
From tyoshino@google.com  Tue May  7 00:26:23 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 544B121F8E3C for <hybi@ietfa.amsl.com>; Tue,  7 May 2013 00:26:23 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.977
X-Spam-Level: 
X-Spam-Status: No, score=-101.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001, 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 jw0T4huf1Xkh for <hybi@ietfa.amsl.com>; Tue,  7 May 2013 00:26:22 -0700 (PDT)
Received: from mail-ea0-x236.google.com (mail-ea0-x236.google.com [IPv6:2a00:1450:4013:c01::236]) by ietfa.amsl.com (Postfix) with ESMTP id 80DF621F855A for <hybi@ietf.org>; Tue,  7 May 2013 00:26:22 -0700 (PDT)
Received: by mail-ea0-f182.google.com with SMTP id z16so116117ead.13 for <hybi@ietf.org>; Tue, 07 May 2013 00:26:21 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:content-type; bh=IhEepBf41RQdzk3x0A1glvDxgvsVFBQp7w8S6zLoUVE=; b=d+adPXKjn6QM9NkvQh8Y8mVF02zwn+LdjZEXGg9PY+ft/B+luvYMNO0GCQiWfhfCte DVSywSk3RhajXgGq04nah7ZOvNakDH57G61zYgTUXOluNcvbMBdPd/kBIjoWlw080oYm quHxn539HqRlowUCeiIXCoM4cp81o6gm7JGx/uQpGsB7HBs1YNgFY/iqmmB3h/fG2jNO c8kEQFmld7oMvrGwS+MyqwBProJa37JcFQIg/A+qcB+Vm++j8dGUVOqmw6cNpcO1johT DFSFUgfCXjbQHAruh7Mo57J5m3Whrf/aP3Je8xvU6wx8EJERBc/FlCdEVuhYHES7hBiQ bSAg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:content-type:x-gm-message-state; bh=IhEepBf41RQdzk3x0A1glvDxgvsVFBQp7w8S6zLoUVE=; b=HPZ7XiXeZ407G3/ChTxSVQLMe4iXXvWVEhge56/ulvwLXI+0VPON5jlCxHv+vfh9sA QxujbMRdFwja6Xu1qpHTUb9y8qf5jjmsXelZTnDhPW9aB/+n9BHtu6uykAtV7lg/M29L H178YIYrVfFKA5/7ilZtslgyvJA3X9hltyVAFCZ7OS48Vgg7H96iQyhAcOMK53OwOv6A hVZsnV82pXBSK48eB9aVpNuQJr3qEZd378b+d4GJUyU4ZDPe2d/H6SNqwX4Bgmwh6bRm 6+vieEjemrrXQ2DkxrRrXMq/T9vyFk6aFniteSVxsZM+I0c3jO0LiRQ5zJ4evi45yYeX NWQA==
X-Received: by 10.14.3.137 with SMTP id 9mr2622721eeh.0.1367911581512; Tue, 07 May 2013 00:26:21 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.85.152 with HTTP; Tue, 7 May 2013 00:26:01 -0700 (PDT)
In-Reply-To: <CAH9hSJa4B6bRmCf8QrrBf9O8a4tR2WApvDuOJ0ih=Rh_NP-FYg@mail.gmail.com>
References: <CAG4zZZB0AFMvfYjegrZ73mMeMwngUV5C7fKkqFNqwgnjD5jz=Q@mail.gmail.com> <CAH9hSJYdT+Aez9D240ZPSAbTumXu1Baovwh-Zk9p35vKSD2=Pw@mail.gmail.com> <CAH9hSJa4B6bRmCf8QrrBf9O8a4tR2WApvDuOJ0ih=Rh_NP-FYg@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Tue, 7 May 2013 16:26:01 +0900
Message-ID: <CAH9hSJYzhLezