>
From: Alan Ford <alan.ford@gmail.com>
To: christoph.paasch@uclouvain.be
Content-Type: multipart/alternative; boundary=001636833eeaf0182004ac5bc2e4
Cc: multipathtcp@ietf.org
Subject: Re: [multipathtcp] Modified MP_CAPABLE handshake
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 07 Sep 2011 15:38:30 -0000

--001636833eeaf0182004ac5bc2e4
Content-Type: text/plain; charset=ISO-8859-1

On 7 September 2011 16:32, Alan Ford <alan.ford@gmail.com> wrote:

> Hi,
>
> 2011/9/7 Christoph Paasch <christoph.paasch@uclouvain.be>
>
>>
>> Mark Handley proposed another solution on this list (
>> http://www.ietf.org/mail-
>> archive/web/multipathtcp/current/msg01529.html<http://www.ietf.org/mail-archive/web/multipathtcp/current/msg01529.html>
>> )
>> I personally prefer this solution to the one in the draft, because adding
>> a
>> retransmission-timer for a non-data packet (the third ack) is not that
>> easy
>> and clean from an implementation point-of-view.
>
>
> Mark's solution is unfortunately not feasible since we do not have enough
> option space to fit both a DSS and a MP_CAPABLE in the same packet.
>

Actually, there might be another solution: As long as a DSS covering that
first packet is sent on a following packet (i.e. if the sender had more than
one packet of data to send, or even if not a DSS could be sent on an empty
ACK), then data could be included in the third ACK that contains the
MP_CAPABLE.

This isn't very neat, however, and it may be simpler to revert to keys in
the SYN. What do people think?

Regards,
Alan

--001636833eeaf0182004ac5bc2e4
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<br><br><div class=3D"gmail_quote">On 7 September 2011 16:32, Alan Ford <sp=
an dir=3D"ltr">&lt;<a href=3D"mailto:alan.ford@gmail.com">alan.ford@gmail.c=
om</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"marg=
in:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Hi,<br><br><div class=3D"gmail_quote"><div class=3D"im">2011/9/7 Christoph =
Paasch <span dir=3D"ltr">&lt;<a href=3D"mailto:christoph.paasch@uclouvain.b=
e" target=3D"_blank">christoph.paasch@uclouvain.be</a>&gt;</span><br><block=
quote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc=
 solid;padding-left:1ex">
<br></blockquote></div><div class=3D"im"><blockquote class=3D"gmail_quote" =
style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Mar=
k Handley proposed another solution on this list (<a href=3D"http://www.iet=
f.org/mail-archive/web/multipathtcp/current/msg01529.html" target=3D"_blank=
">http://www.ietf.org/mail-<br>

archive/web/multipathtcp/current/msg01529.html</a>)<br>
I personally prefer this solution to the one in the draft, because adding a=
<br>
retransmission-timer for a non-data packet (the third ack) is not that easy=
<br>
and clean from an implementation point-of-view.</blockquote><div><br></div>=
</div><div>Mark&#39;s solution is unfortunately not feasible since we do no=
t have enough option space to fit both a DSS and a MP_CAPABLE in the same p=
acket.</div>
</div></blockquote><div><br></div><div>Actually, there might be another sol=
ution: As long as a DSS covering that first packet is sent on a following p=
acket (i.e. if the sender had more than one packet of data to send, or even=
 if not a DSS could be sent on an empty ACK), then data could be included i=
n the third ACK that contains the MP_CAPABLE.</div>
<div><br></div><div>This isn&#39;t very neat, however, and it may be simple=
r to revert to keys in the SYN. What do people think?</div><div><br></div><=
div>Regards,</div><div>Alan</div><div><br></div></div>

--001636833eeaf0182004ac5bc2e4--

From christoph.paasch@uclouvain.be  Wed Sep  7 08:38:43 2011
Return-Path: <christoph.paasch@uclouvain.be>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D268F21F8C6A for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 08:38:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.599
X-Spam-Level: 
X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id M7wzUzGvkg6j for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 08:38:39 -0700 (PDT)
Received: from smtp6.sgsi.ucl.ac.be (smtp.sgsi.ucl.ac.be [130.104.5.67]) by ietfa.amsl.com (Postfix) with ESMTP id 4C9B221F8C82 for <multipathtcp@ietf.org>; Wed,  7 Sep 2011 08:38:39 -0700 (PDT)
Received: from cpaasch-mac.localnet (a84-230-244-21.elisa-laajakaista.fi [84.230.244.21]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: cpaasch@smtp6.sgsi.ucl.ac.be) by smtp6.sgsi.ucl.ac.be (Postfix) with ESMTPSA id B0B631C57FE; Wed,  7 Sep 2011 17:40:24 +0200 (CEST)
X-DKIM: Sendmail DKIM Filter v2.8.3 smtp6.sgsi.ucl.ac.be B0B631C57FE
DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=uclouvain.be; s=selucl; t=1315410024; bh=T4ppkWiGEzerR9mfzfwGrPNX9tWmkmUVrg4b2RzWUfI=; h=From:Reply-To:To:Subject:Date:Cc:References:In-Reply-To: MIME-Version:Content-Type:Content-Transfer-Encoding:Message-Id; b=DDFigEUnw91vRbQjrUmDPY9w44QXgbKkiCZI1Hfe7c6VCfh2ndLEkfJDCw3B2oght C1LOC371EZoMXpKpF26HPV0NF4HTAp0AZyCCavD7KhwzBsW2iq2XAOXpq/R5MzFlgp WBmTFUpDtE8V3aQZZomfcGHu7XcgBv+zQfuKpN90=
From: Christoph Paasch <christoph.paasch@uclouvain.be>
Organization: =?iso-8859-1?q?Universit=E9_Catholique_de?= Louvain
To: Alan Ford <alan.ford@gmail.com>
Date: Wed, 7 Sep 2011 18:40:22 +0300
User-Agent: KMail/1.13.6 (Linux/2.6.38-11-generic; KDE/4.6.2; x86_64; ; )
References: <154773479ED2314980CB638A48FC443486DE8EC0@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <201109071826.10767.christoph.paasch@uclouvain.be> <CAOs_kTZ=BAhvsofwLxqk59Gm8xWZ-1X=CDOoKsBuCHFixCFvVg@mail.gmail.com>
In-Reply-To: <CAOs_kTZ=BAhvsofwLxqk59Gm8xWZ-1X=CDOoKsBuCHFixCFvVg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: Text/Plain; charset="iso-8859-15"
Content-Transfer-Encoding: quoted-printable
Message-Id: <201109071840.22993.christoph.paasch@uclouvain.be>
X-Virus-Scanned: clamav-milter 0.97-exp at smtp-6.sipr-dc.ucl.ac.be
X-Virus-Status: Clean
X-Sgsi-Spamcheck: SASL authenticated, 
X-SGSI-MailScanner-ID: B0B631C57FE.AF693
X-SGSI-MailScanner: Found to be clean
X-SGSI-From: christoph.paasch@uclouvain.be
X-SGSI-Spam-Status: No
Cc: multipathtcp@ietf.org
Subject: Re: [multipathtcp] Modified MP_CAPABLE handshake
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: christoph.paasch@uclouvain.be
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 07 Sep 2011 15:38:43 -0000

On Wednesday 07 September 2011 wrote Alan Ford:
> Mark's solution is unfortunately not feasible since we do not have enough
> option space to fit both a DSS and a MP_CAPABLE in the same packet.

The DSS can also be reduced to 32-bit sequence numbers. Then, this would fi=
t,=20
as the MP_CAPABLE in the third ack is 20 bytes long and the DSS fits in the=
=20
remaining 20 bytes. But there is no more space left for timestamps, which a=
re=20
not necessary on each packet.

> Indeed, that would work.
>=20
> However, I think it may well be safest, and easiest, to go back to the -03
> version where the keys are in both the SYN and SYN/ACK, and only echoed in
> the ACK for reliability.
>=20
> If a server wants to act statelessly and thus rely on the third ACK for i=
ts
> keys, then fallback would be the only sensible option in the event of the
> third packet loss.

Yes, I think that would be best. Put the key back in the SYN and ignore the=
=20
loss of the ACK.

Christoph

=2D-
Christoph Paasch
PhD Student

IP Networking Lab --- http://inl.info.ucl.ac.be
MultiPath TCP in the Linux Kernel --- http://inl.info.ucl.ac.be/mptcp
Universit=E9 Catholique de Louvain

www.rollerbulls.be
=2D-

From georg.hampel@alcatel-lucent.com  Wed Sep  7 11:17:42 2011
Return-Path: <georg.hampel@alcatel-lucent.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D2C3721F8CB3 for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 11:17:42 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.565
X-Spam-Level: 
X-Spam-Status: No, score=-6.565 tagged_above=-999 required=5 tests=[AWL=0.034,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mi2MeVzuDSaJ for <multipathtcp@ietfa.amsl.com>; Wed,  7 Sep 2011 11:17:42 -0700 (PDT)
Received: from ihemail1.lucent.com (ihemail1.lucent.com [135.245.0.33]) by ietfa.amsl.com (Postfix) with ESMTP id D935C21F8CA9 for <multipathtcp@ietf.org>; Wed,  7 Sep 2011 11:17:41 -0700 (PDT)
Received: from usnavsmail2.ndc.alcatel-lucent.com (usnavsmail2.ndc.alcatel-lucent.com [135.3.39.10]) by ihemail1.lucent.com (8.13.8/IER-o) with ESMTP id p87IJUoo019724 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Wed, 7 Sep 2011 13:19:30 -0500 (CDT)
Received: from USNAVSXCHHUB01.ndc.alcatel-lucent.com (usnavsxchhub01.ndc.alcatel-lucent.com [135.3.39.110]) by usnavsmail2.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id p87IJU9Q028818 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT); Wed, 7 Sep 2011 13:19:30 -0500
Received: from USNAVSXCHMBSA2.ndc.alcatel-lucent.com ([135.3.39.124]) by USNAVSXCHHUB01.ndc.alcatel-lucent.com ([135.3.39.110]) with mapi; Wed, 7 Sep 2011 13:19:30 -0500
From: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
To: "christoph.paasch@uclouvain.be" <christoph.paasch@uclouvain.be>, Alan Ford <alan.ford@gmail.com>
Date: Wed, 7 Sep 2011 13:19:28 -0500
Thread-Topic: [multipathtcp] Modified MP_CAPABLE handshake
Thread-Index: AcxtdH44eV37bgKOQB6ruvLpN++FLQAE3oGw
Message-ID: <154773479ED2314980CB638A48FC443486DE8FF3@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <154773479ED2314980CB638A48FC443486DE8EC0@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <201109071826.10767.christoph.paasch@uclouvain.be> <CAOs_kTZ=BAhvsofwLxqk59Gm8xWZ-1X=CDOoKsBuCHFixCFvVg@mail.gmail.com> <201109071840.22993.christoph.paasch@uclouvain.be>
In-Reply-To: <201109071840.22993.christoph.paasch@uclouvain.be>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.33
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.10
Cc: "multipathtcp@ietf.org" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Modified MP_CAPABLE handshake
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 07 Sep 2011 18:17:42 -0000

All,

I think that delaying client requests from 1RTT to 2RTT is pretty unaccepta=
ble. In the tcpm WG, Google tries to push for data on the SYN packet to red=
uce the client request time from 1RTT to 0 RTT. In this light, MPTCP wouldn=
't quite meet the trend.

I would tend to ask the question how much security we gain by pushing the k=
ey A transfer from the first SYN to the second ACK.

I further see a philosophical problem in the 3rd ACK solution: MPTCP-level =
messages (e.g. DSN, data FIN) should be acknowledged via MPTCP-level messag=
es (e.g. data ACKs) and not via subflow level ACKs. The -03 version of the =
draft was consistent in that respect.

My vote: Go back to -03.

Georg


-----Original Message-----
From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org] =
On Behalf Of Christoph Paasch
Sent: Wednesday, September 07, 2011 11:40 AM
To: Alan Ford
Cc: multipathtcp@ietf.org
Subject: Re: [multipathtcp] Modified MP_CAPABLE handshake


On Wednesday 07 September 2011 wrote Alan Ford:
> Mark's solution is unfortunately not feasible since we do not have enough
> option space to fit both a DSS and a MP_CAPABLE in the same packet.

The DSS can also be reduced to 32-bit sequence numbers. Then, this would fi=
t,=20
as the MP_CAPABLE in the third ack is 20 bytes long and the DSS fits in the=
=20
remaining 20 bytes. But there is no more space left for timestamps, which a=
re=20
not necessary on each packet.

> Indeed, that would work.
>=20
> However, I think it may well be safest, and easiest, to go back to the -0=
3
> version where the keys are in both the SYN and SYN/ACK, and only echoed i=
n
> the ACK for reliability.
>=20
> If a server wants to act statelessly and thus rely on the third ACK for i=
ts
> keys, then fallback would be the only sensible option in the event of the
> third packet loss.

Yes, I think that would be best. Put the key back in the SYN and ignore the=
=20
loss of the ACK.

Christoph

--
Christoph Paasch
PhD Student

IP Networking Lab --- http://inl.info.ucl.ac.be
MultiPath TCP in the Linux Kernel --- http://inl.info.ucl.ac.be/mptcp
Universit=E9 Catholique de Louvain

www.rollerbulls.be
--
_______________________________________________
multipathtcp mailing list
multipathtcp@ietf.org
https://www.ietf.org/mailman/listinfo/multipathtcp

From Michael.Scharf@alcatel-lucent.com  Thu Sep  8 01:37:35 2011
Return-Path: <Michael.Scharf@alcatel-lucent.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A9FAC21F8B8D for <multipathtcp@ietfa.amsl.com>; Thu,  8 Sep 2011 01:37:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.207
X-Spam-Level: 
X-Spam-Status: No, score=-6.207 tagged_above=-999 required=5 tests=[AWL=0.042,  BAYES_00=-2.599, HELO_EQ_DE=0.35, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id V3f+SUr1azdo for <multipathtcp@ietfa.amsl.com>; Thu,  8 Sep 2011 01:37:35 -0700 (PDT)
Received: from mailrelay1.alcatel.de (mailrelay1.alcatel.de [194.113.59.95]) by ietfa.amsl.com (Postfix) with ESMTP id DEA2221F8B74 for <multipathtcp@ietf.org>; Thu,  8 Sep 2011 01:37:34 -0700 (PDT)
Received: from SLFSNX.rcs.alcatel-research.de (slfsn1.rcs.de.alcatel-lucent.com [149.204.60.98]) by mailrelay1.alcatel.de (8.14.3/8.14.3/ICT) with ESMTP id p888dKbV008167 for <multipathtcp@ietf.org>; Thu, 8 Sep 2011 10:39:23 +0200
X-MimeOLE: Produced By Microsoft Exchange V6.5
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Date: Thu, 8 Sep 2011 10:38:00 +0200
Message-ID: <133D9897FB9C5E4E9DF2779DC91E947C066835A3@SLFSNX.rcs.alcatel-research.de>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: Hint to IEEE Globecom 2011 paper
Thread-Index: AcxuAp4GHiSR4lo6RludfvgqKXGvPQ==
From: "SCHARF, Michael" <Michael.Scharf@alcatel-lucent.com>
To: "MultiPath TCP - IETF WG" <multipathtcp@ietf.org>
X-Scanned-By: MIMEDefang 2.69 on 149.204.45.72
Subject: [multipathtcp] Hint to IEEE Globecom 2011 paper
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 08 Sep 2011 08:37:35 -0000

Hi,

partly off-topic: I finally managed to write a paper that summarizes
some of our insights and results regarding payload encoding. If someone
is interested in a copy, please contact me offlist.

Michael


Michael Scharf, Thomas-Rolf Banniza, "MCTCP: A Multipath Transport Shim
Layer", accepted for publication in IEEE Globecom 2011, 5-9 December
2011, Houston, Texas.

Abstract: Multipath transport protocols enable the concurrent
use of different paths and resource pooling of their capacity.
Ongoing research and standardization activities extend the
Transmission Control Protocol (TCP) towards a Multipath TCP
solution. The objective is to aggregate multiple subflows between
two endpoints into a single session that can be accessed by an
application like a single TCP connection.
This paper presents and evaluates Multi-Connection TCP
(MCTCP) as a new multipath transport solution. MCTCP
consists of a shim layer on top of several TCP connections and
encodes control information, as far as possible, in their payload.
In this paper, we first motivate MCTCP's design and compare it
to a multipath transport protocol that uses TCP option encoding
only. Second, we show that MCTCP can be implemented in
the Linux stack with very few kernel modifications. We also
demonstrate that MCTCP can be combined with a coupled
multipath congestion control mechanism. Measurement results
prove that MCTCP is an efficient and robust protocol that can
dynamically pool the bandwidth of several paths.

From c.raiciu@cs.ucl.ac.uk  Thu Sep  8 03:35:49 2011
Return-Path: <c.raiciu@cs.ucl.ac.uk>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2742D21F8AF7 for <multipathtcp@ietfa.amsl.com>; Thu,  8 Sep 2011 03:35:49 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.357
X-Spam-Level: 
X-Spam-Status: No, score=-0.357 tagged_above=-999 required=5 tests=[AWL=2.242,  BAYES_00=-2.599]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uu1+-iRPfw4K for <multipathtcp@ietfa.amsl.com>; Thu,  8 Sep 2011 03:35:45 -0700 (PDT)
Received: from bells2.cs.ucl.ac.uk (bells2.cs.ucl.ac.uk [128.16.5.33]) by ietfa.amsl.com (Postfix) with ESMTP id E0B7B21F8AF0 for <multipathtcp@ietf.org>; Thu,  8 Sep 2011 03:35:44 -0700 (PDT)
Received: from [92.80.121.178] (helo=[192.168.1.3]) by bells2.cs.ucl.ac.uk with esmtpsa (TLSv1:AES128-SHA:128) (C.Raiciu authenticated) (Exim 4.54) id 1R1bye-0000XK-A8; Thu, 08 Sep 2011 11:37:00 +0100
Mime-Version: 1.0 (Apple Message framework v1081)
Content-Type: text/plain; charset=iso-8859-1
From: Costin Raiciu <c.raiciu@cs.ucl.ac.uk>
In-Reply-To: <154773479ED2314980CB638A48FC443486DE8FF3@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
Date: Thu, 8 Sep 2011 13:37:26 +0300
Content-Transfer-Encoding: quoted-printable
Message-Id: <EDCE929A-62F9-4FE0-AA51-EA40DC31BAE1@cs.ucl.ac.uk>
References: <154773479ED2314980CB638A48FC443486DE8EC0@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <201109071826.10767.christoph.paasch@uclouvain.be> <CAOs_kTZ=BAhvsofwLxqk59Gm8xWZ-1X=CDOoKsBuCHFixCFvVg@mail.gmail.com> <201109071840.22993.christoph.paasch@uclouvain.be> <154773479ED2314980CB638A48FC443486DE8FF3@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
To: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
X-Mailer: Apple Mail (2.1081)
Cc: "multipathtcp@ietf.org" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Modified MP_CAPABLE handshake
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 08 Sep 2011 10:35:49 -0000

Hi,

I also think moving back to -03 is the way to go for the initial =
handshake.
Costin

On 7 Sep 2011, at 21:19, Hampel, K Georg (K Georg) wrote:

> All,
>=20
> I think that delaying client requests from 1RTT to 2RTT is pretty =
unacceptable. In the tcpm WG, Google tries to push for data on the SYN =
packet to reduce the client request time from 1RTT to 0 RTT. In this =
light, MPTCP wouldn't quite meet the trend.
>=20
> I would tend to ask the question how much security we gain by pushing =
the key A transfer from the first SYN to the second ACK.
>=20
> I further see a philosophical problem in the 3rd ACK solution: =
MPTCP-level messages (e.g. DSN, data FIN) should be acknowledged via =
MPTCP-level messages (e.g. data ACKs) and not via subflow level ACKs. =
The -03 version of the draft was consistent in that respect.
>=20
> My vote: Go back to -03.
>=20
> Georg
>=20
>=20
> -----Original Message-----
> From: multipathtcp-bounces@ietf.org =
[mailto:multipathtcp-bounces@ietf.org] On Behalf Of Christoph Paasch
> Sent: Wednesday, September 07, 2011 11:40 AM
> To: Alan Ford
> Cc: multipathtcp@ietf.org
> Subject: Re: [multipathtcp] Modified MP_CAPABLE handshake
>=20
>=20
> On Wednesday 07 September 2011 wrote Alan Ford:
>> Mark's solution is unfortunately not feasible since we do not have =
enough
>> option space to fit both a DSS and a MP_CAPABLE in the same packet.
>=20
> The DSS can also be reduced to 32-bit sequence numbers. Then, this =
would fit,=20
> as the MP_CAPABLE in the third ack is 20 bytes long and the DSS fits =
in the=20
> remaining 20 bytes. But there is no more space left for timestamps, =
which are=20
> not necessary on each packet.
>=20
>> Indeed, that would work.
>>=20
>> However, I think it may well be safest, and easiest, to go back to =
the -03
>> version where the keys are in both the SYN and SYN/ACK, and only =
echoed in
>> the ACK for reliability.
>>=20
>> If a server wants to act statelessly and thus rely on the third ACK =
for its
>> keys, then fallback would be the only sensible option in the event of =
the
>> third packet loss.
>=20
> Yes, I think that would be best. Put the key back in the SYN and =
ignore the=20
> loss of the ACK.
>=20
> Christoph
>=20
> --
> Christoph Paasch
> PhD Student
>=20
> IP Networking Lab --- http://inl.info.ucl.ac.be
> MultiPath TCP in the Linux Kernel --- http://inl.info.ucl.ac.be/mptcp
> Universit=E9 Catholique de Louvain
>=20
> www.rollerbulls.be
> --
> _______________________________________________
> multipathtcp mailing list
> multipathtcp@ietf.org
> https://www.ietf.org/mailman/listinfo/multipathtcp
> _______________________________________________
> multipathtcp mailing list
> multipathtcp@ietf.org
> https://www.ietf.org/mailman/listinfo/multipathtcp


From C.RAICIU@cs.ucl.ac.uk  Thu Sep  8 12:38:31 2011
Return-Path: <C.RAICIU@cs.ucl.ac.uk>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3E7C821F8B3F for <multipathtcp@ietfa.amsl.com>; Thu,  8 Sep 2011 12:38:31 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.805
X-Spam-Level: 
X-Spam-Status: No, score=-0.805 tagged_above=-999 required=5 tests=[AWL=1.194,  BAYES_00=-2.599, J_CHICKENPOX_72=0.6]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iSfK8RookaSO for <multipathtcp@ietfa.amsl.com>; Thu,  8 Sep 2011 12:38:30 -0700 (PDT)
Received: from bells2.cs.ucl.ac.uk (bells2.cs.ucl.ac.uk [128.16.5.33]) by ietfa.amsl.com (Postfix) with ESMTP id D0E4B21F8B3A for <multipathtcp@ietf.org>; Thu,  8 Sep 2011 12:38:29 -0700 (PDT)
Received: from [92.80.121.178] (helo=[192.168.1.3]) by bells2.cs.ucl.ac.uk with esmtpsa (TLSv1:AES128-SHA:128) (C.Raiciu authenticated) (Exim 4.54) id 1R1kRk-0000lJ-Gu; Thu, 08 Sep 2011 20:39:37 +0100
Content-Type: text/plain; charset=windows-1252
Mime-Version: 1.0 (Apple Message framework v1081)
From: Costin Raiciu <C.RAICIU@cs.ucl.ac.uk>
In-Reply-To: <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk>
Date: Thu, 8 Sep 2011 22:40:14 +0300
Content-Transfer-Encoding: quoted-printable
Message-Id: <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk>
To: Alan Ford <alan.ford@gmail.com>, "philip.eardley Eardley" <philip.eardley@bt.com>, draft-ietf-mptcp-multiaddressed@tools.ietf.org, Yoshifumi Nishida <yoshifumi.nishida@gmail.com>, "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
X-Mailer: Apple Mail (2.1081)
Subject: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 08 Sep 2011 19:38:31 -0000

Hi all,

As you know, we are in the stage of finalizing the MPTCP protocol draft. =
There are two major changes still to be made. Having re-read recently =
all the emails that have gone back and forth, here are the resolutions I =
propose.

The first question relates to the ability of opening new subflows in =
fallback mode (break before make in fallback is a specialization of this =
case).=20

My inclination would be to mandate the following:=20

- when MPTCP goes into fallback mode, it effectively becomes equivalent =
to a regular TCP session; no further subflows can be setup; teardown =
happens only with regular FIN messages, etc.

This does prevent mobility in certain cases that would require a lot of =
complexity to handle; it seems reasonable given my hope that payload =
changing middleboxes are not that ubiquitous.

The second question regards adding support for transparent proxies. In =
particular, each endpoint should know, for each address of their peer, =
if it is talking to a proxy or the destination. This would allow a =
smooth transition from
an initial proxy based deployment in cellular networks to worldwide =
support for MPTCP in servers. In particular, MPTCP could connect to the =
proxy=20

The proposal is to add the P flag to ADD_ADDRESS, to allow middleboxes =
to say they are on path.=20

Currently ADD_ADDRESS does not have flag space;=20
- we could add another octet for flag space (but this is not ideal since =
this option is already big enough for IPv6, as Georg pointed out)
- or we could reduce the number of bits in the address ID by one. 256 =
addresses seem like a lot of interfaces to have at the same time; 128 =
should suffice. We should explicitly allow the number to wrap, though, =
to support long term mobility where a mobile's Wifi interface comes up =
and down hundreds of times. If we do this, we also free one bit on the =
MP_JOIN.

My inclination is to go for the second option.

When I first thought of it, it seemed to me that adding the P flag only =
to add_address is not sufficient; the host does not know upon directly =
opening a connection/subflow whether it is talking directly to the =
destination or a proxy.=20
It turns out that using add address the proxy can make itself known for =
the initial path. This path will have address ID 0; the proxy will just =
send an add address with ID 0 and it's real address. The receiver now =
compares this address with what he thinks the remote endpoint is, and =
can infer it is talking to a proxy on that path. The same applies if a =
new subflow is opened.

What do people think? Are these reasonable solutions? If people agree, I =
will go ahead and rev the draft accordingly.

Cheers,
Costin
>=20
> On 6 Sep 2011, at 18:31, Alan Ford wrote:
>=20
>> Folks,
>>=20
>> Does anybody agree/disagree with my comments below? Can anybody =
contribute text for (2) and (5)? These are the main ones that I don't =
know enough about what was discussed and was intended in order to write =
anything myself.
>>=20
>> Thanks,
>> Alan
>>=20
>> On 22 August 2011 22:08, Alan Ford <alan.ford@gmail.com> wrote:
>> Apologies for the delay in replying.
>>=20
>> (1) - This won't work, unfortunately - we don't have enough option =
space.
>>=20
>> (2) - I'm afraid I'm out of touch on what the conclusion here was. =
When I last thought about this, the only way I could see it working was =
if a subflow was clearly terminated before a new one was started, =
relying on the timeout you talk about in (3) to keep the MPTCP state =
alive. Was there an alternative way?
>>=20
>> (3) - Indeed this is heuristics, although more so than for regular =
TCP applications may make assumptions about session survivability. This =
is one of those issues that we need to say needs thought in the protocol =
draft, and then in the API draft talk about its actual impact.
>>=20
>> (4) - ok
>>=20
>> (5) - Once again I'm out of touch of what was discussed at Quebec so =
am not sure how this would be defined. I remain uneasy about this flag - =
embedding proxies as a fundamental part of the protocol - but am willing =
to bow to consensus at Quebec.
>>=20
>> Cheers,
>> Alan
>>=20
>>=20
>> On 19 August 2011 08:15, <philip.eardley@bt.com> wrote:
>> Hi,
>>=20
>> Just a reminder, please can we try and close these issues
>>=20
>> (1)    no-one complained at Mark=92s suggestion
>>=20
>> (2)    Needs some slightly revised text
>>=20
>> (3)    May need some tweaked text?
>>=20
>> (4)    Needs no action
>>=20
>> (5)    Is the main point requiring a detailed proposal
>>=20
>> Thanks
>>=20
>> phil
>>=20
>>=20
>>=20
>> From: Eardley,PL,Philip,DES8 R=20
>> Sent: 28 July 2011 00:01
>> To: Eardley,PL,Philip,DES8 R; multipathtcp@ietf.org
>> Subject: Open /closed issues from WG meeting
>>=20
>>=20
>>=20
>> =46rom the WG meeting I think we have a proposal on various Issues =
that seemed open before the meeting.
>>=20
>> This email is to check consensus, especially amongst the WG people =
who aren't in Quebec or didn't listen in.
>>=20
>>=20
>>=20
>> Of the issues, (1) was raised in the meeting, (2) & (3) are from =
Mark's slide 12, (4) and (5) are from Georg's slide 34.
>>=20
>>=20
>>=20
>> (1) Ack of MP_CAPABLE.
>>=20
>> See Mark's email to the list a short while ago, which explains & =
proposes a resolution.
>>=20
>>=20
>>=20
>> (2) Mobility in fallback
>>=20
>> The doc needs to correctly describe the case - when it works OK & =
when it doesn't (ie what the limitation is). We accept the limitation =
(it isn't too clear how to do a solution but seems it would be complex)
>>=20
>>=20
>>=20
>> (3) Teardown of state when all subflows fail
>>=20
>> This is a heuristics issue rather than a protocol issue.
>>=20
>> (Hopefully in the future we can write a doc with lessons learnt from =
experiments, with suggestions how to deal with all sorts of heuristics =
issues)
>>=20
>>=20
>>=20
>> (4)  Add Bulk_transfer_optimisation flag to MP-Capable?
>>=20
>> Don't add, seems like extra complexity for not much gain
>>=20
>>=20
>>=20
>> (5) Add support for transparent proxy?
>>=20
>> Rough consensus in the room in Quebec to add an "I'm a proxy" flag to =
ADD_ADDRESS.
>>=20
>> Need the precise text to add to the draft.
>>=20
>>=20
>>=20
>> We'd like to resolve all these issues within the next week or so - =
hoping that proves possible!
>>=20
>>=20
>>=20
>> We will WG last call as soon as there's an updated draft that deals =
with these issues.
>>=20
>>=20
>>=20
>> If I got something wrong, please shout.
>>=20
>>=20
>>=20
>> thanks to everyone for the efforts to try and resolve the hopefully =
last remaining issues with the protocol spec.
>>=20
>> Best wishes,
>>=20
>> phil
>>=20
>> From: multipathtcp-bounces@ietf.org [multipathtcp-bounces@ietf.org] =
On Behalf Of philip.eardley@bt.com [philip.eardley@bt.com]
>> Sent: 27 July 2011 23:34
>> To: multipathtcp@ietf.org
>> Subject: [multipathtcp] Draft minutes of WG meeting
>>=20
>> thanks to Andrew for taking notes.
>>=20
>>=20
>>=20
>> Here are the draft minutes of the meeting, please send corrections =
/improvements
>>=20
>> thanks
>>=20
>> phil/
>>=20
>>=20
>>=20
>> Minutes of MPTCP WG Meeting
>>=20
>> WEDNESDAY, July 27, 2011
>> 0900-1130  Morning Session I
>>=20
>> Main aim:
>> Conclude on all protocol issues. So that we can WG last call protocol =
(& API) docs immediately after IETF.
>>=20
>>=20
>>=20
>> 1. Agenda bashing, note well, Milestones =96 Chairs
>>=20
>>=20
>>=20
>> 2. API update =96 Michael Scharf
>>=20
>> Doc ready for last call, no comments.
>>=20
>>=20
>>=20
>> 3. Congestion control update =96 Mark Handley
>> Done. Minor comments from IESG and David Black have been resolved.
>>=20
>>=20
>>=20
>> 4. Protocol update & discussion of open issues =96 Mark Handley
>>=20
>> Reviewers could review section 2 (overview)
>>=20
>> Small changes to protocol
>>=20
>>=20
>>=20
>> Discussion over sending key-A on the third Ack; there is an issue =
over the reliable delivery of the key.
>>=20
>> Jana: B could echo back the options.
>>=20
>> Mark: The issue is how to trigger stopping option sends.  We will =
double-check if the issue is covered and resolve if required.
>>=20
>>=20
>>=20
>> Discussion on mobility in fallback mode
>>=20
>> It doesn't seem that it is possible to reliably use mobility in =
fallback mode.
>>=20
>> Andrew: Isn't that OK?  Can't you just say, if the ACKs haven't =
arrived, oh well, RST it.
>>=20
>> Georg: The draft says that there's no more multipath signalling in =
fallback.  So you can't get here.
>>=20
>> Mark: So, the only resolution seems to be that if you end up with =
un-ACKed data, you lose.
>>=20
>> Costin: You have signalling in the SYNs
>>=20
>> Mark: Which gives you a way out, if that is possible.
>>=20
>> Georg: I have a proposal for how to cover this later.
>>=20
>> Mark: It probably isn't worth trying to optimise this case.
>>=20
>> What to do if all subflows are in time out?
>>=20
>> Phil: You're saying that is an implementation specific heuristic?
>>=20
>> Andrew: What say no interfaces are up?
>>=20
>> Mark: Yes, doesn't need specified.
>>=20
>>=20
>> Phil: Plan is, get these three issues resolved on the list, give it a =
week or so, and WGLC it.
>>=20
>>=20
>>=20
>>=20
>> Experiments on "Is MPTCP deployable?"
>>=20
>> Georg: You can't name service providers or enterprises, to what =
extent are the harmful boxes installed in providers vs enterprises?
>>=20
>> Mark: Cellular providers have more, but no monopoly.  We can't claim =
this to be completely representative.
>>=20
>>=20
>> 5. (new draft) Enhancements to Improve the Applicability of Multipath =
TCP to Wireless Access Networks =96 Georg Hampel
>>=20
>> Mark: Best path probably has lowest energy per byte
>>=20
>> Georg: If you have money or energy to burn, you might not care
>>=20
>> Mark: Proxy.  One part of me is revolted by supporting transparent =
proxies explicitly, and we should support them.  Calling the flag =
I_AM_A_PROXY is more explicit.
>>=20
>> Andrew: I agree there are always proxy devices "there"
>> Andrew: ... there are good features here (is it important to tag this =
as "min" complexity, images are getting big for this part of the stack)
>> Georg: How about recv buffer reduction?
>> Andrew: good idea
>>=20
>> Mark: I'm not totally convinced that buffer reduction is real.
>>=20
>> Costin: Receive buffer reduction is orthogonal to implementation =
complexity
>>=20
>> Mark: The other question, if you're in path selection mode, do you =
get the right behaviour over mobile networks.  Assert: you should be =
able to tell from signal strength etc. if it is a good link... not =
really.  I can't easily predict from wireless if it's decent, I have to =
actually try it.
>>=20
>> Georg: The signal is not necessarily indicative, could have other =
variables
>>=20
>> Andrew: Wifi links have to train up, which can take a few hundred =
packets
>>=20
>> David: I have in mind a satellite PEP setup, where RTT could be =
radically different.
>>=20
>> Mark: Our intuition is that you should only use one path most of the =
time, but you want to transition make-before-break and do a soft =
transition.  You need to get the links adequately probed.
>>=20
>> Georg: Both of our points are based on speculation, we need to know =
how well you do if you shorten the overlap
>>=20
>> Costin: that paper assumed there is an oracle that knows the best =
path that chooses the best path yes: 15% compared to an optimal path =
selection algorithm that can't be achieved in practice
>>=20
>> Costin: you can also have redundant data if you want to probe and =
avoid hol blocking
>>=20
>> Mark: The spec does not prohibit you sending data on multiple paths
>>=20
>> Phil: We've had quite a lot of discussion now on heuristic stuff, how =
you might make this work in practice.  It would be nice to have more =
experiments.  My impression is that BTO doesn't seem to gain very much
>>=20
>> Andrew: Suggest not doing bulk transfer optimisation, it could be =
really expensive for little gain.
>>=20
>> Andrew: Very much in favour of proxy flags, they're everywhere
>>=20
>> Mark: I am in favour of adding a bit to support transparent proxies, =
1) address of proxy 2) remote supports mptcp - could decide based on =
this
>>=20
>> Andrew: It can be a content inspecting firewall that just wants to do =
its job
>>=20
>> Georg: Proxy is not optimisation, it's a feature
>>=20
>> Michio: No such middle box
>>=20
>> Mark: If we can preserve the benefits in the presence of boxes that =
want to be on path, then we should.  If we don't want the content =
inspected, we need to encrypt.
>>=20
>> Georg: 3G providers still get to bill it.
>>=20
>> Gorry: There are proxies that need to inform L2 what is going on.
>>=20
>> Andrew: Could have network architectures that wouldn't otherwise work =
that way.
>>=20
>> Mark: We should probably try to address this before we move the =
protocol forward.
>>=20
>> Phil: Would like to see people get together in the remainder of the =
week.
>>=20
>> Andrew: I think the bottom three lines of the slide are OK, up to =
naming.
>>=20
>> Costin: +1
>>=20
>>=20
>>=20
>> 6. Call for news on Implementations
>>=20
>> Linux version being updated as http://mptcpinfo.ucl.ac.be, also for =
Nokia N900 & Android Nexus-one
>>=20
>> Tim: Implementation is not SMP capable, does anyone know if it is =
fixed?
>>=20
>> <apparently not, docs still say no SMP>
>>=20
>> Georg Hampel - in progress on 'lightweight' implementation
>>=20
>> Sebastien Roy: Oracle are working on a prototype in Solaris.  Don't =
know if it can be released yet.
>>=20
>>=20
>>=20
>> 7. Discussion on future of WG - Chairs
>>=20
>>=20
>>=20
>> Discussion last time (roughly) concluded that we pause for =
experiments & usage: WG hears about implementations, heuristics etc; =
hears about investigations on possible extensions; not do any protocol =
standards work. Re-charter in ~ 1year when understand better what the =
needs are & where the energy is
>> No comments - Chairs will work with AD on re-chartering text, when =
the current charter is done.
>>=20
>>=20
>=20


From georg.hampel@alcatel-lucent.com  Thu Sep  8 13:44:35 2011
Return-Path: <georg.hampel@alcatel-lucent.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C922621F8B49 for <multipathtcp@ietfa.amsl.com>; Thu,  8 Sep 2011 13:44:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.267
X-Spam-Level: 
X-Spam-Status: No, score=-6.267 tagged_above=-999 required=5 tests=[AWL=-0.268, BAYES_00=-2.599, J_CHICKENPOX_72=0.6, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PDLlGq-EWg6T for <multipathtcp@ietfa.amsl.com>; Thu,  8 Sep 2011 13:44:34 -0700 (PDT)
Received: from ihemail4.lucent.com (ihemail4.lucent.com [135.245.0.39]) by ietfa.amsl.com (Postfix) with ESMTP id 2B29321F8B46 for <multipathtcp@ietf.org>; Thu,  8 Sep 2011 13:44:34 -0700 (PDT)
Received: from usnavsmail3.ndc.alcatel-lucent.com (usnavsmail3.ndc.alcatel-lucent.com [135.3.39.11]) by ihemail4.lucent.com (8.13.8/IER-o) with ESMTP id p88Kk9kp024752 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 8 Sep 2011 15:46:10 -0500 (CDT)
Received: from USNAVSXCHHUB03.ndc.alcatel-lucent.com (usnavsxchhub03.ndc.alcatel-lucent.com [135.3.39.112]) by usnavsmail3.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id p88Kk5Gv025821 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT); Thu, 8 Sep 2011 15:46:09 -0500
Received: from USNAVSXCHMBSA2.ndc.alcatel-lucent.com ([135.3.39.124]) by USNAVSXCHHUB03.ndc.alcatel-lucent.com ([135.3.39.112]) with mapi; Thu, 8 Sep 2011 15:46:06 -0500
From: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
To: Costin Raiciu <C.RAICIU@cs.ucl.ac.uk>, Alan Ford <alan.ford@gmail.com>, "philip.eardley Eardley" <philip.eardley@bt.com>, "draft-ietf-mptcp-multiaddressed@tools.ietf.org" <draft-ietf-mptcp-multiaddressed@tools.ietf.org>, Yoshifumi Nishida <yoshifumi.nishida@gmail.com>, "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
Date: Thu, 8 Sep 2011 15:46:05 -0500
Thread-Topic: [multipathtcp] Proposed changes to the protocol draft
Thread-Index: AcxuXytL5+HGsRd9Shijl8kyPhkoxAAAV4xw
Message-ID: <154773479ED2314980CB638A48FC443486DE94C6@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk>
In-Reply-To: <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.39
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.11
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 08 Sep 2011 20:44:35 -0000

Hi Costin, all,

First some comments on "opening new subflows in fallback mode":

In case checksums are used, I agree with your inclination to restrict fallb=
ack mode to regular TCP. If not checksums are used, however, there are no p=
ayload-rewriting middleboxes (by definition). Under such circumstances, DSS=
 with infinity setting could be interpreted as "use this mapping until upda=
ted via a new DSS". The next DSS can be sent on any packet and on any subfl=
ow.

I strongly advocate such "temporary fallback solution in absence of checksu=
ms" for the following reasons:

- There is currently little information about the likelihood to encounter a=
 payload-rewriting middlebox. Such middleboxes may be very rare in fact. It=
 is further possible that such middleboxes also exhibit other MPTCP-obstruc=
ting behavior, i.e. swallow of TCP options etc, which obsoletes the problem=
.

- According to Andrew McGregor, checksum computation stalls hardware-accele=
rated stack implementations. In an offline discussion, he recommended to av=
oid them.

- Single-path connectivity is still the typical scenario for wireless setti=
ngs. MPTCP provides an opportunity in case a second path shows up. The corr=
esponding opportunity cost, however, should be held at minimum. With 20-28B=
, DSS has an overhead comparable to a TCP-in-TCP tunnel, and it has to be c=
arried over the air interface (a lot for VoIP).

- In my opinion, bulk optimization adds unnecessary delay and processing ov=
erhead.


Here are some comments on your ideas about the transparent proxy. If I unde=
rstand you right, you offer 3 potential solutions:

a) Add a byte to ADD_ADDRESS for the P flag.

b) Use one bit of the address-id for the P flag.

c) Use address-id =3D 0 in ADD_ADDRESS (or MP_JOIN), which is usually the d=
efault for the initial subflow.

I have a preference for solution c). Some issues:
- Q: What happens to the address-id of the interface used for the initial s=
ubflow (whose value was 0)? A: This address cannot be used anymore. The ini=
tial subflow, however, does not have to be torn down. Once torn down, it ca=
nnot be re-established since the address is gone. Is there any problem here=
?
- Q: How to distinguish "do join" from "I'm a proxy"? A: Address-id =3D 0 m=
eans "I'm a proxy". Address > 0 means "do join". Is that right?

Thanks,

Georg







-----Original Message-----
From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org] =
On Behalf Of Costin Raiciu
Sent: Thursday, September 08, 2011 3:40 PM
To: Alan Ford; philip.eardley Eardley; draft-ietf-mptcp-multiaddressed@tool=
s.ietf.org; Yoshifumi Nishida; multipathtcp@ietf.org List
Subject: [multipathtcp] Proposed changes to the protocol draft

Hi all,

As you know, we are in the stage of finalizing the MPTCP protocol draft. Th=
ere are two major changes still to be made. Having re-read recently all the=
 emails that have gone back and forth, here are the resolutions I propose.

The first question relates to the ability of opening new subflows in fallba=
ck mode (break before make in fallback is a specialization of this case).

My inclination would be to mandate the following:

- when MPTCP goes into fallback mode, it effectively becomes equivalent to =
a regular TCP session; no further subflows can be setup; teardown happens o=
nly with regular FIN messages, etc.

This does prevent mobility in certain cases that would require a lot of com=
plexity to handle; it seems reasonable given my hope that payload changing =
middleboxes are not that ubiquitous.

The second question regards adding support for transparent proxies. In part=
icular, each endpoint should know, for each address of their peer, if it is=
 talking to a proxy or the destination. This would allow a smooth transitio=
n from
an initial proxy based deployment in cellular networks to worldwide support=
 for MPTCP in servers. In particular, MPTCP could connect to the proxy

The proposal is to add the P flag to ADD_ADDRESS, to allow middleboxes to s=
ay they are on path.

Currently ADD_ADDRESS does not have flag space;
- we could add another octet for flag space (but this is not ideal since th=
is option is already big enough for IPv6, as Georg pointed out)
- or we could reduce the number of bits in the address ID by one. 256 addre=
sses seem like a lot of interfaces to have at the same time; 128 should suf=
fice. We should explicitly allow the number to wrap, though, to support lon=
g term mobility where a mobile's Wifi interface comes up and down hundreds =
of times. If we do this, we also free one bit on the MP_JOIN.

My inclination is to go for the second option.

When I first thought of it, it seemed to me that adding the P flag only to =
add_address is not sufficient; the host does not know upon directly opening=
 a connection/subflow whether it is talking directly to the destination or =
a proxy.
It turns out that using add address the proxy can make itself known for the=
 initial path. This path will have address ID 0; the proxy will just send a=
n add address with ID 0 and it's real address. The receiver now compares th=
is address with what he thinks the remote endpoint is, and can infer it is =
talking to a proxy on that path. The same applies if a new subflow is opene=
d.

What do people think? Are these reasonable solutions? If people agree, I wi=
ll go ahead and rev the draft accordingly.

Cheers,
Costin
>
> On 6 Sep 2011, at 18:31, Alan Ford wrote:
>
>> Folks,
>>
>> Does anybody agree/disagree with my comments below? Can anybody contribu=
te text for (2) and (5)? These are the main ones that I don't know enough a=
bout what was discussed and was intended in order to write anything myself.
>>
>> Thanks,
>> Alan
>>
>> On 22 August 2011 22:08, Alan Ford <alan.ford@gmail.com> wrote:
>> Apologies for the delay in replying.
>>
>> (1) - This won't work, unfortunately - we don't have enough option space=
.
>>
>> (2) - I'm afraid I'm out of touch on what the conclusion here was. When =
I last thought about this, the only way I could see it working was if a sub=
flow was clearly terminated before a new one was started, relying on the ti=
meout you talk about in (3) to keep the MPTCP state alive. Was there an alt=
ernative way?
>>
>> (3) - Indeed this is heuristics, although more so than for regular TCP a=
pplications may make assumptions about session survivability. This is one o=
f those issues that we need to say needs thought in the protocol draft, and=
 then in the API draft talk about its actual impact.
>>
>> (4) - ok
>>
>> (5) - Once again I'm out of touch of what was discussed at Quebec so am =
not sure how this would be defined. I remain uneasy about this flag - embed=
ding proxies as a fundamental part of the protocol - but am willing to bow =
to consensus at Quebec.
>>
>> Cheers,
>> Alan
>>
>>
>> On 19 August 2011 08:15, <philip.eardley@bt.com> wrote:
>> Hi,
>>
>> Just a reminder, please can we try and close these issues
>>
>> (1)    no-one complained at Mark's suggestion
>>
>> (2)    Needs some slightly revised text
>>
>> (3)    May need some tweaked text?
>>
>> (4)    Needs no action
>>
>> (5)    Is the main point requiring a detailed proposal
>>
>> Thanks
>>
>> phil
>>
>>
>>
>> From: Eardley,PL,Philip,DES8 R
>> Sent: 28 July 2011 00:01
>> To: Eardley,PL,Philip,DES8 R; multipathtcp@ietf.org
>> Subject: Open /closed issues from WG meeting
>>
>>
>>
>> From the WG meeting I think we have a proposal on various Issues that se=
emed open before the meeting.
>>
>> This email is to check consensus, especially amongst the WG people who a=
ren't in Quebec or didn't listen in.
>>
>>
>>
>> Of the issues, (1) was raised in the meeting, (2) & (3) are from Mark's =
slide 12, (4) and (5) are from Georg's slide 34.
>>
>>
>>
>> (1) Ack of MP_CAPABLE.
>>
>> See Mark's email to the list a short while ago, which explains & propose=
s a resolution.
>>
>>
>>
>> (2) Mobility in fallback
>>
>> The doc needs to correctly describe the case - when it works OK & when i=
t doesn't (ie what the limitation is). We accept the limitation (it isn't t=
oo clear how to do a solution but seems it would be complex)
>>
>>
>>
>> (3) Teardown of state when all subflows fail
>>
>> This is a heuristics issue rather than a protocol issue.
>>
>> (Hopefully in the future we can write a doc with lessons learnt from exp=
eriments, with suggestions how to deal with all sorts of heuristics issues)
>>
>>
>>
>> (4)  Add Bulk_transfer_optimisation flag to MP-Capable?
>>
>> Don't add, seems like extra complexity for not much gain
>>
>>
>>
>> (5) Add support for transparent proxy?
>>
>> Rough consensus in the room in Quebec to add an "I'm a proxy" flag to AD=
D_ADDRESS.
>>
>> Need the precise text to add to the draft.
>>
>>
>>
>> We'd like to resolve all these issues within the next week or so - hopin=
g that proves possible!
>>
>>
>>
>> We will WG last call as soon as there's an updated draft that deals with=
 these issues.
>>
>>
>>
>> If I got something wrong, please shout.
>>
>>
>>
>> thanks to everyone for the efforts to try and resolve the hopefully last=
 remaining issues with the protocol spec.
>>
>> Best wishes,
>>
>> phil
>>
>> From: multipathtcp-bounces@ietf.org [multipathtcp-bounces@ietf.org] On B=
ehalf Of philip.eardley@bt.com [philip.eardley@bt.com]
>> Sent: 27 July 2011 23:34
>> To: multipathtcp@ietf.org
>> Subject: [multipathtcp] Draft minutes of WG meeting
>>
>> thanks to Andrew for taking notes.
>>
>>
>>
>> Here are the draft minutes of the meeting, please send corrections /impr=
ovements
>>
>> thanks
>>
>> phil/
>>
>>
>>
>> Minutes of MPTCP WG Meeting
>>
>> WEDNESDAY, July 27, 2011
>> 0900-1130  Morning Session I
>>
>> Main aim:
>> Conclude on all protocol issues. So that we can WG last call protocol (&=
 API) docs immediately after IETF.
>>
>>
>>
>> 1. Agenda bashing, note well, Milestones - Chairs
>>
>>
>>
>> 2. API update - Michael Scharf
>>
>> Doc ready for last call, no comments.
>>
>>
>>
>> 3. Congestion control update - Mark Handley
>> Done. Minor comments from IESG and David Black have been resolved.
>>
>>
>>
>> 4. Protocol update & discussion of open issues - Mark Handley
>>
>> Reviewers could review section 2 (overview)
>>
>> Small changes to protocol
>>
>>
>>
>> Discussion over sending key-A on the third Ack; there is an issue over t=
he reliable delivery of the key.
>>
>> Jana: B could echo back the options.
>>
>> Mark: The issue is how to trigger stopping option sends.  We will double=
-check if the issue is covered and resolve if required.
>>
>>
>>
>> Discussion on mobility in fallback mode
>>
>> It doesn't seem that it is possible to reliably use mobility in fallback=
 mode.
>>
>> Andrew: Isn't that OK?  Can't you just say, if the ACKs haven't arrived,=
 oh well, RST it.
>>
>> Georg: The draft says that there's no more multipath signalling in fallb=
ack.  So you can't get here.
>>
>> Mark: So, the only resolution seems to be that if you end up with un-ACK=
ed data, you lose.
>>
>> Costin: You have signalling in the SYNs
>>
>> Mark: Which gives you a way out, if that is possible.
>>
>> Georg: I have a proposal for how to cover this later.
>>
>> Mark: It probably isn't worth trying to optimise this case.
>>
>> What to do if all subflows are in time out?
>>
>> Phil: You're saying that is an implementation specific heuristic?
>>
>> Andrew: What say no interfaces are up?
>>
>> Mark: Yes, doesn't need specified.
>>
>>
>> Phil: Plan is, get these three issues resolved on the list, give it a we=
ek or so, and WGLC it.
>>
>>
>>
>>
>> Experiments on "Is MPTCP deployable?"
>>
>> Georg: You can't name service providers or enterprises, to what extent a=
re the harmful boxes installed in providers vs enterprises?
>>
>> Mark: Cellular providers have more, but no monopoly.  We can't claim thi=
s to be completely representative.
>>
>>
>> 5. (new draft) Enhancements to Improve the Applicability of Multipath TC=
P to Wireless Access Networks - Georg Hampel
>>
>> Mark: Best path probably has lowest energy per byte
>>
>> Georg: If you have money or energy to burn, you might not care
>>
>> Mark: Proxy.  One part of me is revolted by supporting transparent proxi=
es explicitly, and we should support them.  Calling the flag I_AM_A_PROXY i=
s more explicit.
>>
>> Andrew: I agree there are always proxy devices "there"
>> Andrew: ... there are good features here (is it important to tag this as=
 "min" complexity, images are getting big for this part of the stack)
>> Georg: How about recv buffer reduction?
>> Andrew: good idea
>>
>> Mark: I'm not totally convinced that buffer reduction is real.
>>
>> Costin: Receive buffer reduction is orthogonal to implementation complex=
ity
>>
>> Mark: The other question, if you're in path selection mode, do you get t=
he right behaviour over mobile networks.  Assert: you should be able to tel=
l from signal strength etc. if it is a good link... not really.  I can't ea=
sily predict from wireless if it's decent, I have to actually try it.
>>
>> Georg: The signal is not necessarily indicative, could have other variab=
les
>>
>> Andrew: Wifi links have to train up, which can take a few hundred packet=
s
>>
>> David: I have in mind a satellite PEP setup, where RTT could be radicall=
y different.
>>
>> Mark: Our intuition is that you should only use one path most of the tim=
e, but you want to transition make-before-break and do a soft transition.  =
You need to get the links adequately probed.
>>
>> Georg: Both of our points are based on speculation, we need to know how =
well you do if you shorten the overlap
>>
>> Costin: that paper assumed there is an oracle that knows the best path t=
hat chooses the best path yes: 15% compared to an optimal path selection al=
gorithm that can't be achieved in practice
>>
>> Costin: you can also have redundant data if you want to probe and avoid =
hol blocking
>>
>> Mark: The spec does not prohibit you sending data on multiple paths
>>
>> Phil: We've had quite a lot of discussion now on heuristic stuff, how yo=
u might make this work in practice.  It would be nice to have more experime=
nts.  My impression is that BTO doesn't seem to gain very much
>>
>> Andrew: Suggest not doing bulk transfer optimisation, it could be really=
 expensive for little gain.
>>
>> Andrew: Very much in favour of proxy flags, they're everywhere
>>
>> Mark: I am in favour of adding a bit to support transparent proxies, 1) =
address of proxy 2) remote supports mptcp - could decide based on this
>>
>> Andrew: It can be a content inspecting firewall that just wants to do it=
s job
>>
>> Georg: Proxy is not optimisation, it's a feature
>>
>> Michio: No such middle box
>>
>> Mark: If we can preserve the benefits in the presence of boxes that want=
 to be on path, then we should.  If we don't want the content inspected, we=
 need to encrypt.
>>
>> Georg: 3G providers still get to bill it.
>>
>> Gorry: There are proxies that need to inform L2 what is going on.
>>
>> Andrew: Could have network architectures that wouldn't otherwise work th=
at way.
>>
>> Mark: We should probably try to address this before we move the protocol=
 forward.
>>
>> Phil: Would like to see people get together in the remainder of the week=
.
>>
>> Andrew: I think the bottom three lines of the slide are OK, up to naming=
.
>>
>> Costin: +1
>>
>>
>>
>> 6. Call for news on Implementations
>>
>> Linux version being updated as http://mptcpinfo.ucl.ac.be, also for Noki=
a N900 & Android Nexus-one
>>
>> Tim: Implementation is not SMP capable, does anyone know if it is fixed?
>>
>> <apparently not, docs still say no SMP>
>>
>> Georg Hampel - in progress on 'lightweight' implementation
>>
>> Sebastien Roy: Oracle are working on a prototype in Solaris.  Don't know=
 if it can be released yet.
>>
>>
>>
>> 7. Discussion on future of WG - Chairs
>>
>>
>>
>> Discussion last time (roughly) concluded that we pause for experiments &=
 usage: WG hears about implementations, heuristics etc; hears about investi=
gations on possible extensions; not do any protocol standards work. Re-char=
ter in ~ 1year when understand better what the needs are & where the energy=
 is
>> No comments - Chairs will work with AD on re-chartering text, when the c=
urrent charter is done.
>>
>>
>

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

From christoph.paasch@uclouvain.be  Fri Sep  9 01:00:27 2011
Return-Path: <christoph.paasch@uclouvain.be>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5E0C621F86FF for <multipathtcp@ietfa.amsl.com>; Fri,  9 Sep 2011 01:00:27 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -5.377
X-Spam-Level: 
X-Spam-Status: No, score=-5.377 tagged_above=-999 required=5 tests=[AWL=0.604,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, RCVD_IN_SORBS_WEB=0.619]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 25bwgZ4Lh6pN for <multipathtcp@ietfa.amsl.com>; Fri,  9 Sep 2011 01:00:26 -0700 (PDT)
Received: from smtp5.sgsi.ucl.ac.be (smtp.sgsi.ucl.ac.be [130.104.5.67]) by ietfa.amsl.com (Postfix) with ESMTP id 2981321F86F6 for <multipathtcp@ietf.org>; Fri,  9 Sep 2011 01:00:26 -0700 (PDT)
Received: from cpaasch-mac.localnet (unknown [192.100.124.156]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: cpaasch@smtp5.sgsi.ucl.ac.be) by smtp5.sgsi.ucl.ac.be (Postfix) with ESMTPSA id 1CCD211E52D; Fri,  9 Sep 2011 10:02:15 +0200 (CEST)
X-DKIM: Sendmail DKIM Filter v2.8.3 smtp5.sgsi.ucl.ac.be 1CCD211E52D
DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=uclouvain.be; s=selucl; t=1315555335; bh=IfMoLLGQwHiJKzOEX+46ygFExIa6RTJrZR8qf19kSXA=; h=From:Reply-To:To:Subject:Date:Cc:References:In-Reply-To: MIME-Version:Content-Type:Content-Transfer-Encoding:Message-Id; b=zo2hCUurPM/yzj3OpquF1g5QwykbGW8wwWy4M3HW7FZpFL7T8WrdKlzAonCV56Uy2 b1bCK4xVqvVMt3OqTVHo8eaqBN5vsAVYrUDzXU9b7LkkE7rlDh9pFpFsNLrQZXA+2x Y8GxlHOX9OeYiGy1JmIWD3JQdvPCoKnGSJkBnhts=
From: Christoph Paasch <christoph.paasch@uclouvain.be>
Organization: =?iso-8859-1?q?Universit=E9_Catholique_de?= Louvain
To: multipathtcp@ietf.org
Date: Fri, 9 Sep 2011 11:02:13 +0300
User-Agent: KMail/1.13.6 (Linux/2.6.38-11-generic; KDE/4.6.2; x86_64; ; )
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <154773479ED2314980CB638A48FC443486DE94C6@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
In-Reply-To: <154773479ED2314980CB638A48FC443486DE94C6@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
MIME-Version: 1.0
Content-Type: Text/Plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Message-Id: <201109091102.13578.christoph.paasch@uclouvain.be>
X-Virus-Scanned: clamav-milter 0.97-exp at smtp-5.sipr-dc.ucl.ac.be
X-Virus-Status: Clean
X-Sgsi-Spamcheck: SASL authenticated, 
X-SGSI-MailScanner-ID: 1CCD211E52D.A2268
X-SGSI-MailScanner: Found to be clean
X-SGSI-From: christoph.paasch@uclouvain.be
X-SGSI-Spam-Status: No
Cc: Yoshifumi Nishida <yoshifumi.nishida@gmail.com>, "draft-ietf-mptcp-multiaddressed@tools.ietf.org" <draft-ietf-mptcp-multiaddressed@tools.ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: christoph.paasch@uclouvain.be
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 09 Sep 2011 08:00:27 -0000

Hi George,

On Thursday 08 September 2011 wrote Hampel, K Georg (K Georg):
> In case checksums are used, I agree with your inclination to restrict
> fallback mode to regular TCP. If not checksums are used, however, there
> are no payload-rewriting middleboxes (by definition). Under such
> circumstances, DSS with infinity setting could be interpreted as "use this
> mapping until updated via a new DSS". The next DSS can be sent on any
> packet and on any subflow.

If dss-checksums are not used, how can we (by definition) be sure that ther=
e=20
are no payload-rewriting middleboxes?

If we are in infinity-setting and thus no DSS-checksums are used,  and a=20
payload-rewriting middlebox is adding a byte to the byte-stream. After=20
changing the subflow, the TCP-layer will offer a corrupted byte-stream to t=
he=20
application. Because the normalizing DSS will be one byte less than what we=
=20
already offered to the application. Thus, the first byte on the new subflow=
=20
will not be passed to the application and thus this byte got lost.
I think, we should at all means, avoid a corrupted byte-stream. And we can=
=20
never be sure that this is not going to happen.

As far as I can see, mobility in fallback-mode can only work if we can=20
correctly close a subflow before opening the new one.
But how can we correctly close a subflow in mobility, as we do not know, wh=
en=20
we will loose the access-point?

> - According to Andrew McGregor, checksum computation stalls
> hardware-accelerated stack implementations. In an offline discussion, he
> recommended to avoid them.

The overhead of DSS-checksum calculation is not that much. I only see a=20
difference at very high speeds (about 4Gbps).


Cheers,
Christoph

=2D-
Christoph Paasch
PhD Student

IP Networking Lab --- http://inl.info.ucl.ac.be
MultiPath TCP in the Linux Kernel --- http://inl.info.ucl.ac.be/mptcp
Universit=E9 Catholique de Louvain

www.rollerbulls.be
=2D-

From C.RAICIU@cs.ucl.ac.uk  Fri Sep  9 01:13:21 2011
Return-Path: <C.RAICIU@cs.ucl.ac.uk>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E0D1821F8AB0 for <multipathtcp@ietfa.amsl.com>; Fri,  9 Sep 2011 01:13:21 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.299
X-Spam-Level: 
X-Spam-Status: No, score=-2.299 tagged_above=-999 required=5 tests=[AWL=-0.300, BAYES_00=-2.599, J_CHICKENPOX_72=0.6]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RFkKJ8ysr+NR for <multipathtcp@ietfa.amsl.com>; Fri,  9 Sep 2011 01:13:20 -0700 (PDT)
Received: from bells2.cs.ucl.ac.uk (bells2.cs.ucl.ac.uk [128.16.5.33]) by ietfa.amsl.com (Postfix) with ESMTP id 1612A21F8B1E for <multipathtcp@ietf.org>; Fri,  9 Sep 2011 01:13:19 -0700 (PDT)
Received: from [141.85.225.204] (helo=[192.168.1.3]) by bells2.cs.ucl.ac.uk with esmtpsa (TLSv1:AES128-SHA:128) (C.Raiciu authenticated) (Exim 4.54) id 1R1wDi-0000vt-O5; Fri, 09 Sep 2011 09:13:55 +0100
Mime-Version: 1.0 (Apple Message framework v1081)
Content-Type: text/plain; charset=us-ascii
From: Costin Raiciu <C.RAICIU@cs.ucl.ac.uk>
In-Reply-To: <154773479ED2314980CB638A48FC443486DE94C6@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
Date: Fri, 9 Sep 2011 11:14:50 +0300
Content-Transfer-Encoding: quoted-printable
Message-Id: <1FC9F558-8730-4C64-B8A8-F90947B1178C@cs.ucl.ac.uk>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <154773479ED2314980CB638A48FC443486DE94C6@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
To: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
X-Mailer: Apple Mail (2.1081)
Cc: "multipathtcp@ietf.org List" <multipathtcp@ietf.org>, "draft-ietf-mptcp-multiaddressed@tools.ietf.org" <draft-ietf-mptcp-multiaddressed@tools.ietf.org>, Yoshifumi Nishida <yoshifumi.nishida@gmail.com>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 09 Sep 2011 08:13:22 -0000

Hi,

> First some comments on "opening new subflows in fallback mode":
>=20
> In case checksums are used, I agree with your inclination to restrict =
fallback mode to regular TCP. If not checksums are used, however, there =
are no payload-rewriting middleboxes (by definition). Under such =
circumstances, DSS with infinity setting could be interpreted as "use =
this mapping until updated via a new DSS". The next DSS can be sent on =
any packet and on any subflow.
>=20
> I strongly advocate such "temporary fallback solution in absence of =
checksums"=20

If checksums are not used, MPTCP should not go into fallback mode - =
failed checksums were the only ways MPTCP would go into fallback mode.=20=

Since fallback cannot really happen (or will happen very rarely, if the =
spec is updated), there is no point in allowing  mobility in fallback =
mode.=20

Fallback mode was designed as a way to signal to endpoints running MPTCP =
to revert to TCP from a certain sequence number onwards, in the middle =
of the connection. We expect this to be very rare in practice, so there =
is no need to add unnecessary complexity for this fallback mode.

MPTCP can revert to TCP in the beginning of the connection also; after =
this it never comes back to MPTCP mode. Why should we allow MPTCP to =
revert back from TCP in fallback mode (i.e. mid connection downgrade), =
and not allow it to revert back in some cases when the connection has =
been downgraded in the beginning (for instance because options did not =
get thourgh on data segments)?

Adding complexity to an already pretty complex protocol is probably the =
wrong thing to do. I think we should keep things simple and mandate =
that:=20

"Once MPTCP reverts to TCP, it MUST NOT revert back to MPTCP aftewards".=20=



> Here are some comments on your ideas about the transparent proxy. If I =
understand you right, you offer 3 potential solutions:
>=20
> a) Add a byte to ADD_ADDRESS for the P flag.
>=20
> b) Use one bit of the address-id for the P flag.
>=20
> c) Use address-id =3D 0 in ADD_ADDRESS (or MP_JOIN), which is usually =
the default for the initial subflow.
>=20
> I have a preference for solution c). Some issues:
> - Q: What happens to the address-id of the interface used for the =
initial subflow (whose value was 0)? A: This address cannot be used =
anymore. The initial subflow, however, does not have to be torn down. =
Once torn down, it cannot be re-established since the address is gone. =
Is there any problem here?
> - Q: How to distinguish "do join" from "I'm a proxy"? A: Address-id =3D =
0 means "I'm a proxy". Address > 0 means "do join". Is that right?

Actually, the only usable solutions in my mind are a) and b)
They would allow a proxy to send an add_address with a new address_id =
and a proxy flag, telling the endpoint that the advertised address =
belongs to a proxy.

However, this by itself does not allow the endpoint to tell if it is =
talking to a proxy on its initial subflow, or any other subflows for =
which is has not received information via ADD_ADDRESS with the P flag =
set and the corresponding address ID.

I was using solution (c) for this exact purpose. Once an address is =
first signalled implicitly (via subflow creation with MP_CAPABLE =
(address_id =3D 0) or MP_JOIN (address_id =3D x)), the proxy can also =
send an ADD_ADDRESS with the same address_id (0 or x), thus informing =
the endpoint that is a proxy address.

We thus need to change the rules on accepting ADD_ADDRESS options: when =
duplicate options are received for the same address ID, they should be =
processed nevertheless. This shouldn't introduce new vulnerabilities =
because subflow setup is the thing that matters and that uses crypto.

Cheers,
Costin




> -----Original Message-----
> From: multipathtcp-bounces@ietf.org =
[mailto:multipathtcp-bounces@ietf.org] On Behalf Of Costin Raiciu
> Sent: Thursday, September 08, 2011 3:40 PM
> To: Alan Ford; philip.eardley Eardley; =
draft-ietf-mptcp-multiaddressed@tools.ietf.org; Yoshifumi Nishida; =
multipathtcp@ietf.org List
> Subject: [multipathtcp] Proposed changes to the protocol draft
>=20
> Hi all,
>=20
> As you know, we are in the stage of finalizing the MPTCP protocol =
draft. There are two major changes still to be made. Having re-read =
recently all the emails that have gone back and forth, here are the =
resolutions I propose.
>=20
> The first question relates to the ability of opening new subflows in =
fallback mode (break before make in fallback is a specialization of this =
case).
>=20
> My inclination would be to mandate the following:
>=20
> - when MPTCP goes into fallback mode, it effectively becomes =
equivalent to a regular TCP session; no further subflows can be setup; =
teardown happens only with regular FIN messages, etc.
>=20
> This does prevent mobility in certain cases that would require a lot =
of complexity to handle; it seems reasonable given my hope that payload =
changing middleboxes are not that ubiquitous.
>=20
> The second question regards adding support for transparent proxies. In =
particular, each endpoint should know, for each address of their peer, =
if it is talking to a proxy or the destination. This would allow a =
smooth transition from
> an initial proxy based deployment in cellular networks to worldwide =
support for MPTCP in servers. In particular, MPTCP could connect to the =
proxy
>=20
> The proposal is to add the P flag to ADD_ADDRESS, to allow middleboxes =
to say they are on path.
>=20
> Currently ADD_ADDRESS does not have flag space;
> - we could add another octet for flag space (but this is not ideal =
since this option is already big enough for IPv6, as Georg pointed out)
> - or we could reduce the number of bits in the address ID by one. 256 =
addresses seem like a lot of interfaces to have at the same time; 128 =
should suffice. We should explicitly allow the number to wrap, though, =
to support long term mobility where a mobile's Wifi interface comes up =
and down hundreds of times. If we do this, we also free one bit on the =
MP_JOIN.
>=20
> My inclination is to go for the second option.
>=20
> When I first thought of it, it seemed to me that adding the P flag =
only to add_address is not sufficient; the host does not know upon =
directly opening a connection/subflow whether it is talking directly to =
the destination or a proxy.
> It turns out that using add address the proxy can make itself known =
for the initial path. This path will have address ID 0; the proxy will =
just send an add address with ID 0 and it's real address. The receiver =
now compares this address with what he thinks the remote endpoint is, =
and can infer it is talking to a proxy on that path. The same applies if =
a new subflow is opened.
>=20
> What do people think? Are these reasonable solutions? If people agree, =
I will go ahead and rev the draft accordingly.
>=20
> Cheers,
> Costin
>>=20
>> On 6 Sep 2011, at 18:31, Alan Ford wrote:
>>=20
>>> Folks,
>>>=20
>>> Does anybody agree/disagree with my comments below? Can anybody =
contribute text for (2) and (5)? These are the main ones that I don't =
know enough about what was discussed and was intended in order to write =
anything myself.
>>>=20
>>> Thanks,
>>> Alan
>>>=20
>>> On 22 August 2011 22:08, Alan Ford <alan.ford@gmail.com> wrote:
>>> Apologies for the delay in replying.
>>>=20
>>> (1) - This won't work, unfortunately - we don't have enough option =
space.
>>>=20
>>> (2) - I'm afraid I'm out of touch on what the conclusion here was. =
When I last thought about this, the only way I could see it working was =
if a subflow was clearly terminated before a new one was started, =
relying on the timeout you talk about in (3) to keep the MPTCP state =
alive. Was there an alternative way?
>>>=20
>>> (3) - Indeed this is heuristics, although more so than for regular =
TCP applications may make assumptions about session survivability. This =
is one of those issues that we need to say needs thought in the protocol =
draft, and then in the API draft talk about its actual impact.
>>>=20
>>> (4) - ok
>>>=20
>>> (5) - Once again I'm out of touch of what was discussed at Quebec so =
am not sure how this would be defined. I remain uneasy about this flag - =
embedding proxies as a fundamental part of the protocol - but am willing =
to bow to consensus at Quebec.
>>>=20
>>> Cheers,
>>> Alan
>>>=20
>>>=20
>>> On 19 August 2011 08:15, <philip.eardley@bt.com> wrote:
>>> Hi,
>>>=20
>>> Just a reminder, please can we try and close these issues
>>>=20
>>> (1)    no-one complained at Mark's suggestion
>>>=20
>>> (2)    Needs some slightly revised text
>>>=20
>>> (3)    May need some tweaked text?
>>>=20
>>> (4)    Needs no action
>>>=20
>>> (5)    Is the main point requiring a detailed proposal
>>>=20
>>> Thanks
>>>=20
>>> phil
>>>=20
>>>=20
>>>=20
>>> From: Eardley,PL,Philip,DES8 R
>>> Sent: 28 July 2011 00:01
>>> To: Eardley,PL,Philip,DES8 R; multipathtcp@ietf.org
>>> Subject: Open /closed issues from WG meeting
>>>=20
>>>=20
>>>=20
>>> =46rom the WG meeting I think we have a proposal on various Issues =
that seemed open before the meeting.
>>>=20
>>> This email is to check consensus, especially amongst the WG people =
who aren't in Quebec or didn't listen in.
>>>=20
>>>=20
>>>=20
>>> Of the issues, (1) was raised in the meeting, (2) & (3) are from =
Mark's slide 12, (4) and (5) are from Georg's slide 34.
>>>=20
>>>=20
>>>=20
>>> (1) Ack of MP_CAPABLE.
>>>=20
>>> See Mark's email to the list a short while ago, which explains & =
proposes a resolution.
>>>=20
>>>=20
>>>=20
>>> (2) Mobility in fallback
>>>=20
>>> The doc needs to correctly describe the case - when it works OK & =
when it doesn't (ie what the limitation is). We accept the limitation =
(it isn't too clear how to do a solution but seems it would be complex)
>>>=20
>>>=20
>>>=20
>>> (3) Teardown of state when all subflows fail
>>>=20
>>> This is a heuristics issue rather than a protocol issue.
>>>=20
>>> (Hopefully in the future we can write a doc with lessons learnt from =
experiments, with suggestions how to deal with all sorts of heuristics =
issues)
>>>=20
>>>=20
>>>=20
>>> (4)  Add Bulk_transfer_optimisation flag to MP-Capable?
>>>=20
>>> Don't add, seems like extra complexity for not much gain
>>>=20
>>>=20
>>>=20
>>> (5) Add support for transparent proxy?
>>>=20
>>> Rough consensus in the room in Quebec to add an "I'm a proxy" flag =
to ADD_ADDRESS.
>>>=20
>>> Need the precise text to add to the draft.
>>>=20
>>>=20
>>>=20
>>> We'd like to resolve all these issues within the next week or so - =
hoping that proves possible!
>>>=20
>>>=20
>>>=20
>>> We will WG last call as soon as there's an updated draft that deals =
with these issues.
>>>=20
>>>=20
>>>=20
>>> If I got something wrong, please shout.
>>>=20
>>>=20
>>>=20
>>> thanks to everyone for the efforts to try and resolve the hopefully =
last remaining issues with the protocol spec.
>>>=20
>>> Best wishes,
>>>=20
>>> phil
>>>=20
>>> From: multipathtcp-bounces@ietf.org [multipathtcp-bounces@ietf.org] =
On Behalf Of philip.eardley@bt.com [philip.eardley@bt.com]
>>> Sent: 27 July 2011 23:34
>>> To: multipathtcp@ietf.org
>>> Subject: [multipathtcp] Draft minutes of WG meeting
>>>=20
>>> thanks to Andrew for taking notes.
>>>=20
>>>=20
>>>=20
>>> Here are the draft minutes of the meeting, please send corrections =
/improvements
>>>=20
>>> thanks
>>>=20
>>> phil/
>>>=20
>>>=20
>>>=20
>>> Minutes of MPTCP WG Meeting
>>>=20
>>> WEDNESDAY, July 27, 2011
>>> 0900-1130  Morning Session I
>>>=20
>>> Main aim:
>>> Conclude on all protocol issues. So that we can WG last call =
protocol (& API) docs immediately after IETF.
>>>=20
>>>=20
>>>=20
>>> 1. Agenda bashing, note well, Milestones - Chairs
>>>=20
>>>=20
>>>=20
>>> 2. API update - Michael Scharf
>>>=20
>>> Doc ready for last call, no comments.
>>>=20
>>>=20
>>>=20
>>> 3. Congestion control update - Mark Handley
>>> Done. Minor comments from IESG and David Black have been resolved.
>>>=20
>>>=20
>>>=20
>>> 4. Protocol update & discussion of open issues - Mark Handley
>>>=20
>>> Reviewers could review section 2 (overview)
>>>=20
>>> Small changes to protocol
>>>=20
>>>=20
>>>=20
>>> Discussion over sending key-A on the third Ack; there is an issue =
over the reliable delivery of the key.
>>>=20
>>> Jana: B could echo back the options.
>>>=20
>>> Mark: The issue is how to trigger stopping option sends.  We will =
double-check if the issue is covered and resolve if required.
>>>=20
>>>=20
>>>=20
>>> Discussion on mobility in fallback mode
>>>=20
>>> It doesn't seem that it is possible to reliably use mobility in =
fallback mode.
>>>=20
>>> Andrew: Isn't that OK?  Can't you just say, if the ACKs haven't =
arrived, oh well, RST it.
>>>=20
>>> Georg: The draft says that there's no more multipath signalling in =
fallback.  So you can't get here.
>>>=20
>>> Mark: So, the only resolution seems to be that if you end up with =
un-ACKed data, you lose.
>>>=20
>>> Costin: You have signalling in the SYNs
>>>=20
>>> Mark: Which gives you a way out, if that is possible.
>>>=20
>>> Georg: I have a proposal for how to cover this later.
>>>=20
>>> Mark: It probably isn't worth trying to optimise this case.
>>>=20
>>> What to do if all subflows are in time out?
>>>=20
>>> Phil: You're saying that is an implementation specific heuristic?
>>>=20
>>> Andrew: What say no interfaces are up?
>>>=20
>>> Mark: Yes, doesn't need specified.
>>>=20
>>>=20
>>> Phil: Plan is, get these three issues resolved on the list, give it =
a week or so, and WGLC it.
>>>=20
>>>=20
>>>=20
>>>=20
>>> Experiments on "Is MPTCP deployable?"
>>>=20
>>> Georg: You can't name service providers or enterprises, to what =
extent are the harmful boxes installed in providers vs enterprises?
>>>=20
>>> Mark: Cellular providers have more, but no monopoly.  We can't claim =
this to be completely representative.
>>>=20
>>>=20
>>> 5. (new draft) Enhancements to Improve the Applicability of =
Multipath TCP to Wireless Access Networks - Georg Hampel
>>>=20
>>> Mark: Best path probably has lowest energy per byte
>>>=20
>>> Georg: If you have money or energy to burn, you might not care
>>>=20
>>> Mark: Proxy.  One part of me is revolted by supporting transparent =
proxies explicitly, and we should support them.  Calling the flag =
I_AM_A_PROXY is more explicit.
>>>=20
>>> Andrew: I agree there are always proxy devices "there"
>>> Andrew: ... there are good features here (is it important to tag =
this as "min" complexity, images are getting big for this part of the =
stack)
>>> Georg: How about recv buffer reduction?
>>> Andrew: good idea
>>>=20
>>> Mark: I'm not totally convinced that buffer reduction is real.
>>>=20
>>> Costin: Receive buffer reduction is orthogonal to implementation =
complexity
>>>=20
>>> Mark: The other question, if you're in path selection mode, do you =
get the right behaviour over mobile networks.  Assert: you should be =
able to tell from signal strength etc. if it is a good link... not =
really.  I can't easily predict from wireless if it's decent, I have to =
actually try it.
>>>=20
>>> Georg: The signal is not necessarily indicative, could have other =
variables
>>>=20
>>> Andrew: Wifi links have to train up, which can take a few hundred =
packets
>>>=20
>>> David: I have in mind a satellite PEP setup, where RTT could be =
radically different.
>>>=20
>>> Mark: Our intuition is that you should only use one path most of the =
time, but you want to transition make-before-break and do a soft =
transition.  You need to get the links adequately probed.
>>>=20
>>> Georg: Both of our points are based on speculation, we need to know =
how well you do if you shorten the overlap
>>>=20
>>> Costin: that paper assumed there is an oracle that knows the best =
path that chooses the best path yes: 15% compared to an optimal path =
selection algorithm that can't be achieved in practice
>>>=20
>>> Costin: you can also have redundant data if you want to probe and =
avoid hol blocking
>>>=20
>>> Mark: The spec does not prohibit you sending data on multiple paths
>>>=20
>>> Phil: We've had quite a lot of discussion now on heuristic stuff, =
how you might make this work in practice.  It would be nice to have more =
experiments.  My impression is that BTO doesn't seem to gain very much
>>>=20
>>> Andrew: Suggest not doing bulk transfer optimisation, it could be =
really expensive for little gain.
>>>=20
>>> Andrew: Very much in favour of proxy flags, they're everywhere
>>>=20
>>> Mark: I am in favour of adding a bit to support transparent proxies, =
1) address of proxy 2) remote supports mptcp - could decide based on =
this
>>>=20
>>> Andrew: It can be a content inspecting firewall that just wants to =
do its job
>>>=20
>>> Georg: Proxy is not optimisation, it's a feature
>>>=20
>>> Michio: No such middle box
>>>=20
>>> Mark: If we can preserve the benefits in the presence of boxes that =
want to be on path, then we should.  If we don't want the content =
inspected, we need to encrypt.
>>>=20
>>> Georg: 3G providers still get to bill it.
>>>=20
>>> Gorry: There are proxies that need to inform L2 what is going on.
>>>=20
>>> Andrew: Could have network architectures that wouldn't otherwise =
work that way.
>>>=20
>>> Mark: We should probably try to address this before we move the =
protocol forward.
>>>=20
>>> Phil: Would like to see people get together in the remainder of the =
week.
>>>=20
>>> Andrew: I think the bottom three lines of the slide are OK, up to =
naming.
>>>=20
>>> Costin: +1
>>>=20
>>>=20
>>>=20
>>> 6. Call for news on Implementations
>>>=20
>>> Linux version being updated as http://mptcpinfo.ucl.ac.be, also for =
Nokia N900 & Android Nexus-one
>>>=20
>>> Tim: Implementation is not SMP capable, does anyone know if it is =
fixed?
>>>=20
>>> <apparently not, docs still say no SMP>
>>>=20
>>> Georg Hampel - in progress on 'lightweight' implementation
>>>=20
>>> Sebastien Roy: Oracle are working on a prototype in Solaris.  Don't =
know if it can be released yet.
>>>=20
>>>=20
>>>=20
>>> 7. Discussion on future of WG - Chairs
>>>=20
>>>=20
>>>=20
>>> Discussion last time (roughly) concluded that we pause for =
experiments & usage: WG hears about implementations, heuristics etc; =
hears about investigations on possible extensions; not do any protocol =
standards work. Re-charter in ~ 1year when understand better what the =
needs are & where the energy is
>>> No comments - Chairs will work with AD on re-chartering text, when =
the current charter is done.
>>>=20
>>>=20
>>=20
>=20
> _______________________________________________
> multipathtcp mailing list
> multipathtcp@ietf.org
> https://www.ietf.org/mailman/listinfo/multipathtcp


From yoshifumi.nishida@gmail.com  Fri Sep  9 03:13:17 2011
Return-Path: <yoshifumi.nishida@gmail.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C059C21F869C for <multipathtcp@ietfa.amsl.com>; Fri,  9 Sep 2011 03:13:16 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.91
X-Spam-Level: 
X-Spam-Status: No, score=-102.91 tagged_above=-999 required=5 tests=[AWL=0.067, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id nuWQe054x+Vb for <multipathtcp@ietfa.amsl.com>; Fri,  9 Sep 2011 03:13:16 -0700 (PDT)
Received: from mail-yw0-f44.google.com (mail-yw0-f44.google.com [209.85.213.44]) by ietfa.amsl.com (Postfix) with ESMTP id 1962721F8514 for <multipathtcp@ietf.org>; Fri,  9 Sep 2011 03:13:16 -0700 (PDT)
Received: by ywe9 with SMTP id 9so1618748ywe.31 for <multipathtcp@ietf.org>; Fri, 09 Sep 2011 03:15:10 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; bh=vYe3F2oryXXTWaT2/AMH95DeGumMS+ePdsAtuE8fQoU=; b=Ws5Hy/43hKKHrlDZLHFy2BJVogYohF8Tcq9RXV0/6NVk8iO53UM/nrSFrecGrq+N3u hjLPflsb5d1YrJ4xL7A5YAi2DQAhm6STMvTeBipKKkN0wKTHK3Vwu/nI6yHlZ15Z9RBO cKAubPqi5MkEc5mut3BzwhN2D4ezawX0QUhCw=
MIME-Version: 1.0
Received: by 10.231.47.17 with SMTP id l17mr2004887ibf.24.1315563310353; Fri, 09 Sep 2011 03:15:10 -0700 (PDT)
Sender: yoshifumi.nishida@gmail.com
Received: by 10.231.199.80 with HTTP; Fri, 9 Sep 2011 03:15:10 -0700 (PDT)
In-Reply-To: <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk>
Date: Fri, 9 Sep 2011 03:15:10 -0700
X-Google-Sender-Auth: BsJxPr-ccgga1WfIC5Oq3Dn6OiA
Message-ID: <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com>
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
To: "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 09 Sep 2011 10:13:17 -0000

Hi Costin,

2011/9/8 Costin Raiciu <C.RAICIU@cs.ucl.ac.uk>:

> The second question regards adding support for transparent proxies. In pa=
rticular, each endpoint should know, for each address of their peer, if it =
is talking to a proxy or the destination. This would allow a smooth transit=
ion from
> an initial proxy based deployment in cellular networks to worldwide suppo=
rt for MPTCP in servers. In particular, MPTCP could connect to the proxy
>
> The proposal is to add the P flag to ADD_ADDRESS, to allow middleboxes to=
 say they are on path.
>
> Currently ADD_ADDRESS does not have flag space;
> - we could add another octet for flag space (but this is not ideal since =
this option is already big enough for IPv6, as Georg pointed out)
> - or we could reduce the number of bits in the address ID by one. 256 add=
resses seem like a lot of interfaces to have at the same time; 128 should s=
uffice. We should explicitly allow the number to wrap, though, to support l=
ong term mobility where a mobile's Wifi interface comes up and down hundred=
s of times. If we do this, we also free one bit on the MP_JOIN.
>
> My inclination is to go for the second option.
>
> When I first thought of it, it seemed to me that adding the P flag only t=
o add_address is not sufficient; the host does not know upon directly openi=
ng a connection/subflow whether it is talking directly to the destination o=
r a proxy.
> It turns out that using add address the proxy can make itself known for t=
he initial path. This path will have address ID 0; the proxy will just send=
 an add address with ID 0 and it's real address. The receiver now compares =
this address with what he thinks the remote endpoint is, and can infer it i=
s talking to a proxy on that path. The same applies if a new subflow is ope=
ned.
>
> What do people think? Are these reasonable solutions? If people agree, I =
will go ahead and rev the draft accordingly.

I tend to prefer the second option, but you don't want to use another
1 or 2 bits for future purpose?
If we should explicitly allow the number to wrap, it might not be a
big problem for having smaller ID space.

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp

From georg.hampel@alcatel-lucent.com  Fri Sep  9 06:56:07 2011
Return-Path: <georg.hampel@alcatel-lucent.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 585BC21F8490 for <multipathtcp@ietfa.amsl.com>; Fri,  9 Sep 2011 06:56:07 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.554
X-Spam-Level: 
X-Spam-Status: No, score=-6.554 tagged_above=-999 required=5 tests=[AWL=0.045,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bDKB8-YZj0Ba for <multipathtcp@ietfa.amsl.com>; Fri,  9 Sep 2011 06:56:06 -0700 (PDT)
Received: from ihemail4.lucent.com (ihemail4.lucent.com [135.245.0.39]) by ietfa.amsl.com (Postfix) with ESMTP id 9267421F8463 for <multipathtcp@ietf.org>; Fri,  9 Sep 2011 06:56:06 -0700 (PDT)
Received: from usnavsmail1.ndc.alcatel-lucent.com (usnavsmail1.ndc.alcatel-lucent.com [135.3.39.9]) by ihemail4.lucent.com (8.13.8/IER-o) with ESMTP id p89DvxKT025869 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Fri, 9 Sep 2011 08:57:59 -0500 (CDT)
Received: from USNAVSXCHHUB02.ndc.alcatel-lucent.com (usnavsxchhub02.ndc.alcatel-lucent.com [135.3.39.111]) by usnavsmail1.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id p89DvtWu024822 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT); Fri, 9 Sep 2011 08:57:59 -0500
Received: from USNAVSXCHMBSA2.ndc.alcatel-lucent.com ([135.3.39.124]) by USNAVSXCHHUB02.ndc.alcatel-lucent.com ([135.3.39.111]) with mapi; Fri, 9 Sep 2011 08:57:57 -0500
From: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>, "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
Date: Fri, 9 Sep 2011 08:57:56 -0500
Thread-Topic: [multipathtcp] Proposed changes to the protocol draft
Thread-Index: Acxu2V8YUnu0TB8RRyCXkD0WxY6RbQAGgu3w
Message-ID: <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com>
In-Reply-To: <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.39
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.9
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 09 Sep 2011 13:56:07 -0000

Hi Christoph, Costin, Yoshifumi, all,

Fallback issue: I agree that when checksums are not used, there is no way f=
or the end hosts to tell if payload has been rewritten, and hence, there's =
no reason to enter fallback due to payload rewriting. Obviously, if payload=
 rewriting still occurs, the connection may get corrupted or it may break.

I'm striving for something else: If checksums are *not* used and only one s=
ubflow runs traffic, DSS can principally be omitted. This reduces overhead =
and improves processing speed. Omission of DSS can be announced via DSS wit=
h infinity setting. DSS must be re-inserted as soon as another subflow carr=
ies traffic. This is done via conventional DSSs.=20

What problems does this simplification raise?


Proxy issue:=20
First a clarification (in my opinion): In presence of a proxy, there can *o=
nly* *ever* be two independent "end points" that do MPTCP signaling. If hos=
t A and host B do MPTCP, the proxy must remain out of the picture and let a=
ll MPTCP signaling pass. If the proxy wants to be in the picture, it has to=
 shun host A's MPTCP signaling from host B and respond on behalf of host A.

Let's assume the proxy is in the picture. In this case, the first subflow r=
uns between host A's and host B's IP addresses with address-id=3D0 assigned=
 to both addresses. Host B's address *cannot* be used for any further subfl=
ow. If the first subflow gets torn down, it cannot be reestablished within =
the same connection. Therefore, apart from supporting the first subflow, ho=
st A has *no* use for Host-B's address. If the proxy forwarded to host A it=
s own address with address-id=3D0, it would clearly mark that (1) its own a=
ddress should replace address-id=3D0 and (2) host B's IP address cannot be =
used for any further subflows. No need for any further flag or byte.

Am I missing something? Thanks.


Georg
=20



-----Original Message-----
From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org] =
On Behalf Of Yoshifumi Nishida
Sent: Friday, September 09, 2011 6:15 AM
To: multipathtcp@ietf.org List
Subject: Re: [multipathtcp] Proposed changes to the protocol draft

Hi Costin,

2011/9/8 Costin Raiciu <C.RAICIU@cs.ucl.ac.uk>:

> The second question regards adding support for transparent proxies. In pa=
rticular, each endpoint should know, for each address of their peer, if it =
is talking to a proxy or the destination. This would allow a smooth transit=
ion from
> an initial proxy based deployment in cellular networks to worldwide suppo=
rt for MPTCP in servers. In particular, MPTCP could connect to the proxy
>
> The proposal is to add the P flag to ADD_ADDRESS, to allow middleboxes to=
 say they are on path.
>
> Currently ADD_ADDRESS does not have flag space;
> - we could add another octet for flag space (but this is not ideal since =
this option is already big enough for IPv6, as Georg pointed out)
> - or we could reduce the number of bits in the address ID by one. 256 add=
resses seem like a lot of interfaces to have at the same time; 128 should s=
uffice. We should explicitly allow the number to wrap, though, to support l=
ong term mobility where a mobile's Wifi interface comes up and down hundred=
s of times. If we do this, we also free one bit on the MP_JOIN.
>
> My inclination is to go for the second option.
>
> When I first thought of it, it seemed to me that adding the P flag only t=
o add_address is not sufficient; the host does not know upon directly openi=
ng a connection/subflow whether it is talking directly to the destination o=
r a proxy.
> It turns out that using add address the proxy can make itself known for t=
he initial path. This path will have address ID 0; the proxy will just send=
 an add address with ID 0 and it's real address. The receiver now compares =
this address with what he thinks the remote endpoint is, and can infer it i=
s talking to a proxy on that path. The same applies if a new subflow is ope=
ned.
>
> What do people think? Are these reasonable solutions? If people agree, I =
will go ahead and rev the draft accordingly.

I tend to prefer the second option, but you don't want to use another
1 or 2 bits for future purpose?
If we should explicitly allow the number to wrap, it might not be a
big problem for having smaller ID space.

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp
_______________________________________________
multipathtcp mailing list
multipathtcp@ietf.org
https://www.ietf.org/mailman/listinfo/multipathtcp

From yoshifumi.nishida@gmail.com  Fri Sep  9 16:28:04 2011
Return-Path: <yoshifumi.nishida@gmail.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6B46321F87FC for <multipathtcp@ietfa.amsl.com>; Fri,  9 Sep 2011 16:28:04 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.914
X-Spam-Level: 
X-Spam-Status: No, score=-102.914 tagged_above=-999 required=5 tests=[AWL=0.063, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id v8+8KVGTQK2S for <multipathtcp@ietfa.amsl.com>; Fri,  9 Sep 2011 16:28:03 -0700 (PDT)
Received: from mail-gy0-f172.google.com (mail-gy0-f172.google.com [209.85.160.172]) by ietfa.amsl.com (Postfix) with ESMTP id C9A6A21F8744 for <multipathtcp@ietf.org>; Fri,  9 Sep 2011 16:28:03 -0700 (PDT)
Received: by gyd12 with SMTP id 12so2262836gyd.31 for <multipathtcp@ietf.org>; Fri, 09 Sep 2011 16:29:57 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=bI0hfK4XbDnawPvoHspEiS+ZZIK77XjlJghWip8wVxs=; b=rqqIpDVVvhYjkEVc15d0VdkoPO1F6mWuSbG7vZSc4ow6QG/jKIxMaOgAl2DYCweE9p AGc1QeJJruVyhzbr1+S/aD57QLAFV/Vw0AT5ZypY2Hskaz7Mo68q6zo2CKgCGCVz72+s 8I5NMAjROASBfU1kLD7YOCocbgPN6x2iwxedM=
MIME-Version: 1.0
Received: by 10.231.24.66 with SMTP id u2mr3118073ibb.11.1315610997690; Fri, 09 Sep 2011 16:29:57 -0700 (PDT)
Sender: yoshifumi.nishida@gmail.com
Received: by 10.231.199.80 with HTTP; Fri, 9 Sep 2011 16:29:57 -0700 (PDT)
In-Reply-To: <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
Date: Fri, 9 Sep 2011 16:29:57 -0700
X-Google-Sender-Auth: UbOJbKmzKqXtd8_P7e1M8w40K7Q
Message-ID: <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com>
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
To: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 09 Sep 2011 23:28:04 -0000

Hi Georg,

2011/9/9 Hampel, K Georg (K Georg) <georg.hampel@alcatel-lucent.com>:
> Hi Christoph, Costin, Yoshifumi, all,
>
> Fallback issue: I agree that when checksums are not used, there is no way=
 for the end hosts to tell if payload has been rewritten, and hence, there'=
s no reason to enter fallback due to payload rewriting. Obviously, if paylo=
ad rewriting still occurs, the connection may get corrupted or it may break=
.
>
> I'm striving for something else: If checksums are *not* used and only one=
 subflow runs traffic, DSS can principally be omitted. This reduces overhea=
d and improves processing speed. Omission of DSS can be announced via DSS w=
ith infinity setting. DSS must be re-inserted as soon as another subflow ca=
rries traffic. This is done via conventional DSSs.
>
> What problems does this simplification raise?

Please correct me if I'm wrong.
In my understanding, there seems to be two distinct modes in this.

   fallback mode:  where we want to omit MPTCP signaling since it
might not be very safe.
   one path mode: where we want to suppress MPTCP signaling to reduce
the overhead and improve the performance.

I'm so far inclined to disable all MPTCP features (including opening
subflows) in the fallback mode.
For one path mode, I'm thinking that it's possible to implement it
without violating the current spec.
So, I think that it's quite fine to have this mode in implementations.

> Proxy issue:
> First a clarification (in my opinion): In presence of a proxy, there can =
*only* *ever* be two independent "end points" that do MPTCP signaling. If h=
ost A and host B do MPTCP, the proxy must remain out of the picture and let=
 all MPTCP signaling pass. If the proxy wants to be in the picture, it has =
to shun host A's MPTCP signaling from host B and respond on behalf of host =
A.

Well, I'm not sure if this is a good example, But, a mobile MPTCP node
in future might want to say something like this...
"I'm going to leave this network and will come back with new address
after for a while. In the meantime, you can send data to this proxy if
you want"

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp

From olivier.bonaventure@uclouvain.be  Sat Sep 10 02:41:30 2011
Return-Path: <olivier.bonaventure@uclouvain.be>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2EF9621F856A for <multipathtcp@ietfa.amsl.com>; Sat, 10 Sep 2011 02:41:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -5.999
X-Spam-Level: 
X-Spam-Status: No, score=-5.999 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, J_CHICKENPOX_38=0.6, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kiYgH71BXlBI for <multipathtcp@ietfa.amsl.com>; Sat, 10 Sep 2011 02:41:29 -0700 (PDT)
Received: from smtp5.sgsi.ucl.ac.be (smtp.sgsi.ucl.ac.be [130.104.5.67]) by ietfa.amsl.com (Postfix) with ESMTP id F16F421F84DF for <multipathtcp@ietf.org>; Sat, 10 Sep 2011 02:41:19 -0700 (PDT)
Received: from mbpobo.local (host-85-27-91-91.brutele.be [85.27.91.91]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: obonaventure@smtp5.sgsi.ucl.ac.be) by smtp5.sgsi.ucl.ac.be (Postfix) with ESMTPSA id 6A89A11E45C; Sat, 10 Sep 2011 11:43:10 +0200 (CEST)
X-DKIM: Sendmail DKIM Filter v2.8.3 smtp5.sgsi.ucl.ac.be 6A89A11E45C
DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=uclouvain.be; s=selucl; t=1315647790; bh=CUg+lHPq7yVR3W77Je9IFVD72PRbutr5mpl/xuVAGlQ=; h=Message-ID:Date:From:Reply-To:MIME-Version:To:CC:Subject: References:In-Reply-To:Content-Type:Content-Transfer-Encoding; b=dIObngRD61HhEPazQwD6hA+XmqtWulF9LAZMyjZ/H0ZI2Qy0DkROksTIiG1AgFrqo ax8DKnYTYOvlZiJVAT8L+d3WGY44YkQbP1eRDslY9ezfOGKCpXBAMv+ITzqWO4Uo6U izFnuGNvbiou9ws12JxHWEDrkyBV356QsQATL1xA=
Message-ID: <4E6B312E.8060707@uclouvain.be>
Date: Sat, 10 Sep 2011 11:43:10 +0200
From: Olivier Bonaventure <Olivier.Bonaventure@uclouvain.be>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:6.0.1) Gecko/20110830 Thunderbird/6.0.1
MIME-Version: 1.0
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com>
In-Reply-To: <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com>
X-Enigmail-Version: 1.3.1
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: clamav-milter 0.97-exp at smtp-5.sipr-dc.ucl.ac.be
X-Virus-Status: Clean
X-Sgsi-Spamcheck: SASL authenticated, 
X-SGSI-MailScanner-ID: 6A89A11E45C.A367D
X-SGSI-MailScanner: Found to be clean
X-SGSI-From: olivier.bonaventure@uclouvain.be
X-SGSI-Spam-Status: No
Cc: "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: Olivier.Bonaventure@uclouvain.be
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sat, 10 Sep 2011 09:41:30 -0000

Hello,
> 
> 2011/9/9 Hampel, K Georg (K Georg) <georg.hampel@alcatel-lucent.com>:
>> Hi Christoph, Costin, Yoshifumi, all,
>>
>> Fallback issue: I agree that when checksums are not used, there is no way for the end hosts to tell if payload has been rewritten, and hence, there's no reason to enter fallback due to payload rewriting. Obviously, if payload rewriting still occurs, the connection may get corrupted or it may break.
>>
>> I'm striving for something else: If checksums are *not* used and only one subflow runs traffic, DSS can principally be omitted. 

We'd need to define precisely define when there is a single subflow that
runs traffic.

Does this means that only one flow has been established ? or only one
subflow is active (e.g. previous subflows have timedout) or there
unacknowledged data on only one subflow ? or another definition ?

We need an unambiguous definition that can be easily supported by
implementations

> This reduces overhead and improves processing speed. 

But this increases the implementation complexity and the number of
corner cases that have to be considered by implementors.

>Omission of DSS can be announced via DSS with infinity setting. DSS must be re-inserted as soon as another subflow carries traffic. This is done via conventional DSSs.

Implementations will also need to hanlde the case when :
- at time t0, there is a single subflow and thus dss=infinite
- at time t1, a second subflow is used and thus we send dss
- at time t2, we go back to a single subflow and dss=infinite

I fear that this would add unnecessary complexity and would prefer to
wait for more implementation experience before looking at such
optimisations in the main spec


>> What problems does this simplification raise?
> 
> Please correct me if I'm wrong.
> In my understanding, there seems to be two distinct modes in this.
> 
>    fallback mode:  where we want to omit MPTCP signaling since it
> might not be very safe.

Agree. As soon as we go in fallback mode, we must stop all MPTCP signalling.

>    one path mode: where we want to suppress MPTCP signaling to reduce
> the overhead and improve the performance.
> 
> I'm so far inclined to disable all MPTCP features (including opening
> subflows) in the fallback mode.

Agreed


>> Proxy issue:
>> First a clarification (in my opinion): In presence of a proxy, there can *only* *ever* be two independent "end points" that do MPTCP signaling. If host A and host B do MPTCP, the proxy must remain out of the picture and let all MPTCP signaling pass. If the proxy wants to be in the picture, it has to shun host A's MPTCP signaling from host B and respond on behalf of host A.
> 
> Well, I'm not sure if this is a good example, But, a mobile MPTCP node
> in future might want to say something like this...
> "I'm going to leave this network and will come back with new address
> after for a while. In the meantime, you can send data to this proxy if
> you want"

Adding proxy to the main spec will add complexity to the protocol.
Wouldn't it be better to :
- reserve a few bits in the main spec to have some space to add the
proxy bits that are being discussed but leave proxy out of the main spec
- start a new draft that works on these proxy issues

MPTCP has already several use cases without proxies and I think that it
is important to stabilise the main MPTCP spec based on implementation
experience before trying to add new features and new complexity.



Olivier


-- 
INL, ICTEAM, UCLouvain, Belgium, http://inl.info.ucl.ac.be

From c.raiciu@cs.ucl.ac.uk  Sun Sep 11 07:07:27 2011
Return-Path: <c.raiciu@cs.ucl.ac.uk>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C7DF821F84ED for <multipathtcp@ietfa.amsl.com>; Sun, 11 Sep 2011 07:07:27 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.403
X-Spam-Level: 
X-Spam-Status: No, score=-1.403 tagged_above=-999 required=5 tests=[AWL=1.196,  BAYES_00=-2.599]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 61kJa+E+9F54 for <multipathtcp@ietfa.amsl.com>; Sun, 11 Sep 2011 07:07:27 -0700 (PDT)
Received: from bells2.cs.ucl.ac.uk (bells2.cs.ucl.ac.uk [128.16.5.33]) by ietfa.amsl.com (Postfix) with ESMTP id 44E8E21F85F1 for <multipathtcp@ietf.org>; Sun, 11 Sep 2011 07:07:27 -0700 (PDT)
Received: from [92.80.121.178] (helo=[192.168.1.7]) by bells2.cs.ucl.ac.uk with esmtpsa (TLSv1:AES128-SHA:128) (C.Raiciu authenticated) (Exim 4.54) id 1R2kij-0001Yn-HY; Sun, 11 Sep 2011 15:09:17 +0100
Mime-Version: 1.0 (Apple Message framework v1078)
Content-Type: text/plain; charset=us-ascii
From: Costin Raiciu <c.raiciu@cs.ucl.ac.uk>
In-Reply-To: <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com>
Date: Sun, 11 Sep 2011 17:09:20 +0300
Content-Transfer-Encoding: quoted-printable
Message-Id: <6E1F1C00-4D8D-464A-861A-0D70FD763428@cs.ucl.ac.uk>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com>
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
X-Mailer: Apple Mail (2.1078)
Cc: "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 11 Sep 2011 14:07:27 -0000

Hi,

I agree that single path mode could be made to work if we use infinite =
mappings and a lot of care to cover the cases when there are two =
subflows.

But we can cover the single path mode by using really long data sequence =
mappings - as long as you want them. And that needs to further thought =
on corner cases, and does not add further complexity to MPTCP.
Sending one option every few RTTs seems like a reasonable overhead to =
me.

As for the proxies, all I am proposing is to add a bit to ADD_ADDRESS; =
that seems to cover all the signaling we need for transparent proxies, =
which for sure will be used in cellular nets if mptcp is deployed on =
mobiles.

Cheers.
Costin=

From lars.eggert@nokia.com  Sun Sep 11 23:25:35 2011
Return-Path: <lars.eggert@nokia.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 981C121F84BD for <multipathtcp@ietfa.amsl.com>; Sun, 11 Sep 2011 23:25:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.884
X-Spam-Level: 
X-Spam-Status: No, score=-102.884 tagged_above=-999 required=5 tests=[AWL=-0.285, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RnhWPdFxn2CY for <multipathtcp@ietfa.amsl.com>; Sun, 11 Sep 2011 23:25:34 -0700 (PDT)
Received: from mgw-da02.nokia.com (smtp.nokia.com [147.243.128.26]) by ietfa.amsl.com (Postfix) with ESMTP id A09AA21F84A9 for <multipathtcp@ietf.org>; Sun, 11 Sep 2011 23:25:34 -0700 (PDT)
Received: from mail.fit.nokia.com (esdhcp030222.research.nokia.com [172.21.30.222]) by mgw-da02.nokia.com (Switch-3.4.4/Switch-3.4.3) with ESMTP id p8C6RX5m019992 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Mon, 12 Sep 2011 09:27:35 +0300
X-Virus-Status: Clean
X-Virus-Scanned: clamav-milter 0.97.2 at fit.nokia.com
Mime-Version: 1.0 (Apple Message framework v1244.3)
Content-Type: multipart/signed; boundary="Apple-Mail=_7122305A-5AFD-494E-BC91-D07609618840"; protocol="application/pkcs7-signature"; micalg=sha1
From: Lars Eggert <lars.eggert@nokia.com>
In-Reply-To: <4E6B312E.8060707@uclouvain.be>
Date: Mon, 12 Sep 2011 09:27:31 +0300
Message-Id: <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com> <4E6B312E.8060707@uclouvain.be>
To: <Olivier.Bonaventure@uclouvain.be>
X-Mailer: Apple Mail (2.1244.3)
X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.7 (mail.fit.nokia.com); Mon, 12 Sep 2011 09:27:31 +0300 (EEST)
X-Nokia-AV: Clean
Cc: "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 12 Sep 2011 06:25:35 -0000

--Apple-Mail=_7122305A-5AFD-494E-BC91-D07609618840
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

On 2011-9-10, at 12:43, Olivier Bonaventure wrote:
> I fear that this would add unnecessary complexity and would prefer to
> wait for more implementation experience before looking at such
> optimisations in the main spec

This is my feeling at the moment, too. I think it'd be useful to have =
this written up in an individual ID so that folks can discuss (and maybe =
try and implement) it, but I think it's too early to put this into the =
main spec.

Lars=

--Apple-Mail=_7122305A-5AFD-494E-BC91-D07609618840
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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--Apple-Mail=_7122305A-5AFD-494E-BC91-D07609618840--

From georg.hampel@alcatel-lucent.com  Mon Sep 12 06:38:35 2011
Return-Path: <georg.hampel@alcatel-lucent.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5741C21F8B53 for <multipathtcp@ietfa.amsl.com>; Mon, 12 Sep 2011 06:38:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.556
X-Spam-Level: 
X-Spam-Status: No, score=-6.556 tagged_above=-999 required=5 tests=[AWL=0.043,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OEaRHetmFpOx for <multipathtcp@ietfa.amsl.com>; Mon, 12 Sep 2011 06:38:34 -0700 (PDT)
Received: from ihemail1.lucent.com (ihemail1.lucent.com [135.245.0.33]) by ietfa.amsl.com (Postfix) with ESMTP id B5EBD21F8B49 for <multipathtcp@ietf.org>; Mon, 12 Sep 2011 06:38:34 -0700 (PDT)
Received: from usnavsmail3.ndc.alcatel-lucent.com (usnavsmail3.ndc.alcatel-lucent.com [135.3.39.11]) by ihemail1.lucent.com (8.13.8/IER-o) with ESMTP id p8CDebi7019511 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 12 Sep 2011 08:40:37 -0500 (CDT)
Received: from USNAVSXCHHUB02.ndc.alcatel-lucent.com (usnavsxchhub02.ndc.alcatel-lucent.com [135.3.39.111]) by usnavsmail3.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id p8CDeXpX009394 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT); Mon, 12 Sep 2011 08:40:36 -0500
Received: from USNAVSXCHMBSA2.ndc.alcatel-lucent.com ([135.3.39.124]) by USNAVSXCHHUB02.ndc.alcatel-lucent.com ([135.3.39.111]) with mapi; Mon, 12 Sep 2011 08:40:35 -0500
From: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
To: Lars Eggert <lars.eggert@nokia.com>, "Olivier.Bonaventure@uclouvain.be" <Olivier.Bonaventure@uclouvain.be>
Date: Mon, 12 Sep 2011 08:40:33 -0500
Thread-Topic: [multipathtcp] Proposed changes to the protocol draft
Thread-Index: AcxxFRTLA0u06POXT5C3FALYMUJCXgAOSriA
Message-ID: <154773479ED2314980CB638A48FC443486E8996A@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com> <4E6B312E.8060707@uclouvain.be> <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com>
In-Reply-To: <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.33
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.11
Cc: "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 12 Sep 2011 13:38:35 -0000

Hi Yoshifumi, Olivier, Costin, Lars, all,

I tend to agree with Olivier and Lars: Transparent proxy as well as "single=
-path mode" are new features, which extend the present architecture/design.=
 We should add room for extensions to ADD_ADDRESS and leave it up to furthe=
r IDs, discussions, implementations etc to decide on further details.


On the discussion of these features:

I agree that it makes sense to differentiate between "single-path" mode and=
 "fallback mode". The expression "single-path" mode leads to misunderstandi=
ngs, however, as Olivier pointed out. In fact, it should be allowed to have=
 multiple subflows up, but only one of them can carry traffic.

Using bulk optimization with large range as proposed by Costin may lead to =
problems if sudden handover is needed (e.g. sudden break before make).

Working on an implementation right now, I canNOT confirm that "single path"=
 mode is more complex than the present bulk optimization.

I think that proxy issues brought up by Yoshifumi ("use proxy until I'm bac=
k") require further discussion. I don't understand the scenario: Where woul=
d the proxy ship the data if the host temporarily moves to a different addr=
ess? Why can't he host send his temporary address instead of the proxy's ad=
dress?
Thanks.

Regards,

Georg


-----Original Message-----
From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org] =
On Behalf Of Lars Eggert
Sent: Monday, September 12, 2011 2:28 AM
To: Olivier.Bonaventure@uclouvain.be
Cc: multipathtcp@ietf.org List
Subject: Re: [multipathtcp] Proposed changes to the protocol draft

On 2011-9-10, at 12:43, Olivier Bonaventure wrote:
> I fear that this would add unnecessary complexity and would prefer to
> wait for more implementation experience before looking at such
> optimisations in the main spec

This is my feeling at the moment, too. I think it'd be useful to have this =
written up in an individual ID so that folks can discuss (and maybe try and=
 implement) it, but I think it's too early to put this into the main spec.

Lars

From yoshifumi.nishida@gmail.com  Tue Sep 13 02:33:45 2011
Return-Path: <yoshifumi.nishida@gmail.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5BAC721F8B54 for <multipathtcp@ietfa.amsl.com>; Tue, 13 Sep 2011 02:33:45 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -100.989
X-Spam-Level: 
X-Spam-Status: No, score=-100.989 tagged_above=-999 required=5 tests=[AWL=-1.870, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, FRT_PROFIT1=3.858, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BTEMYqhXf3jr for <multipathtcp@ietfa.amsl.com>; Tue, 13 Sep 2011 02:33:44 -0700 (PDT)
Received: from mail-yw0-f44.google.com (mail-yw0-f44.google.com [209.85.213.44]) by ietfa.amsl.com (Postfix) with ESMTP id B5E7A21F8ADE for <multipathtcp@ietf.org>; Tue, 13 Sep 2011 02:33:38 -0700 (PDT)
Received: by ywa6 with SMTP id 6so311427ywa.31 for <multipathtcp@ietf.org>; Tue, 13 Sep 2011 02:35:44 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=XYg9jkEw1+VBw8TfrO9m+/WVXXpgpMeHnP4joIZE3ho=; b=WxgzhfVg9OStzbEbDiIpL14lk91clltq5CV8Ql/GrtYSnV+AWEprfETsRI4F+sGgk6 Ylv7qHU5edQa5MGjuroeT5yZzEKZV20Pwd7bhkRMRuwwj4aCd7LgkgAOS2VVAVTr/TyE IarnVLcW2B9qmjY8lF7amAar+kE4Su91DgRJk=
MIME-Version: 1.0
Received: by 10.231.42.139 with SMTP id s11mr5129538ibe.24.1315906544059; Tue, 13 Sep 2011 02:35:44 -0700 (PDT)
Sender: yoshifumi.nishida@gmail.com
Received: by 10.231.199.80 with HTTP; Tue, 13 Sep 2011 02:35:43 -0700 (PDT)
In-Reply-To: <154773479ED2314980CB638A48FC443486E8996A@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com> <4E6B312E.8060707@uclouvain.be> <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com> <154773479ED2314980CB638A48FC443486E8996A@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
Date: Tue, 13 Sep 2011 02:35:44 -0700
X-Google-Sender-Auth: WtGdy1twAoo657gR88dNOX8SCvo
Message-ID: <CABxaRLHmQt_kN1mj5mfdozcoe03+PPiiZN4d4OVGonzWdUNkpQ@mail.gmail.com>
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
To: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 13 Sep 2011 09:33:45 -0000

Hi Georg,

2011/9/12 Hampel, K Georg (K Georg) <georg.hampel@alcatel-lucent.com>:
> Hi Yoshifumi, Olivier, Costin, Lars, all,
>
> I tend to agree with Olivier and Lars: Transparent proxy as well as "sing=
le-path mode" are new features, which extend the present architecture/desig=
n. We should add room for extensions to ADD_ADDRESS and leave it up to furt=
her IDs, discussions, implementations etc to decide on further details.
>
>
> On the discussion of these features:
>
> I agree that it makes sense to differentiate between "single-path" mode a=
nd "fallback mode". The expression "single-path" mode leads to misunderstan=
dings, however, as Olivier pointed out. In fact, it should be allowed to ha=
ve multiple subflows up, but only one of them can carry traffic.
>
> Using bulk optimization with large range as proposed by Costin may lead t=
o problems if sudden handover is needed (e.g. sudden break before make).
>
> Working on an implementation right now, I canNOT confirm that "single pat=
h" mode is more complex than the present bulk optimization.

I cannot confirm, either.
Probably, the important point at this moment is we don't have to
change the main spec for it.
You can proof it through your experiments later and we can discuss it.
But, If you find problems in fallback mode, I think we'll need to
discuss now since it's currently in the spec.

>
> I think that proxy issues brought up by Yoshifumi ("use proxy until I'm b=
ack") require further discussion. I don't understand the scenario: Where wo=
uld the proxy ship the data if the host temporarily moves to a different ad=
dress? Why can't he host send his temporary address instead of the proxy's =
address?

It was just an example.. I may be wrong, but I was trying to say that
there might be more than 2 nodes in future mptcp signaling exchange.
I thought presuming only 2 nodes in mptcp signaling might not be a
good idea if we are trying to make the spec extensible.
Sorry if this brings confusion.

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp

From georg.hampel@alcatel-lucent.com  Tue Sep 13 06:27:42 2011
Return-Path: <georg.hampel@alcatel-lucent.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D3F6321F8AE9 for <multipathtcp@ietfa.amsl.com>; Tue, 13 Sep 2011 06:27:42 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.629
X-Spam-Level: 
X-Spam-Status: No, score=-4.629 tagged_above=-999 required=5 tests=[AWL=-1.888, BAYES_00=-2.599, FRT_PROFIT1=3.858, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id W2Jd-BmDrkVv for <multipathtcp@ietfa.amsl.com>; Tue, 13 Sep 2011 06:27:42 -0700 (PDT)
Received: from ihemail3.lucent.com (ihemail3.lucent.com [135.245.0.37]) by ietfa.amsl.com (Postfix) with ESMTP id 36B1D21F8AF2 for <multipathtcp@ietf.org>; Tue, 13 Sep 2011 06:27:41 -0700 (PDT)
Received: from usnavsmail2.ndc.alcatel-lucent.com (usnavsmail2.ndc.alcatel-lucent.com [135.3.39.10]) by ihemail3.lucent.com (8.13.8/IER-o) with ESMTP id p8DDTiCd026847 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Tue, 13 Sep 2011 08:29:45 -0500 (CDT)
Received: from USNAVSXCHHUB02.ndc.alcatel-lucent.com (usnavsxchhub02.ndc.alcatel-lucent.com [135.3.39.111]) by usnavsmail2.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id p8DDTgPS026434 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT); Tue, 13 Sep 2011 08:29:44 -0500
Received: from USNAVSXCHMBSA2.ndc.alcatel-lucent.com ([135.3.39.124]) by USNAVSXCHHUB02.ndc.alcatel-lucent.com ([135.3.39.111]) with mapi; Tue, 13 Sep 2011 08:29:42 -0500
From: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
Date: Tue, 13 Sep 2011 08:29:41 -0500
Thread-Topic: [multipathtcp] Proposed changes to the protocol draft
Thread-Index: Acxx+IWq44uAsdiSQNukQ7yu9SycBAAHooDA
Message-ID: <154773479ED2314980CB638A48FC443486E89E2E@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com> <4E6B312E.8060707@uclouvain.be> <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com> <154773479ED2314980CB638A48FC443486E8996A@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLHmQt_kN1mj5mfdozcoe03+PPiiZN4d4OVGonzWdUNkpQ@mail.gmail.com>
In-Reply-To: <CABxaRLHmQt_kN1mj5mfdozcoe03+PPiiZN4d4OVGonzWdUNkpQ@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.37
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.10
Cc: "multipathtcp@ietf.org List" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 13 Sep 2011 13:27:42 -0000

Hi Yoshifumi,

There seems to be the general notion that implementation and measurements h=
ave to catch up with the present specs before new features are added.

Both, "single path mode" as well as transparent-proxy support, would be suc=
h features.

Single-path mode further relies on the absence of payload-rewriting middle =
boxes which needs to be confirmed via trials.

Your comments on the transparent proxy confirm that more discussion is need=
ed in this area.

Thanks.

Georg

-----Original Message-----
From: yoshifumi.nishida@gmail.com [mailto:yoshifumi.nishida@gmail.com] On B=
ehalf Of Yoshifumi Nishida
Sent: Tuesday, September 13, 2011 5:36 AM
To: Hampel, K Georg (K Georg)
Cc: multipathtcp@ietf.org List
Subject: Re: [multipathtcp] Proposed changes to the protocol draft

Hi Georg,

2011/9/12 Hampel, K Georg (K Georg) <georg.hampel@alcatel-lucent.com>:
> Hi Yoshifumi, Olivier, Costin, Lars, all,
>
> I tend to agree with Olivier and Lars: Transparent proxy as well as "sing=
le-path mode" are new features, which extend the present architecture/desig=
n. We should add room for extensions to ADD_ADDRESS and leave it up to furt=
her IDs, discussions, implementations etc to decide on further details.
>
>
> On the discussion of these features:
>
> I agree that it makes sense to differentiate between "single-path" mode a=
nd "fallback mode". The expression "single-path" mode leads to misunderstan=
dings, however, as Olivier pointed out. In fact, it should be allowed to ha=
ve multiple subflows up, but only one of them can carry traffic.
>
> Using bulk optimization with large range as proposed by Costin may lead t=
o problems if sudden handover is needed (e.g. sudden break before make).
>
> Working on an implementation right now, I canNOT confirm that "single pat=
h" mode is more complex than the present bulk optimization.

I cannot confirm, either.
Probably, the important point at this moment is we don't have to
change the main spec for it.
You can proof it through your experiments later and we can discuss it.
But, If you find problems in fallback mode, I think we'll need to
discuss now since it's currently in the spec.


>
> I think that proxy issues brought up by Yoshifumi ("use proxy until I'm b=
ack") require further discussion. I don't understand the scenario: Where wo=
uld the proxy ship the data if the host temporarily moves to a different ad=
dress? Why can't he host send his temporary address instead of the proxy's =
address?

It was just an example.. I may be wrong, but I was trying to say that
there might be more than 2 nodes in future mptcp signaling exchange.
I thought presuming only 2 nodes in mptcp signaling might not be a
good idea if we are trying to make the spec extensible.
Sorry if this brings confusion.

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp

From yoshifumi.nishida@gmail.com  Fri Sep 16 00:45:39 2011
Return-Path: <yoshifumi.nishida@gmail.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7837B21F8BF9 for <multipathtcp@ietfa.amsl.com>; Fri, 16 Sep 2011 00:45:39 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.999
X-Spam-Level: 
X-Spam-Status: No, score=-2.999 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, J_CHICKENPOX_72=0.6, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZVKYqIWoa5vT for <multipathtcp@ietfa.amsl.com>; Fri, 16 Sep 2011 00:45:37 -0700 (PDT)
Received: from mail-iy0-f172.google.com (mail-iy0-f172.google.com [209.85.210.172]) by ietfa.amsl.com (Postfix) with ESMTP id AA1AF21F8BB9 for <multipathtcp@ietf.org>; Fri, 16 Sep 2011 00:45:37 -0700 (PDT)
Received: by iaby26 with SMTP id y26so2483894iab.31 for <multipathtcp@ietf.org>; Fri, 16 Sep 2011 00:47:51 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type:content-transfer-encoding; bh=XJnjGSfE/ky90DcgDThRiEQBO/LJQVv9ONcmssy650c=; b=k1ca+KZrTRY+JL2WgJXf2tMNxdKYaXu2DGTfk9gU/peuV78o8ALZ9cDZ+n+HT8FMD1 3tVAfanKdTR2jjmzZpFotkErcLJbXtp5NwAQe2ezdHQFPPjnFOnzoP9krpqdXCV7EnIH LKQ7Ytn2p3DZcIfO4vLyVxhiD5AZNzs6TwMOY=
MIME-Version: 1.0
Received: by 10.231.69.18 with SMTP id x18mr3488870ibi.37.1316159271304; Fri, 16 Sep 2011 00:47:51 -0700 (PDT)
Received: by 10.231.199.80 with HTTP; Fri, 16 Sep 2011 00:47:51 -0700 (PDT)
In-Reply-To: <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk>
Date: Fri, 16 Sep 2011 00:47:51 -0700
Message-ID: <CABxaRLG0yYkDGycjqfWGFNs46eHkrVZCFj2X1dy9ahBh89br_Q@mail.gmail.com>
From: Yoshifumi Nishida <yoshifumi.nishida@gmail.com>
To: multipathtcp <multipathtcp@ietf.org>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable
X-Mailman-Approved-At: Fri, 16 Sep 2011 00:47:21 -0700
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 16 Sep 2011 07:45:39 -0000

Hello,

It seems that we're reaching the stage to finalize the spec.
Please correct me if I'm wrong, but here's the current status I think..

Constin's proposal 1:
    "when MPTCP goes into fallback mode, it effectively becomes
equivalent to a regular TCP session; no further subflows can be setup;
teardown happens only with regular FIN messages, etc."

 seems to be reaching consensus as we don't see any objections so far.

Constin's proposal 2:
    "add the P flag to ADD_ADDRESS, to allow middleboxes to say they
are on path"

 seems to need more inputs to discuss where to put the flag, although
this looks relatively a minor issue.

In addition, I guess we need to discuss initial handshake. Or, the
authors already got the conclusion about this?

If someone has any other concerns, please point out!

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp


2011/9/8 Costin Raiciu <C.RAICIU@cs.ucl.ac.uk>:
> Hi all,
>
> As you know, we are in the stage of finalizing the MPTCP protocol draft. =
There are two major changes still to be made. Having re-read recently all t=
he emails that have gone back and forth, here are the resolutions I propose=
.
>
> The first question relates to the ability of opening new subflows in fall=
back mode (break before make in fallback is a specialization of this case).
>
> My inclination would be to mandate the following:
>
> - when MPTCP goes into fallback mode, it effectively becomes equivalent t=
o a regular TCP session; no further subflows can be setup; teardown happens=
 only with regular FIN messages, etc.
>
> This does prevent mobility in certain cases that would require a lot of c=
omplexity to handle; it seems reasonable given my hope that payload changin=
g middleboxes are not that ubiquitous.
>
> The second question regards adding support for transparent proxies. In pa=
rticular, each endpoint should know, for each address of their peer, if it =
is talking to a proxy or the destination. This would allow a smooth transit=
ion from
> an initial proxy based deployment in cellular networks to worldwide suppo=
rt for MPTCP in servers. In particular, MPTCP could connect to the proxy
>
> The proposal is to add the P flag to ADD_ADDRESS, to allow middleboxes to=
 say they are on path.
>
> Currently ADD_ADDRESS does not have flag space;
> - we could add another octet for flag space (but this is not ideal since =
this option is already big enough for IPv6, as Georg pointed out)
> - or we could reduce the number of bits in the address ID by one. 256 add=
resses seem like a lot of interfaces to have at the same time; 128 should s=
uffice. We should explicitly allow the number to wrap, though, to support l=
ong term mobility where a mobile's Wifi interface comes up and down hundred=
s of times. If we do this, we also free one bit on the MP_JOIN.
>
> My inclination is to go for the second option.
>
> When I first thought of it, it seemed to me that adding the P flag only t=
o add_address is not sufficient; the host does not know upon directly openi=
ng a connection/subflow whether it is talking directly to the destination o=
r a proxy.
> It turns out that using add address the proxy can make itself known for t=
he initial path. This path will have address ID 0; the proxy will just send=
 an add address with ID 0 and it's real address. The receiver now compares =
this address with what he thinks the remote endpoint is, and can infer it i=
s talking to a proxy on that path. The same applies if a new subflow is ope=
ned.
>
> What do people think? Are these reasonable solutions? If people agree, I =
will go ahead and rev the draft accordingly.
>
> Cheers,
> Costin
>>
>> On 6 Sep 2011, at 18:31, Alan Ford wrote:
>>
>>> Folks,
>>>
>>> Does anybody agree/disagree with my comments below? Can anybody contrib=
ute text for (2) and (5)? These are the main ones that I don't know enough =
about what was discussed and was intended in order to write anything myself=
.
>>>
>>> Thanks,
>>> Alan
>>>
>>> On 22 August 2011 22:08, Alan Ford <alan.ford@gmail.com> wrote:
>>> Apologies for the delay in replying.
>>>
>>> (1) - This won't work, unfortunately - we don't have enough option spac=
e.
>>>
>>> (2) - I'm afraid I'm out of touch on what the conclusion here was. When=
 I last thought about this, the only way I could see it working was if a su=
bflow was clearly terminated before a new one was started, relying on the t=
imeout you talk about in (3) to keep the MPTCP state alive. Was there an al=
ternative way?
>>>
>>> (3) - Indeed this is heuristics, although more so than for regular TCP =
applications may make assumptions about session survivability. This is one =
of those issues that we need to say needs thought in the protocol draft, an=
d then in the API draft talk about its actual impact.
>>>
>>> (4) - ok
>>>
>>> (5) - Once again I'm out of touch of what was discussed at Quebec so am=
 not sure how this would be defined. I remain uneasy about this flag - embe=
dding proxies as a fundamental part of the protocol - but am willing to bow=
 to consensus at Quebec.
>>>
>>> Cheers,
>>> Alan
>>>
>>>
>>> On 19 August 2011 08:15, <philip.eardley@bt.com> wrote:
>>> Hi,
>>>
>>> Just a reminder, please can we try and close these issues
>>>
>>> (1) =A0 =A0no-one complained at Mark=92s suggestion
>>>
>>> (2) =A0 =A0Needs some slightly revised text
>>>
>>> (3) =A0 =A0May need some tweaked text?
>>>
>>> (4) =A0 =A0Needs no action
>>>
>>> (5) =A0 =A0Is the main point requiring a detailed proposal
>>>
>>> Thanks
>>>
>>> phil
>>>
>>>
>>>
>>> From: Eardley,PL,Philip,DES8 R
>>> Sent: 28 July 2011 00:01
>>> To: Eardley,PL,Philip,DES8 R; multipathtcp@ietf.org
>>> Subject: Open /closed issues from WG meeting
>>>
>>>
>>>
>>> From the WG meeting I think we have a proposal on various Issues that s=
eemed open before the meeting.
>>>
>>> This email is to check consensus, especially amongst the WG people who =
aren't in Quebec or didn't listen in.
>>>
>>>
>>>
>>> Of the issues, (1) was raised in the meeting, (2) & (3) are from Mark's=
 slide 12, (4) and (5) are from Georg's slide 34.
>>>
>>>
>>>
>>> (1) Ack of MP_CAPABLE.
>>>
>>> See Mark's email to the list a short while ago, which explains & propos=
es a resolution.
>>>
>>>
>>>
>>> (2) Mobility in fallback
>>>
>>> The doc needs to correctly describe the case - when it works OK & when =
it doesn't (ie what the limitation is). We accept the limitation (it isn't =
too clear how to do a solution but seems it would be complex)
>>>
>>>
>>>
>>> (3) Teardown of state when all subflows fail
>>>
>>> This is a heuristics issue rather than a protocol issue.
>>>
>>> (Hopefully in the future we can write a doc with lessons learnt from ex=
periments, with suggestions how to deal with all sorts of heuristics issues=
)
>>>
>>>
>>>
>>> (4) =A0Add Bulk_transfer_optimisation flag to MP-Capable?
>>>
>>> Don't add, seems like extra complexity for not much gain
>>>
>>>
>>>
>>> (5) Add support for transparent proxy?
>>>
>>> Rough consensus in the room in Quebec to add an "I'm a proxy" flag to A=
DD_ADDRESS.
>>>
>>> Need the precise text to add to the draft.
>>>
>>>
>>>
>>> We'd like to resolve all these issues within the next week or so - hopi=
ng that proves possible!
>>>
>>>
>>>
>>> We will WG last call as soon as there's an updated draft that deals wit=
h these issues.
>>>
>>>
>>>
>>> If I got something wrong, please shout.
>>>
>>>
>>>
>>> thanks to everyone for the efforts to try and resolve the hopefully las=
t remaining issues with the protocol spec.
>>>
>>> Best wishes,
>>>
>>> phil
>>>
>>> From: multipathtcp-bounces@ietf.org [multipathtcp-bounces@ietf.org] On =
Behalf Of philip.eardley@bt.com [philip.eardley@bt.com]
>>> Sent: 27 July 2011 23:34
>>> To: multipathtcp@ietf.org
>>> Subject: [multipathtcp] Draft minutes of WG meeting
>>>
>>> thanks to Andrew for taking notes.
>>>
>>>
>>>
>>> Here are the draft minutes of the meeting, please send corrections /imp=
rovements
>>>
>>> thanks
>>>
>>> phil/
>>>
>>>
>>>
>>> Minutes of MPTCP WG Meeting
>>>
>>> WEDNESDAY, July 27, 2011
>>> 0900-1130 =A0Morning Session I
>>>
>>> Main aim:
>>> Conclude on all protocol issues. So that we can WG last call protocol (=
& API) docs immediately after IETF.
>>>
>>>
>>>
>>> 1. Agenda bashing, note well, Milestones =96 Chairs
>>>
>>>
>>>
>>> 2. API update =96 Michael Scharf
>>>
>>> Doc ready for last call, no comments.
>>>
>>>
>>>
>>> 3. Congestion control update =96 Mark Handley
>>> Done. Minor comments from IESG and David Black have been resolved.
>>>
>>>
>>>
>>> 4. Protocol update & discussion of open issues =96 Mark Handley
>>>
>>> Reviewers could review section 2 (overview)
>>>
>>> Small changes to protocol
>>>
>>>
>>>
>>> Discussion over sending key-A on the third Ack; there is an issue over =
the reliable delivery of the key.
>>>
>>> Jana: B could echo back the options.
>>>
>>> Mark: The issue is how to trigger stopping option sends. =A0We will dou=
ble-check if the issue is covered and resolve if required.
>>>
>>>
>>>
>>> Discussion on mobility in fallback mode
>>>
>>> It doesn't seem that it is possible to reliably use mobility in fallbac=
k mode.
>>>
>>> Andrew: Isn't that OK? =A0Can't you just say, if the ACKs haven't arriv=
ed, oh well, RST it.
>>>
>>> Georg: The draft says that there's no more multipath signalling in fall=
back. =A0So you can't get here.
>>>
>>> Mark: So, the only resolution seems to be that if you end up with un-AC=
Ked data, you lose.
>>>
>>> Costin: You have signalling in the SYNs
>>>
>>> Mark: Which gives you a way out, if that is possible.
>>>
>>> Georg: I have a proposal for how to cover this later.
>>>
>>> Mark: It probably isn't worth trying to optimise this case.
>>>
>>> What to do if all subflows are in time out?
>>>
>>> Phil: You're saying that is an implementation specific heuristic?
>>>
>>> Andrew: What say no interfaces are up?
>>>
>>> Mark: Yes, doesn't need specified.
>>>
>>>
>>> Phil: Plan is, get these three issues resolved on the list, give it a w=
eek or so, and WGLC it.
>>>
>>>
>>>
>>>
>>> Experiments on "Is MPTCP deployable?"
>>>
>>> Georg: You can't name service providers or enterprises, to what extent =
are the harmful boxes installed in providers vs enterprises?
>>>
>>> Mark: Cellular providers have more, but no monopoly. =A0We can't claim =
this to be completely representative.
>>>
>>>
>>> 5. (new draft) Enhancements to Improve the Applicability of Multipath T=
CP to Wireless Access Networks =96 Georg Hampel
>>>
>>> Mark: Best path probably has lowest energy per byte
>>>
>>> Georg: If you have money or energy to burn, you might not care
>>>
>>> Mark: Proxy. =A0One part of me is revolted by supporting transparent pr=
oxies explicitly, and we should support them. =A0Calling the flag I_AM_A_PR=
OXY is more explicit.
>>>
>>> Andrew: I agree there are always proxy devices "there"
>>> Andrew: ... there are good features here (is it important to tag this a=
s "min" complexity, images are getting big for this part of the stack)
>>> Georg: How about recv buffer reduction?
>>> Andrew: good idea
>>>
>>> Mark: I'm not totally convinced that buffer reduction is real.
>>>
>>> Costin: Receive buffer reduction is orthogonal to implementation comple=
xity
>>>
>>> Mark: The other question, if you're in path selection mode, do you get =
the right behaviour over mobile networks. =A0Assert: you should be able to =
tell from signal strength etc. if it is a good link... not really. =A0I can=
't easily predict from wireless if it's decent, I have to actually try it.
>>>
>>> Georg: The signal is not necessarily indicative, could have other varia=
bles
>>>
>>> Andrew: Wifi links have to train up, which can take a few hundred packe=
ts
>>>
>>> David: I have in mind a satellite PEP setup, where RTT could be radical=
ly different.
>>>
>>> Mark: Our intuition is that you should only use one path most of the ti=
me, but you want to transition make-before-break and do a soft transition. =
=A0You need to get the links adequately probed.
>>>
>>> Georg: Both of our points are based on speculation, we need to know how=
 well you do if you shorten the overlap
>>>
>>> Costin: that paper assumed there is an oracle that knows the best path =
that chooses the best path yes: 15% compared to an optimal path selection a=
lgorithm that can't be achieved in practice
>>>
>>> Costin: you can also have redundant data if you want to probe and avoid=
 hol blocking
>>>
>>> Mark: The spec does not prohibit you sending data on multiple paths
>>>
>>> Phil: We've had quite a lot of discussion now on heuristic stuff, how y=
ou might make this work in practice. =A0It would be nice to have more exper=
iments. =A0My impression is that BTO doesn't seem to gain very much
>>>
>>> Andrew: Suggest not doing bulk transfer optimisation, it could be reall=
y expensive for little gain.
>>>
>>> Andrew: Very much in favour of proxy flags, they're everywhere
>>>
>>> Mark: I am in favour of adding a bit to support transparent proxies, 1)=
 address of proxy 2) remote supports mptcp - could decide based on this
>>>
>>> Andrew: It can be a content inspecting firewall that just wants to do i=
ts job
>>>
>>> Georg: Proxy is not optimisation, it's a feature
>>>
>>> Michio: No such middle box
>>>
>>> Mark: If we can preserve the benefits in the presence of boxes that wan=
t to be on path, then we should. =A0If we don't want the content inspected,=
 we need to encrypt.
>>>
>>> Georg: 3G providers still get to bill it.
>>>
>>> Gorry: There are proxies that need to inform L2 what is going on.
>>>
>>> Andrew: Could have network architectures that wouldn't otherwise work t=
hat way.
>>>
>>> Mark: We should probably try to address this before we move the protoco=
l forward.
>>>
>>> Phil: Would like to see people get together in the remainder of the wee=
k.
>>>
>>> Andrew: I think the bottom three lines of the slide are OK, up to namin=
g.
>>>
>>> Costin: +1
>>>
>>>
>>>
>>> 6. Call for news on Implementations
>>>
>>> Linux version being updated as http://mptcpinfo.ucl.ac.be, also for Nok=
ia N900 & Android Nexus-one
>>>
>>> Tim: Implementation is not SMP capable, does anyone know if it is fixed=
?
>>>
>>> <apparently not, docs still say no SMP>
>>>
>>> Georg Hampel - in progress on 'lightweight' implementation
>>>
>>> Sebastien Roy: Oracle are working on a prototype in Solaris. =A0Don't k=
now if it can be released yet.
>>>
>>>
>>>
>>> 7. Discussion on future of WG - Chairs
>>>
>>>
>>>
>>> Discussion last time (roughly) concluded that we pause for experiments =
& usage: WG hears about implementations, heuristics etc; hears about invest=
igations on possible extensions; not do any protocol standards work. Re-cha=
rter in ~ 1year when understand better what the needs are & where the energ=
y is
>>> No comments - Chairs will work with AD on re-chartering text, when the =
current charter is done.
>>>
>>>
>>
>
>

From georg.hampel@alcatel-lucent.com  Fri Sep 16 07:49:54 2011
Return-Path: <georg.hampel@alcatel-lucent.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B77F221F8B5A for <multipathtcp@ietfa.amsl.com>; Fri, 16 Sep 2011 07:49:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.176
X-Spam-Level: 
X-Spam-Status: No, score=-6.176 tagged_above=-999 required=5 tests=[AWL=-0.177, BAYES_00=-2.599, J_CHICKENPOX_72=0.6, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zj0vM7ITWI09 for <multipathtcp@ietfa.amsl.com>; Fri, 16 Sep 2011 07:49:53 -0700 (PDT)
Received: from ihemail1.lucent.com (ihemail1.lucent.com [135.245.0.33]) by ietfa.amsl.com (Postfix) with ESMTP id 4774321F8B36 for <multipathtcp@ietf.org>; Fri, 16 Sep 2011 07:49:52 -0700 (PDT)
Received: from usnavsmail1.ndc.alcatel-lucent.com (usnavsmail1.ndc.alcatel-lucent.com [135.3.39.9]) by ihemail1.lucent.com (8.13.8/IER-o) with ESMTP id p8GEq5j7003282 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Fri, 16 Sep 2011 09:52:06 -0500 (CDT)
Received: from USNAVSXCHHUB02.ndc.alcatel-lucent.com (usnavsxchhub02.ndc.alcatel-lucent.com [135.3.39.111]) by usnavsmail1.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id p8GEq43V007775 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT); Fri, 16 Sep 2011 09:52:05 -0500
Received: from USNAVSXCHMBSA2.ndc.alcatel-lucent.com ([135.3.39.124]) by USNAVSXCHHUB02.ndc.alcatel-lucent.com ([135.3.39.111]) with mapi; Fri, 16 Sep 2011 09:52:05 -0500
From: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
To: Yoshifumi Nishida <yoshifumi.nishida@gmail.com>, multipathtcp <multipathtcp@ietf.org>
Date: Fri, 16 Sep 2011 09:52:03 -0500
Thread-Topic: [multipathtcp] Proposed changes to the protocol draft
Thread-Index: Acx0RTElNpmJY51RQyOWQ4cj/dJMcwAMgCgg
Message-ID: <154773479ED2314980CB638A48FC443486F04711@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLG0yYkDGycjqfWGFNs46eHkrVZCFj2X1dy9ahBh89br_Q@mail.gmail.com>
In-Reply-To: <CABxaRLG0yYkDGycjqfWGFNs46eHkrVZCFj2X1dy9ahBh89br_Q@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.33
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.9
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 16 Sep 2011 14:49:54 -0000

Yoshifumi, all,

I have a problem with the quick fix of "I'm a proxy"-bit squeezed into ADD_=
ADDRESS. I propose to keep the ADD_ADDRESS option as in draft-ietf-mptcp-mu=
ltiaddressed-04.txt and to discuss further extensions over time.

Here are the reasons:

Currently, the ADD_ADDRESS option covers two meanings, namely:
(1) Cache this address together with address id.
(2) Do MP_JOIN from this address right away.

This works as long as the request to do MP_JOIN is the only purpose for ADD=
_ADDRESS. That's fine for now, but probably not in the future.

With the introduction of the transparent proxy, ADD_ADDRESS obtains at leas=
t one additional meaning:
(3) I'm a proxy address.

We still have to discuss what the implications of this statement "I'm a pro=
xy" are. One implication could be: ONLY use this proxy address for future s=
ubflows. Are we certain that this is the only meaning? I'm not. (Yoshifumi =
doesn't seem to be either).

Further, a MPTCP host may want to provide alternative addresses as backup, =
e.g. of different IP version, WITHOUT requiring a new subflow right away. T=
his is currently not covered.

I think what we ultimately need is a separation between two functions:
(a) Provide a new address together with address id
(b) Assign a specific interpretation or request-for-action to an address id

Function (a) can be covered by ADD_ADDRESS. Function (b) deserves its own E=
XTENDIBLE option, which allows adding more interpretations/requests over ti=
me.

Thanks,

Georg



-----Original Message-----
From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org] =
On Behalf Of Yoshifumi Nishida
Sent: Friday, September 16, 2011 3:48 AM
To: multipathtcp
Subject: Re: [multipathtcp] Proposed changes to the protocol draft

Hello,

It seems that we're reaching the stage to finalize the spec.
Please correct me if I'm wrong, but here's the current status I think..

Constin's proposal 1:
    "when MPTCP goes into fallback mode, it effectively becomes
equivalent to a regular TCP session; no further subflows can be setup;
teardown happens only with regular FIN messages, etc."

 seems to be reaching consensus as we don't see any objections so far.

Constin's proposal 2:
    "add the P flag to ADD_ADDRESS, to allow middleboxes to say they
are on path"

 seems to need more inputs to discuss where to put the flag, although
this looks relatively a minor issue.

In addition, I guess we need to discuss initial handshake. Or, the
authors already got the conclusion about this?

If someone has any other concerns, please point out!

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp


2011/9/8 Costin Raiciu <C.RAICIU@cs.ucl.ac.uk>:
> Hi all,
>
> As you know, we are in the stage of finalizing the MPTCP protocol draft. =
There are two major changes still to be made. Having re-read recently all t=
he emails that have gone back and forth, here are the resolutions I propose=
.
>
> The first question relates to the ability of opening new subflows in fall=
back mode (break before make in fallback is a specialization of this case).
>
> My inclination would be to mandate the following:
>
> - when MPTCP goes into fallback mode, it effectively becomes equivalent t=
o a regular TCP session; no further subflows can be setup; teardown happens=
 only with regular FIN messages, etc.
>
> This does prevent mobility in certain cases that would require a lot of c=
omplexity to handle; it seems reasonable given my hope that payload changin=
g middleboxes are not that ubiquitous.
>
> The second question regards adding support for transparent proxies. In pa=
rticular, each endpoint should know, for each address of their peer, if it =
is talking to a proxy or the destination. This would allow a smooth transit=
ion from
> an initial proxy based deployment in cellular networks to worldwide suppo=
rt for MPTCP in servers. In particular, MPTCP could connect to the proxy
>
> The proposal is to add the P flag to ADD_ADDRESS, to allow middleboxes to=
 say they are on path.
>
> Currently ADD_ADDRESS does not have flag space;
> - we could add another octet for flag space (but this is not ideal since =
this option is already big enough for IPv6, as Georg pointed out)
> - or we could reduce the number of bits in the address ID by one. 256 add=
resses seem like a lot of interfaces to have at the same time; 128 should s=
uffice. We should explicitly allow the number to wrap, though, to support l=
ong term mobility where a mobile's Wifi interface comes up and down hundred=
s of times. If we do this, we also free one bit on the MP_JOIN.
>
> My inclination is to go for the second option.
>
> When I first thought of it, it seemed to me that adding the P flag only t=
o add_address is not sufficient; the host does not know upon directly openi=
ng a connection/subflow whether it is talking directly to the destination o=
r a proxy.
> It turns out that using add address the proxy can make itself known for t=
he initial path. This path will have address ID 0; the proxy will just send=
 an add address with ID 0 and it's real address. The receiver now compares =
this address with what he thinks the remote endpoint is, and can infer it i=
s talking to a proxy on that path. The same applies if a new subflow is ope=
ned.
>
> What do people think? Are these reasonable solutions? If people agree, I =
will go ahead and rev the draft accordingly.
>
> Cheers,
> Costin
>>
>> On 6 Sep 2011, at 18:31, Alan Ford wrote:
>>
>>> Folks,
>>>
>>> Does anybody agree/disagree with my comments below? Can anybody contrib=
ute text for (2) and (5)? These are the main ones that I don't know enough =
about what was discussed and was intended in order to write anything myself=
.
>>>
>>> Thanks,
>>> Alan
>>>
>>> On 22 August 2011 22:08, Alan Ford <alan.ford@gmail.com> wrote:
>>> Apologies for the delay in replying.
>>>
>>> (1) - This won't work, unfortunately - we don't have enough option spac=
e.
>>>
>>> (2) - I'm afraid I'm out of touch on what the conclusion here was. When=
 I last thought about this, the only way I could see it working was if a su=
bflow was clearly terminated before a new one was started, relying on the t=
imeout you talk about in (3) to keep the MPTCP state alive. Was there an al=
ternative way?
>>>
>>> (3) - Indeed this is heuristics, although more so than for regular TCP =
applications may make assumptions about session survivability. This is one =
of those issues that we need to say needs thought in the protocol draft, an=
d then in the API draft talk about its actual impact.
>>>
>>> (4) - ok
>>>
>>> (5) - Once again I'm out of touch of what was discussed at Quebec so am=
 not sure how this would be defined. I remain uneasy about this flag - embe=
dding proxies as a fundamental part of the protocol - but am willing to bow=
 to consensus at Quebec.
>>>
>>> Cheers,
>>> Alan
>>>
>>>
>>> On 19 August 2011 08:15, <philip.eardley@bt.com> wrote:
>>> Hi,
>>>
>>> Just a reminder, please can we try and close these issues
>>>
>>> (1)    no-one complained at Mark's suggestion
>>>
>>> (2)    Needs some slightly revised text
>>>
>>> (3)    May need some tweaked text?
>>>
>>> (4)    Needs no action
>>>
>>> (5)    Is the main point requiring a detailed proposal
>>>
>>> Thanks
>>>
>>> phil
>>>
>>>
>>>
>>> From: Eardley,PL,Philip,DES8 R
>>> Sent: 28 July 2011 00:01
>>> To: Eardley,PL,Philip,DES8 R; multipathtcp@ietf.org
>>> Subject: Open /closed issues from WG meeting
>>>
>>>
>>>
>>> From the WG meeting I think we have a proposal on various Issues that s=
eemed open before the meeting.
>>>
>>> This email is to check consensus, especially amongst the WG people who =
aren't in Quebec or didn't listen in.
>>>
>>>
>>>
>>> Of the issues, (1) was raised in the meeting, (2) & (3) are from Mark's=
 slide 12, (4) and (5) are from Georg's slide 34.
>>>
>>>
>>>
>>> (1) Ack of MP_CAPABLE.
>>>
>>> See Mark's email to the list a short while ago, which explains & propos=
es a resolution.
>>>
>>>
>>>
>>> (2) Mobility in fallback
>>>
>>> The doc needs to correctly describe the case - when it works OK & when =
it doesn't (ie what the limitation is). We accept the limitation (it isn't =
too clear how to do a solution but seems it would be complex)
>>>
>>>
>>>
>>> (3) Teardown of state when all subflows fail
>>>
>>> This is a heuristics issue rather than a protocol issue.
>>>
>>> (Hopefully in the future we can write a doc with lessons learnt from ex=
periments, with suggestions how to deal with all sorts of heuristics issues=
)
>>>
>>>
>>>
>>> (4)  Add Bulk_transfer_optimisation flag to MP-Capable?
>>>
>>> Don't add, seems like extra complexity for not much gain
>>>
>>>
>>>
>>> (5) Add support for transparent proxy?
>>>
>>> Rough consensus in the room in Quebec to add an "I'm a proxy" flag to A=
DD_ADDRESS.
>>>
>>> Need the precise text to add to the draft.
>>>
>>>
>>>
>>> We'd like to resolve all these issues within the next week or so - hopi=
ng that proves possible!
>>>
>>>
>>>
>>> We will WG last call as soon as there's an updated draft that deals wit=
h these issues.
>>>
>>>
>>>
>>> If I got something wrong, please shout.
>>>
>>>
>>>
>>> thanks to everyone for the efforts to try and resolve the hopefully las=
t remaining issues with the protocol spec.
>>>
>>> Best wishes,
>>>
>>> phil
>>>
>>> From: multipathtcp-bounces@ietf.org [multipathtcp-bounces@ietf.org] On =
Behalf Of philip.eardley@bt.com [philip.eardley@bt.com]
>>> Sent: 27 July 2011 23:34
>>> To: multipathtcp@ietf.org
>>> Subject: [multipathtcp] Draft minutes of WG meeting
>>>
>>> thanks to Andrew for taking notes.
>>>
>>>
>>>
>>> Here are the draft minutes of the meeting, please send corrections /imp=
rovements
>>>
>>> thanks
>>>
>>> phil/
>>>
>>>
>>>
>>> Minutes of MPTCP WG Meeting
>>>
>>> WEDNESDAY, July 27, 2011
>>> 0900-1130  Morning Session I
>>>
>>> Main aim:
>>> Conclude on all protocol issues. So that we can WG last call protocol (=
& API) docs immediately after IETF.
>>>
>>>
>>>
>>> 1. Agenda bashing, note well, Milestones - Chairs
>>>
>>>
>>>
>>> 2. API update - Michael Scharf
>>>
>>> Doc ready for last call, no comments.
>>>
>>>
>>>
>>> 3. Congestion control update - Mark Handley
>>> Done. Minor comments from IESG and David Black have been resolved.
>>>
>>>
>>>
>>> 4. Protocol update & discussion of open issues - Mark Handley
>>>
>>> Reviewers could review section 2 (overview)
>>>
>>> Small changes to protocol
>>>
>>>
>>>
>>> Discussion over sending key-A on the third Ack; there is an issue over =
the reliable delivery of the key.
>>>
>>> Jana: B could echo back the options.
>>>
>>> Mark: The issue is how to trigger stopping option sends.  We will doubl=
e-check if the issue is covered and resolve if required.
>>>
>>>
>>>
>>> Discussion on mobility in fallback mode
>>>
>>> It doesn't seem that it is possible to reliably use mobility in fallbac=
k mode.
>>>
>>> Andrew: Isn't that OK?  Can't you just say, if the ACKs haven't arrived=
, oh well, RST it.
>>>
>>> Georg: The draft says that there's no more multipath signalling in fall=
back.  So you can't get here.
>>>
>>> Mark: So, the only resolution seems to be that if you end up with un-AC=
Ked data, you lose.
>>>
>>> Costin: You have signalling in the SYNs
>>>
>>> Mark: Which gives you a way out, if that is possible.
>>>
>>> Georg: I have a proposal for how to cover this later.
>>>
>>> Mark: It probably isn't worth trying to optimise this case.
>>>
>>> What to do if all subflows are in time out?
>>>
>>> Phil: You're saying that is an implementation specific heuristic?
>>>
>>> Andrew: What say no interfaces are up?
>>>
>>> Mark: Yes, doesn't need specified.
>>>
>>>
>>> Phil: Plan is, get these three issues resolved on the list, give it a w=
eek or so, and WGLC it.
>>>
>>>
>>>
>>>
>>> Experiments on "Is MPTCP deployable?"
>>>
>>> Georg: You can't name service providers or enterprises, to what extent =
are the harmful boxes installed in providers vs enterprises?
>>>
>>> Mark: Cellular providers have more, but no monopoly.  We can't claim th=
is to be completely representative.
>>>
>>>
>>> 5. (new draft) Enhancements to Improve the Applicability of Multipath T=
CP to Wireless Access Networks - Georg Hampel
>>>
>>> Mark: Best path probably has lowest energy per byte
>>>
>>> Georg: If you have money or energy to burn, you might not care
>>>
>>> Mark: Proxy.  One part of me is revolted by supporting transparent prox=
ies explicitly, and we should support them.  Calling the flag I_AM_A_PROXY =
is more explicit.
>>>
>>> Andrew: I agree there are always proxy devices "there"
>>> Andrew: ... there are good features here (is it important to tag this a=
s "min" complexity, images are getting big for this part of the stack)
>>> Georg: How about recv buffer reduction?
>>> Andrew: good idea
>>>
>>> Mark: I'm not totally convinced that buffer reduction is real.
>>>
>>> Costin: Receive buffer reduction is orthogonal to implementation comple=
xity
>>>
>>> Mark: The other question, if you're in path selection mode, do you get =
the right behaviour over mobile networks.  Assert: you should be able to te=
ll from signal strength etc. if it is a good link... not really.  I can't e=
asily predict from wireless if it's decent, I have to actually try it.
>>>
>>> Georg: The signal is not necessarily indicative, could have other varia=
bles
>>>
>>> Andrew: Wifi links have to train up, which can take a few hundred packe=
ts
>>>
>>> David: I have in mind a satellite PEP setup, where RTT could be radical=
ly different.
>>>
>>> Mark: Our intuition is that you should only use one path most of the ti=
me, but you want to transition make-before-break and do a soft transition. =
 You need to get the links adequately probed.
>>>
>>> Georg: Both of our points are based on speculation, we need to know how=
 well you do if you shorten the overlap
>>>
>>> Costin: that paper assumed there is an oracle that knows the best path =
that chooses the best path yes: 15% compared to an optimal path selection a=
lgorithm that can't be achieved in practice
>>>
>>> Costin: you can also have redundant data if you want to probe and avoid=
 hol blocking
>>>
>>> Mark: The spec does not prohibit you sending data on multiple paths
>>>
>>> Phil: We've had quite a lot of discussion now on heuristic stuff, how y=
ou might make this work in practice.  It would be nice to have more experim=
ents.  My impression is that BTO doesn't seem to gain very much
>>>
>>> Andrew: Suggest not doing bulk transfer optimisation, it could be reall=
y expensive for little gain.
>>>
>>> Andrew: Very much in favour of proxy flags, they're everywhere
>>>
>>> Mark: I am in favour of adding a bit to support transparent proxies, 1)=
 address of proxy 2) remote supports mptcp - could decide based on this
>>>
>>> Andrew: It can be a content inspecting firewall that just wants to do i=
ts job
>>>
>>> Georg: Proxy is not optimisation, it's a feature
>>>
>>> Michio: No such middle box
>>>
>>> Mark: If we can preserve the benefits in the presence of boxes that wan=
t to be on path, then we should.  If we don't want the content inspected, w=
e need to encrypt.
>>>
>>> Georg: 3G providers still get to bill it.
>>>
>>> Gorry: There are proxies that need to inform L2 what is going on.
>>>
>>> Andrew: Could have network architectures that wouldn't otherwise work t=
hat way.
>>>
>>> Mark: We should probably try to address this before we move the protoco=
l forward.
>>>
>>> Phil: Would like to see people get together in the remainder of the wee=
k.
>>>
>>> Andrew: I think the bottom three lines of the slide are OK, up to namin=
g.
>>>
>>> Costin: +1
>>>
>>>
>>>
>>> 6. Call for news on Implementations
>>>
>>> Linux version being updated as http://mptcpinfo.ucl.ac.be, also for Nok=
ia N900 & Android Nexus-one
>>>
>>> Tim: Implementation is not SMP capable, does anyone know if it is fixed=
?
>>>
>>> <apparently not, docs still say no SMP>
>>>
>>> Georg Hampel - in progress on 'lightweight' implementation
>>>
>>> Sebastien Roy: Oracle are working on a prototype in Solaris.  Don't kno=
w if it can be released yet.
>>>
>>>
>>>
>>> 7. Discussion on future of WG - Chairs
>>>
>>>
>>>
>>> Discussion last time (roughly) concluded that we pause for experiments =
& usage: WG hears about implementations, heuristics etc; hears about invest=
igations on possible extensions; not do any protocol standards work. Re-cha=
rter in ~ 1year when understand better what the needs are & where the energ=
y is
>>> No comments - Chairs will work with AD on re-chartering text, when the =
current charter is done.
>>>
>>>
>>
>
>
_______________________________________________
multipathtcp mailing list
multipathtcp@ietf.org
https://www.ietf.org/mailman/listinfo/multipathtcp

From philip.eardley@bt.com  Fri Sep 23 08:05:55 2011
Return-Path: <philip.eardley@bt.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9139821F8CAE for <multipathtcp@ietfa.amsl.com>; Fri, 23 Sep 2011 08:05:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.83
X-Spam-Level: 
X-Spam-Status: No, score=-102.83 tagged_above=-999 required=5 tests=[AWL=0.169, BAYES_00=-2.599, J_CHICKENPOX_21=0.6, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id FyzzkGUAILuJ for <multipathtcp@ietfa.amsl.com>; Fri, 23 Sep 2011 08:05:55 -0700 (PDT)
Received: from smtpe1.intersmtp.com (smtp61.intersmtp.com [62.239.224.234]) by ietfa.amsl.com (Postfix) with ESMTP id CF54821F8CB1 for <multipathtcp@ietf.org>; Fri, 23 Sep 2011 08:05:54 -0700 (PDT)
Received: from EVMHT65-UKRD.domain1.systemhost.net (10.36.3.102) by RDW083A005ED61.smtp-e1.hygiene.service (10.187.98.10) with Microsoft SMTP Server (TLS) id 8.3.159.2; Fri, 23 Sep 2011 16:08:28 +0100
Received: from EMV65-UKRD.domain1.systemhost.net ([169.254.1.20]) by EVMHT65-UKRD.domain1.systemhost.net ([10.36.3.102]) with mapi; Fri, 23 Sep 2011 16:08:28 +0100
From: <philip.eardley@bt.com>
To: <multipathtcp@ietf.org>
Date: Fri, 23 Sep 2011 16:08:27 +0100
Thread-Topic: [multipathtcp] Proposed changes to the protocol draft
Thread-Index: AcxxFRPp7K1k2f9/Rs6jvlA+aheI5AI7Lk4g
Message-ID: <9510D26531EF184D9017DF24659BB87F330B6BBA95@EMV65-UKRD.domain1.systemhost.net>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com> <4E6B312E.8060707@uclouvain.be> <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com>
In-Reply-To: <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com>
Accept-Language: en-US, en-GB
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US, en-GB
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 23 Sep 2011 15:05:55 -0000

There seems to be consensus around completing the main spec without trying =
to solve the proxy issue (since it needs significant more work), but making=
 the spec flexible enough that a 'proxy capability' can be added later.=20
Question is how to leave room for this capability. One suggestion was to ke=
ep aside one bit of the address-id - does that have consensus? Would any of=
 the spec have to be re-thought?

Thanks
phil

Philip Eardley=A0=A0=A0
Research and Technology Strategy

This email contains BT information, which may be privileged or confidential=
. It's meant only for the individual(s) or entity named above. If you're no=
t the intended recipient, note that disclosing, copying, distributing or us=
ing this information is prohibited. If you've received this email in error,=
 please let me know immediately on the email address above. Thank you. We m=
onitor our email system, and may record your emails.
British Telecommunications plc
Registered office: 81 Newgate Street London EC1A 7AJ
Registered in England no: 1800000


-----Original Message-----
From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org] =
On Behalf Of Lars Eggert
Sent: 12 September 2011 07:28
To: Olivier.Bonaventure@uclouvain.be
Cc: multipathtcp@ietf.org List
Subject: Re: [multipathtcp] Proposed changes to the protocol draft

On 2011-9-10, at 12:43, Olivier Bonaventure wrote:
> I fear that this would add unnecessary complexity and would prefer to=20
> wait for more implementation experience before looking at such=20
> optimisations in the main spec

This is my feeling at the moment, too. I think it'd be useful to have this =
written up in an individual ID so that folks can discuss (and maybe try and=
 implement) it, but I think it's too early to put this into the main spec.

Lars

From georg.hampel@alcatel-lucent.com  Mon Sep 26 07:58:02 2011
Return-Path: <georg.hampel@alcatel-lucent.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 762B821F8D15 for <multipathtcp@ietfa.amsl.com>; Mon, 26 Sep 2011 07:58:02 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.168
X-Spam-Level: 
X-Spam-Status: No, score=-6.168 tagged_above=-999 required=5 tests=[AWL=-0.169, BAYES_00=-2.599, J_CHICKENPOX_21=0.6, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OQ2nNG93NpYL for <multipathtcp@ietfa.amsl.com>; Mon, 26 Sep 2011 07:58:01 -0700 (PDT)
Received: from ihemail4.lucent.com (ihemail4.lucent.com [135.245.0.39]) by ietfa.amsl.com (Postfix) with ESMTP id D470021F8CF7 for <multipathtcp@ietf.org>; Mon, 26 Sep 2011 07:58:01 -0700 (PDT)
Received: from usnavsmail2.ndc.alcatel-lucent.com (usnavsmail2.ndc.alcatel-lucent.com [135.3.39.10]) by ihemail4.lucent.com (8.13.8/IER-o) with ESMTP id p8QF0eKR019177 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 26 Sep 2011 10:00:40 -0500 (CDT)
Received: from USNAVSXCHHUB03.ndc.alcatel-lucent.com (usnavsxchhub03.ndc.alcatel-lucent.com [135.3.39.112]) by usnavsmail2.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id p8QF0cLn021161 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT); Mon, 26 Sep 2011 10:00:40 -0500
Received: from USNAVSXCHMBSA2.ndc.alcatel-lucent.com ([135.3.39.124]) by USNAVSXCHHUB03.ndc.alcatel-lucent.com ([135.3.39.112]) with mapi; Mon, 26 Sep 2011 10:00:39 -0500
From: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
To: "philip.eardley@bt.com" <philip.eardley@bt.com>, "multipathtcp@ietf.org" <multipathtcp@ietf.org>
Date: Mon, 26 Sep 2011 10:00:37 -0500
Thread-Topic: [multipathtcp] Proposed changes to the protocol draft
Thread-Index: AcxxFRPp7K1k2f9/Rs6jvlA+aheI5AI7Lk4gAJXDNzA=
Message-ID: <154773479ED2314980CB638A48FC443486FE3311@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com> <4E6B312E.8060707@uclouvain.be> <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com> <9510D26531EF184D9017DF24659BB87F330B6BBA95@EMV65-UKRD.domain1.systemhost.net>
In-Reply-To: <9510D26531EF184D9017DF24659BB87F330B6BBA95@EMV65-UKRD.domain1.systemhost.net>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.39
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.10
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 26 Sep 2011 14:58:02 -0000

Hi Phil, all,

Are we certain that 1 bit is enough to cover all purposes and implications =
of ADD_ADDRESS?=20

Based on recent proxy-related drafts and discussions, ADD_ADDRESS would alr=
eady have to support (1) "Join using this address" and (2) "I'm a proxy". S=
ince both implications are independent from each other and may occur in uni=
son, 1 bit would not be enough.

A more detailed discussion on proxy functions may reveal additional implica=
tions of this option.=20

Adding one byte to the option instead of carving out 1 bit should provide s=
ufficient room for proxy and other extensions.=20

Thanks.

Regards,

Georg



-----Original Message-----
From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org] =
On Behalf Of philip.eardley@bt.com
Sent: Friday, September 23, 2011 11:08 AM
To: multipathtcp@ietf.org
Subject: Re: [multipathtcp] Proposed changes to the protocol draft

There seems to be consensus around completing the main spec without trying =
to solve the proxy issue (since it needs significant more work), but making=
 the spec flexible enough that a 'proxy capability' can be added later.=20
Question is how to leave room for this capability. One suggestion was to ke=
ep aside one bit of the address-id - does that have consensus? Would any of=
 the spec have to be re-thought?

Thanks
phil

Philip Eardley=A0=A0=A0
Research and Technology Strategy

This email contains BT information, which may be privileged or confidential=
. It's meant only for the individual(s) or entity named above. If you're no=
t the intended recipient, note that disclosing, copying, distributing or us=
ing this information is prohibited. If you've received this email in error,=
 please let me know immediately on the email address above. Thank you. We m=
onitor our email system, and may record your emails.
British Telecommunications plc
Registered office: 81 Newgate Street London EC1A 7AJ
Registered in England no: 1800000


-----Original Message-----
From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org] =
On Behalf Of Lars Eggert
Sent: 12 September 2011 07:28
To: Olivier.Bonaventure@uclouvain.be
Cc: multipathtcp@ietf.org List
Subject: Re: [multipathtcp] Proposed changes to the protocol draft

On 2011-9-10, at 12:43, Olivier Bonaventure wrote:
> I fear that this would add unnecessary complexity and would prefer to=20
> wait for more implementation experience before looking at such=20
> optimisations in the main spec

This is my feeling at the moment, too. I think it'd be useful to have this =
written up in an individual ID so that folks can discuss (and maybe try and=
 implement) it, but I think it's too early to put this into the main spec.

Lars
_______________________________________________
multipathtcp mailing list
multipathtcp@ietf.org
https://www.ietf.org/mailman/listinfo/multipathtcp

From yoshifumi.nishida@gmail.com  Tue Sep 27 02:03:01 2011
Return-Path: <yoshifumi.nishida@gmail.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 204F421F8C32 for <multipathtcp@ietfa.amsl.com>; Tue, 27 Sep 2011 02:03:01 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.474
X-Spam-Level: 
X-Spam-Status: No, score=-102.474 tagged_above=-999 required=5 tests=[AWL=-0.097, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, J_CHICKENPOX_21=0.6, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jYaJYub3aC-Q for <multipathtcp@ietfa.amsl.com>; Tue, 27 Sep 2011 02:03:00 -0700 (PDT)
Received: from mail-iy0-f172.google.com (mail-iy0-f172.google.com [209.85.210.172]) by ietfa.amsl.com (Postfix) with ESMTP id 8941B21F8C3E for <multipathtcp@ietf.org>; Tue, 27 Sep 2011 02:02:57 -0700 (PDT)
Received: by iaby26 with SMTP id y26so6836979iab.31 for <multipathtcp@ietf.org>; Tue, 27 Sep 2011 02:05:42 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=ZkQ/LO3NIfYSQoQzqjycK67BfT0RNulX3VQMhLcDNZs=; b=BSwY8mIklbc/UvuAWRIlSUmTVapZ5ht9cX0vPkcvZyceVsNSZQAUeXob3Dq+Lw4Oph EhGpTmOqvCBoBAUuOpemMCBErZKVXw/t0yZcvpVggnokfwGhwS4DakSfU+WfaRZEy62I YQoCb06f6PEP6t5zs6fNLSzxFHv+mIjdFsG28=
MIME-Version: 1.0
Received: by 10.231.50.204 with SMTP id a12mr10067682ibg.11.1317114341128; Tue, 27 Sep 2011 02:05:41 -0700 (PDT)
Sender: yoshifumi.nishida@gmail.com
Received: by 10.231.199.80 with HTTP; Tue, 27 Sep 2011 02:05:41 -0700 (PDT)
In-Reply-To: <154773479ED2314980CB638A48FC443486FE3311@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com> <4E6B312E.8060707@uclouvain.be> <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com> <9510D26531EF184D9017DF24659BB87F330B6BBA95@EMV65-UKRD.domain1.systemhost.net> <154773479ED2314980CB638A48FC443486FE3311@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
Date: Tue, 27 Sep 2011 02:05:41 -0700
X-Google-Sender-Auth: lSo6ENbpEP1f9DsVZjcVG9yLQNw
Message-ID: <CABxaRLEsQ9MQ-CSfx_uxpcRzxomicjbC+doEGy46zU+8Vv+4-A@mail.gmail.com>
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
To: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: "multipathtcp@ietf.org" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 27 Sep 2011 09:03:01 -0000

Hi Georg, All,

I think we need more inputs or opinions for this.
I'm trying to create a simple summary for the proposals from the
previous discussion.
Please correct me if I'm missing or misunderstanding something.


 1) Add a byte to ADD_ADDRESS
     pros: simple change to the current spec. good extensibility
     cons: need one more byte

 2) Use one bit of the address-id
     pros: simple change to the current spec.
     cons: less address-id space, low extensibility

 3) Use more bits of the address-id
     pros: simple change to the current spec, some extensibility
     cons: much less address-id space,

 4) Keep the current format and develop new option to describe the
address (e.g DESC_ADDR)
     pros: great extensibility. no need to change the current spec.
     cons: might need another mechanism to sync 2 options reliably.

Also, I personally think the following can be an another possibility,
although it might not be a good idea.

 5) Keep the current format and develop new option to specify proxy or
other addresses (e. g ADD_ADDR2)
    pros: great extensibility. no need to change the current spec
    cons: having 2 options for mostly the same feature might be a bit
ugly design.

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp

2011/9/26 Hampel, K Georg (K Georg) <georg.hampel@alcatel-lucent.com>:
> Hi Phil, all,
>
> Are we certain that 1 bit is enough to cover all purposes and implication=
s of ADD_ADDRESS?
>
> Based on recent proxy-related drafts and discussions, ADD_ADDRESS would a=
lready have to support (1) "Join using this address" and (2) "I'm a proxy".=
 Since both implications are independent from each other and may occur in u=
nison, 1 bit would not be enough.
>
> A more detailed discussion on proxy functions may reveal additional impli=
cations of this option.
>
> Adding one byte to the option instead of carving out 1 bit should provide=
 sufficient room for proxy and other extensions.
>
> Thanks.
>
> Regards,
>
> Georg
>
>
>
> -----Original Message-----
> From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org=
] On Behalf Of philip.eardley@bt.com
> Sent: Friday, September 23, 2011 11:08 AM
> To: multipathtcp@ietf.org
> Subject: Re: [multipathtcp] Proposed changes to the protocol draft
>
> There seems to be consensus around completing the main spec without tryin=
g to solve the proxy issue (since it needs significant more work), but maki=
ng the spec flexible enough that a 'proxy capability' can be added later.
> Question is how to leave room for this capability. One suggestion was to =
keep aside one bit of the address-id - does that have consensus? Would any =
of the spec have to be re-thought?
>
> Thanks
> phil
>
> Philip Eardley
> Research and Technology Strategy
>
> This email contains BT information, which may be privileged or confidenti=
al. It's meant only for the individual(s) or entity named above. If you're =
not the intended recipient, note that disclosing, copying, distributing or =
using this information is prohibited. If you've received this email in erro=
r, please let me know immediately on the email address above. Thank you. We=
 monitor our email system, and may record your emails.
> British Telecommunications plc
> Registered office: 81 Newgate Street London EC1A 7AJ
> Registered in England no: 1800000
>
>
> -----Original Message-----
> From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org=
] On Behalf Of Lars Eggert
> Sent: 12 September 2011 07:28
> To: Olivier.Bonaventure@uclouvain.be
> Cc: multipathtcp@ietf.org List
> Subject: Re: [multipathtcp] Proposed changes to the protocol draft
>
> On 2011-9-10, at 12:43, Olivier Bonaventure wrote:
>> I fear that this would add unnecessary complexity and would prefer to
>> wait for more implementation experience before looking at such
>> optimisations in the main spec
>
> This is my feeling at the moment, too. I think it'd be useful to have thi=
s written up in an individual ID so that folks can discuss (and maybe try a=
nd implement) it, but I think it's too early to put this into the main spec=
.
>
> Lars
> _______________________________________________
> multipathtcp mailing list
> multipathtcp@ietf.org
> https://www.ietf.org/mailman/listinfo/multipathtcp
> _______________________________________________
> multipathtcp mailing list
> multipathtcp@ietf.org
> https://www.ietf.org/mailman/listinfo/multipathtcp
>

From yoshifumi.nishida@gmail.com  Tue Sep 27 02:20:34 2011
Return-Path: <yoshifumi.nishida@gmail.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D3D1221F8D9D for <multipathtcp@ietfa.amsl.com>; Tue, 27 Sep 2011 02:20:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.77
X-Spam-Level: 
X-Spam-Status: No, score=-102.77 tagged_above=-999 required=5 tests=[AWL=0.207, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3atFeaviJdSJ for <multipathtcp@ietfa.amsl.com>; Tue, 27 Sep 2011 02:20:34 -0700 (PDT)
Received: from mail-iy0-f172.google.com (mail-iy0-f172.google.com [209.85.210.172]) by ietfa.amsl.com (Postfix) with ESMTP id 60C9C21F8D94 for <multipathtcp@ietf.org>; Tue, 27 Sep 2011 02:20:34 -0700 (PDT)
Received: by iaby26 with SMTP id y26so6855871iab.31 for <multipathtcp@ietf.org>; Tue, 27 Sep 2011 02:23:19 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:date:x-google-sender-auth:message-id:subject :from:to:content-type; bh=uXSXlapjm9qWnmzVyZt6BL4IXBOntNRsqoZmlwXhRas=; b=WPFShHEIYPDzl/IYeJpxwmlEJubkzDUz7QUJS7XNY6SIwspBbW1U293nRzrp7PvZWq iDiq3ooqxCrtt7/iRyphQHBJKP+jjZcScqrumPioUDuaJ3Tv8PbKsWoZpSN0Gyt0irX9 wMklmMmpMpJeU9wk4yKXCYqUJH2ny/QtXhr20=
MIME-Version: 1.0
Received: by 10.42.155.134 with SMTP id u6mr7321212icw.146.1317115399380; Tue, 27 Sep 2011 02:23:19 -0700 (PDT)
Sender: yoshifumi.nishida@gmail.com
Received: by 10.231.199.80 with HTTP; Tue, 27 Sep 2011 02:23:19 -0700 (PDT)
Date: Tue, 27 Sep 2011 02:23:19 -0700
X-Google-Sender-Auth: gS_txLdxQXAmgV9Q1QCfCHPdbkY
Message-ID: <CABxaRLExsfwr-zStiP==faBBgiFX9c7jfybvfshBuXPNH28X4A@mail.gmail.com>
From: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
To: multipathtcp <multipathtcp@ietf.org>
Content-Type: text/plain; charset=ISO-8859-1
Subject: [multipathtcp] meeting at Taipei?
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 27 Sep 2011 09:20:34 -0000

Hi folks,

Phil and I would like to know the opinions for having a WG meeting at Taipei.
We tentatively sent a request for 1.5 hour meeting today and need to
decide it soon.
Another option would be a having interim audio meeting in October to
finalize the protocol spec and skip the Taipei meeting.

Please let us know your opinion on this.

Thanks,
--
Yoshifumi & Phil

From olivier.bonaventure@uclouvain.be  Tue Sep 27 06:46:46 2011
Return-Path: <olivier.bonaventure@uclouvain.be>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 442DF21F8C6B for <multipathtcp@ietfa.amsl.com>; Tue, 27 Sep 2011 06:46:46 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.299
X-Spam-Level: 
X-Spam-Status: No, score=-6.299 tagged_above=-999 required=5 tests=[AWL=0.300,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id vRBKbgqd0jfU for <multipathtcp@ietfa.amsl.com>; Tue, 27 Sep 2011 06:46:45 -0700 (PDT)
Received: from smtp6.sgsi.ucl.ac.be (smtp.sgsi.ucl.ac.be [130.104.5.67]) by ietfa.amsl.com (Postfix) with ESMTP id 524CF21F888A for <multipathtcp@ietf.org>; Tue, 27 Sep 2011 06:46:45 -0700 (PDT)
Received: from mbpobo.local (host-85-27-91-91.brutele.be [85.27.91.91]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: obonaventure@smtp6.sgsi.ucl.ac.be) by smtp6.sgsi.ucl.ac.be (Postfix) with ESMTPSA id 11AB31C560F; Tue, 27 Sep 2011 15:49:27 +0200 (CEST)
X-DKIM: Sendmail DKIM Filter v2.8.3 smtp6.sgsi.ucl.ac.be 11AB31C560F
DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=uclouvain.be; s=selucl; t=1317131367; bh=Hkv2JC2RbhCuNiWuWSJoe6A7gPXMU2kl/8B6h3jpIQg=; h=Message-ID:Date:From:Reply-To:MIME-Version:To:CC:Subject: References:In-Reply-To:Content-Type:Content-Transfer-Encoding; b=QV3hPMG/DNVbk0/fYumyBdicGTYkeUyQyjVQ5chvV65lru6/3Ok9QWPCvzoyjmsMy gzeYhJKl/3no2q8nRxdr8BkIVNCzjDBXCCZsmAGuSvLF4gFfuBlNjqzyXpYG0e9sU3 9erjJA4G5qnbzogNTgyaE28XUmyYrm8kHjKGpecQ=
Message-ID: <4E81D467.7000103@uclouvain.be>
Date: Tue, 27 Sep 2011 15:49:27 +0200
From: Olivier Bonaventure <Olivier.Bonaventure@uclouvain.be>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:6.0.2) Gecko/20110902 Thunderbird/6.0.2
MIME-Version: 1.0
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com> <4E6B312E.8060707@uclouvain.be> <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com> <9510D26531EF184D9017DF24659BB87F330B6BBA95@EMV65-UKRD.domain1.systemhost.net> <154773479ED2314980CB638A48FC443486FE3311@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEsQ9MQ-CSfx_uxpcRzxomicjbC+doEGy46zU+8Vv+4-A@mail.gmail.com>
In-Reply-To: <CABxaRLEsQ9MQ-CSfx_uxpcRzxomicjbC+doEGy46zU+8Vv+4-A@mail.gmail.com>
X-Enigmail-Version: 1.3.1
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: clamav-milter 0.97-exp at smtp-6.sipr-dc.ucl.ac.be
X-Virus-Status: Clean
X-Sgsi-Spamcheck: SASL authenticated, 
X-SGSI-MailScanner-ID: 11AB31C560F.A5F39
X-SGSI-MailScanner: Found to be clean
X-SGSI-From: olivier.bonaventure@uclouvain.be
X-SGSI-Spam-Status: No
Cc: "multipathtcp@ietf.org" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: Olivier.Bonaventure@uclouvain.be
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 27 Sep 2011 13:46:46 -0000

Yoshifumi,
> 
> I think we need more inputs or opinions for this.
> I'm trying to create a simple summary for the proposals from the
> previous discussion.
> Please correct me if I'm missing or misunderstanding something.
> 
> 
>  1) Add a byte to ADD_ADDRESS
>      pros: simple change to the current spec. good extensibility
>      cons: need one more byte
> 
>  2) Use one bit of the address-id
>      pros: simple change to the current spec.
>      cons: less address-id space, low extensibility


Looking at the ADD_ADDRESS format, we might also revise the IPver nibble :

                           1                   2                   3
       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
      +---------------+---------------+-------+-------+---------------+
      |     Kind      |     Length    |Subtype| IPVer |  Address ID   |
      +---------------+---------------+-------+-------+---------------+
      |          Address (IPv4 - 4 octets / IPv6 - 16 octets)         |
      +-------------------------------+---------------+---------------+
      |   Port (2 octets, optional)   |
      +-------------------------------+

         Figure 12: Add Address (ADD_ADDR) option (shown for IPv4)


Currently, we use 4 bits to specify either IPv4 or IPv6. This is a bit
expensive for one bit of information. We could replace these 4 bits by
two bits (to have some flexibility in the future, say IPv4=00 and
IPv6=01) and leave two bits reserved for future use.



Olivier


-- 
INL, ICTEAM, UCLouvain, Belgium, http://inl.info.ucl.ac.be

From georg.hampel@alcatel-lucent.com  Tue Sep 27 07:33:26 2011
Return-Path: <georg.hampel@alcatel-lucent.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 48E9C21F8B3F for <multipathtcp@ietfa.amsl.com>; Tue, 27 Sep 2011 07:33:26 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.162
X-Spam-Level: 
X-Spam-Status: No, score=-6.162 tagged_above=-999 required=5 tests=[AWL=-0.163, BAYES_00=-2.599, J_CHICKENPOX_21=0.6, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2NxoOb3YrZor for <multipathtcp@ietfa.amsl.com>; Tue, 27 Sep 2011 07:33:25 -0700 (PDT)
Received: from ihemail2.lucent.com (ihemail2.lucent.com [135.245.0.35]) by ietfa.amsl.com (Postfix) with ESMTP id 7377321F8B8D for <multipathtcp@ietf.org>; Tue, 27 Sep 2011 07:33:25 -0700 (PDT)
Received: from usnavsmail2.ndc.alcatel-lucent.com (usnavsmail2.ndc.alcatel-lucent.com [135.3.39.10]) by ihemail2.lucent.com (8.13.8/IER-o) with ESMTP id p8REa9vH029814 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Tue, 27 Sep 2011 09:36:10 -0500 (CDT)
Received: from USNAVSXCHHUB01.ndc.alcatel-lucent.com (usnavsxchhub01.ndc.alcatel-lucent.com [135.3.39.110]) by usnavsmail2.ndc.alcatel-lucent.com (8.14.3/8.14.3/GMO) with ESMTP id p8REa4D3032091 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT); Tue, 27 Sep 2011 09:36:09 -0500
Received: from USNAVSXCHMBSA2.ndc.alcatel-lucent.com ([135.3.39.124]) by USNAVSXCHHUB01.ndc.alcatel-lucent.com ([135.3.39.110]) with mapi; Tue, 27 Sep 2011 09:36:04 -0500
From: "Hampel, K Georg (K Georg)" <georg.hampel@alcatel-lucent.com>
To: Yoshifumi Nishida <nishida@sfc.wide.ad.jp>
Date: Tue, 27 Sep 2011 09:36:03 -0500
Thread-Topic: [multipathtcp] Proposed changes to the protocol draft
Thread-Index: Acx89Kb2aC8VDbx0RC6LIjzj5NHOTAAIyxmQ
Message-ID: <154773479ED2314980CB638A48FC443486FE3810@USNAVSXCHMBSA2.ndc.alcatel-lucent.com>
References: <9510D26531EF184D9017DF24659BB87F330AFB97A2@EMV65-UKRD.domain1.systemhost.net> <CAOs_kTark+NQ1Js9iBSwhwetXWHm6QAXneRmGZWiRpM7uLJpSQ@mail.gmail.com> <CAOs_kTbnSk4=nBxCdVSAmCcCyscmXDdDBGWuEud_hYxjwRUehA@mail.gmail.com> <461C6AC0-D817-4908-85EB-EAB3ABFB04AD@cs.ucl.ac.uk> <57AB2D9B-1902-4215-8CA4-5FA14C56ACCF@cs.ucl.ac.uk> <CABxaRLFMRsoUkgBOaj2PO=iPWKqV6jH9fqic5X9HDgEp_pmNHw@mail.gmail.com> <154773479ED2314980CB638A48FC443486DE961D@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEqKv962SWya-wyL-UcrJT4TeNn6KewKX=UCn1SqtQZUg@mail.gmail.com> <4E6B312E.8060707@uclouvain.be> <3087A9A7-6B59-431F-B344-35E8E7E1527E@nokia.com> <9510D26531EF184D9017DF24659BB87F330B6BBA95@EMV65-UKRD.domain1.systemhost.net> <154773479ED2314980CB638A48FC443486FE3311@USNAVSXCHMBSA2.ndc.alcatel-lucent.com> <CABxaRLEsQ9MQ-CSfx_uxpcRzxomicjbC+doEGy46zU+8Vv+4-A@mail.gmail.com>
In-Reply-To: <CABxaRLEsQ9MQ-CSfx_uxpcRzxomicjbC+doEGy46zU+8Vv+4-A@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Scanned-By: MIMEDefang 2.57 on 135.245.2.35
X-Scanned-By: MIMEDefang 2.64 on 135.3.39.10
Cc: "multipathtcp@ietf.org" <multipathtcp@ietf.org>
Subject: Re: [multipathtcp] Proposed changes to the protocol draft
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 27 Sep 2011 14:33:26 -0000

Hi Yoshifumi, all,

I think you summarized the various options and their respective tradeoffs v=
ery well.

I'm not completely certain, however, on what we are trying to accomplish. W=
hile the transparent proxy proposed in our I-D demands one additional bit i=
n ADD_ADDRESS, I'm not sure that one bit is enough to cover *all* (present =
and/or future) meanings of the ADD_ADDRESS option.=20

The WG further seems to have agreed that the proxy concept requires more wo=
rk. I'm not sure we can anticipate the associated signaling changes without=
 having gone through this work. Why should the present WG draft try to pree=
mpt this work?

The proxy discussions may have created the feeling that ADD_ADDRESS *princi=
pally* lacks sufficient space to cover future implications/actions/descript=
ions associated with an address. Such space extension should indeed be capt=
ured within the present WG draft. However, I think it should not be guided =
by preemptive guesses on what a proxy may need.


-Georg




-----Original Message-----
From: yoshifumi.nishida@gmail.com [mailto:yoshifumi.nishida@gmail.com] On B=
ehalf Of Yoshifumi Nishida
Sent: Tuesday, September 27, 2011 5:06 AM
To: Hampel, K Georg (K Georg)
Cc: multipathtcp@ietf.org
Subject: Re: [multipathtcp] Proposed changes to the protocol draft

Hi Georg, All,

I think we need more inputs or opinions for this.
I'm trying to create a simple summary for the proposals from the
previous discussion.
Please correct me if I'm missing or misunderstanding something.


 1) Add a byte to ADD_ADDRESS
     pros: simple change to the current spec. good extensibility
     cons: need one more byte

 2) Use one bit of the address-id
     pros: simple change to the current spec.
     cons: less address-id space, low extensibility

 3) Use more bits of the address-id
     pros: simple change to the current spec, some extensibility
     cons: much less address-id space,

 4) Keep the current format and develop new option to describe the
address (e.g DESC_ADDR)
     pros: great extensibility. no need to change the current spec.
     cons: might need another mechanism to sync 2 options reliably.

Also, I personally think the following can be an another possibility,
although it might not be a good idea.

 5) Keep the current format and develop new option to specify proxy or
other addresses (e. g ADD_ADDR2)
    pros: great extensibility. no need to change the current spec
    cons: having 2 options for mostly the same feature might be a bit
ugly design.

Thanks,
--
Yoshifumi Nishida
nishida@sfc.wide.ad.jp

2011/9/26 Hampel, K Georg (K Georg) <georg.hampel@alcatel-lucent.com>:
> Hi Phil, all,
>
> Are we certain that 1 bit is enough to cover all purposes and implication=
s of ADD_ADDRESS?
>
> Based on recent proxy-related drafts and discussions, ADD_ADDRESS would a=
lready have to support (1) "Join using this address" and (2) "I'm a proxy".=
 Since both implications are independent from each other and may occur in u=
nison, 1 bit would not be enough.
>
> A more detailed discussion on proxy functions may reveal additional impli=
cations of this option.
>
> Adding one byte to the option instead of carving out 1 bit should provide=
 sufficient room for proxy and other extensions.
>
> Thanks.
>
> Regards,
>
> Georg
>
>
>
> -----Original Message-----
> From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org=
] On Behalf Of philip.eardley@bt.com
> Sent: Friday, September 23, 2011 11:08 AM
> To: multipathtcp@ietf.org
> Subject: Re: [multipathtcp] Proposed changes to the protocol draft
>
> There seems to be consensus around completing the main spec without tryin=
g to solve the proxy issue (since it needs significant more work), but maki=
ng the spec flexible enough that a 'proxy capability' can be added later.
> Question is how to leave room for this capability. One suggestion was to =
keep aside one bit of the address-id - does that have consensus? Would any =
of the spec have to be re-thought?
>
> Thanks
> phil
>
> Philip Eardley
> Research and Technology Strategy
>
> This email contains BT information, which may be privileged or confidenti=
al. It's meant only for the individual(s) or entity named above. If you're =
not the intended recipient, note that disclosing, copying, distributing or =
using this information is prohibited. If you've received this email in erro=
r, please let me know immediately on the email address above. Thank you. We=
 monitor our email system, and may record your emails.
> British Telecommunications plc
> Registered office: 81 Newgate Street London EC1A 7AJ
> Registered in England no: 1800000
>
>
> -----Original Message-----
> From: multipathtcp-bounces@ietf.org [mailto:multipathtcp-bounces@ietf.org=
] On Behalf Of Lars Eggert
> Sent: 12 September 2011 07:28
> To: Olivier.Bonaventure@uclouvain.be
> Cc: multipathtcp@ietf.org List
> Subject: Re: [multipathtcp] Proposed changes to the protocol draft
>
> On 2011-9-10, at 12:43, Olivier Bonaventure wrote:
>> I fear that this would add unnecessary complexity and would prefer to
>> wait for more implementation experience before looking at such
>> optimisations in the main spec
>
> This is my feeling at the moment, too. I think it'd be useful to have thi=
s written up in an individual ID so that folks can discuss (and maybe try a=
nd implement) it, but I think it's too early to put this into the main spec=
.
>
> Lars
> _______________________________________________
> multipathtcp mailing list
> multipathtcp@ietf.org
> https://www.ietf.org/mailman/listinfo/multipathtcp
> _______________________________________________
> multipathtcp mailing list
> multipathtcp@ietf.org
> https://www.ietf.org/mailman/listinfo/multipathtcp
>

From philip.eardley@bt.com  Thu Sep 29 02:08:32 2011
Return-Path: <philip.eardley@bt.com>
X-Original-To: multipathtcp@ietfa.amsl.com
Delivered-To: multipathtcp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9122321F8D23 for <multipathtcp@ietfa.amsl.com>; Thu, 29 Sep 2011 02:08:32 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.835
X-Spam-Level: 
X-Spam-Status: No, score=-102.835 tagged_above=-999 required=5 tests=[AWL=0.163, BAYES_00=-2.599, HTML_MESSAGE=0.001, J_CHICKENPOX_21=0.6, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kTiSmXrESCXt for <multipathtcp@ietfa.amsl.com>; Thu, 29 Sep 2011 02:08:28 -0700 (PDT)
Received: from smtpe1.intersmtp.com (smtp64.intersmtp.com [62.239.224.237]) by ietfa.amsl.com (Postfix) with ESMTP id 76AD621F8CB0 for <multipathtcp@ietf.org>; Thu, 29 Sep 2011 02:08:27 -0700 (PDT)
Received: from EVMHT64-UKRD.domain1.systemhost.net (10.36.3.101) by RDW083A008ED64.smtp-e4.hygiene.service (10.187.98.13) with Microsoft SMTP Server (TLS) id 8.3.159.2; Thu, 29 Sep 2011 10:11:24 +0100
Received: from EMV65-UKRD.domain1.systemhost.net ([169.254.2.5]) by EVMHT64-UKRD.domain1.systemhost.net ([10.36.3.101]) with mapi; Thu, 29 Sep 2011 10:11:16 +0100
From: <philip.eardley@bt.com>
To: <multipathtcp@ietf.org>
Date: Thu, 29 Sep 2011 10:11:15 +0100
Thread-Topic: Interim (audio) meeting
Thread-Index: Acx+h73xOmObcsR1QQCFBnTG4kay9A==
Message-ID: <9510D26531EF184D9017DF24659BB87F330C2475F9@EMV65-UKRD.domain1.systemhost.net>
Accept-Language: en-US, en-GB
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US, en-GB
Content-Type: multipart/alternative; boundary="_000_9510D26531EF184D9017DF24659BB87F330C2475F9EMV65UKRDdoma_"
MIME-Version: 1.0
Subject: [multipathtcp] Interim (audio) meeting
X-BeenThere: multipathtcp@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multi-path extensions for TCP <multipathtcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/multipathtcp>
List-Post: <mailto:multipathtcp@ietf.org>
List-Help: <mailto:multipathtcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/multipathtcp>, <mailto:multipathtcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 29 Sep 2011 09:08:32 -0000

--_000_9510D26531EF184D9017DF24659BB87F330C2475F9EMV65UKRDdoma_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

We would like to hold an interim audio meeting to conclude on the remaining=
 open issues on the protocol doc, in particular concluding on the proxy iss=
ues.
Please keep discussing on the list in the meantime, then the meeting can be=
 very short!

For the previous interims, discussion about the time has suggested that 3pm=
 GMT is a good time, suggest we keep the same time (=3D 4pm BST, 5pm CEST, =
11am EDT, 8am PDT, etc). (Please let us know if this is a particularly bad =
choice for you, and we can try to re-optimise)

Suggestion is week of 17th October.
Please fill in the doodle ASAP - this week - so we can give the 2 weeks not=
ice.

http://www.doodle.com/9wztb29pithgy9pe#table

Thanks
Phil & Yoshifumi.

Philip Eardley
Research and Technology Strategy

This email contains BT information, which may be privileged or confidential=
. It's meant only for the individual(s) or entity named above. If you're no=
t the intended recipient, note that disclosing, copying, distributing or us=
ing this information is prohibited. If you've received this email in error,=
 please let me know immediately on the email address above. Thank you. We m=
onitor our email system, and may record your emails.
British Telecommunications plc
Registered office: 81 Newgate Street London EC1A 7AJ
Registered in England no: 1800000


--_000_9510D26531EF184D9017DF24659BB87F330C2475F9EMV65UKRDdoma_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:=
//www.w3.org/TR/REC-html40"><head><meta http-equiv=3DContent-Type content=
=3D"text/html; charset=3Dus-ascii"><meta name=3DGenerator content=3D"Micros=
oft Word 12 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0cm;
	margin-bottom:.0001pt;
	font-size:11.0pt;
	font-family:"Calibri","sans-serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-compose;
	font-family:"Calibri","sans-serif";
	color:windowtext;}
.MsoChpDefault
	{mso-style-type:export-only;}
@page WordSection1
	{size:612.0pt 792.0pt;
	margin:72.0pt 72.0pt 72.0pt 72.0pt;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-GB link=3Dblue vli=
nk=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal>We would like to=
 hold an interim audio meeting to conclude on the remaining open issues on =
the protocol doc, in particular concluding on the proxy issues.<o:p></o:p><=
/p><p class=3DMsoNormal>Please keep discussing on the list in the meantime,=
 then the meeting can be very short!<o:p></o:p></p><p class=3DMsoNormal><o:=
p>&nbsp;</o:p></p><p class=3DMsoNormal>For the previous interims, discussio=
n about the time has suggested that 3pm GMT is a good time, suggest we keep=
 the same time (=3D 4pm BST, 5pm CEST, 11am EDT, 8am PDT, etc). (Please let=
 us know if this is a particularly bad choice for you, and we can try to re=
-optimise)<o:p></o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=
=3DMsoNormal>Suggestion is week of 17<sup>th</sup> October.<o:p></o:p></p><=
p class=3DMsoNormal>Please fill in the doodle ASAP &#8211; this week &#8211=
; so we can give the 2 weeks notice.<o:p></o:p></p><p class=3DMsoNormal><o:=
p>&nbsp;</o:p></p><p class=3DMsoNormal><a href=3D"http://www.doodle.com/9wz=
tb29pithgy9pe#table">http://www.doodle.com/9wztb29pithgy9pe#table</a> <o:p>=
</o:p></p><p class=3DMsoNormal><o:p>&nbsp;</o:p></p><p class=3DMsoNormal>Th=
anks<o:p></o:p></p><p class=3DMsoNormal>Phil &amp; Yoshifumi.<b><span style=
=3D'font-size:9.0pt;font-family:"Arial","sans-serif"'><o:p></o:p></span></b=
></p><p class=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-botto=
m-alt:auto'><b><span style=3D'font-size:9.0pt;font-family:"Arial","sans-ser=
if"'><o:p>&nbsp;</o:p></span></b></p><p class=3DMsoNormal style=3D'mso-marg=
in-top-alt:auto;mso-margin-bottom-alt:auto'><b><span style=3D'font-size:9.0=
pt;font-family:"Arial","sans-serif"'>Philip Eardley&nbsp;&nbsp;&nbsp;</span=
></b><span style=3D'font-size:9.0pt;font-family:"Arial","sans-serif"'><br><=
/span><b><span style=3D'font-size:9.0pt;font-family:"Arial","sans-serif"'>R=
esearch and Technology Strategy</span></b><b><span style=3D'font-size:9.0pt=
;font-family:"Arial","sans-serif"'><br><br></span></b><span style=3D'font-s=
ize:9.0pt;font-family:"Arial","sans-serif"'><o:p></o:p></span></p><p class=
=3DMsoNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><=
span style=3D'font-size:9.0pt;font-family:"Arial","sans-serif"'>This email =
contains BT information, which may be privileged or confidential. It's mean=
t only for the individual(s) or entity named above. If you're not the inten=
ded recipient, note that disclosing, copying, distributing or using this in=
formation is prohibited. If you've received this email in error, please let=
 me know immediately on the email address above. Thank you. We monitor our =
email system, and may record your emails.<o:p></o:p></span></p><p class=3DM=
soNormal style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span=
 style=3D'font-size:9.0pt;font-family:"Arial","sans-serif"'>British Telecom=
munications plc<br>Registered office: 81 Newgate Street London EC1A 7AJ<br>=
Registered in England no: 1800000<o:p></o:p></span></p><p class=3DMsoNormal=
><o:p>&nbsp;</o:p></p></div></body></html>=

--_000_9510D26531EF184D9017DF24659BB87F330C2475F9EMV65UKRDdoma_--
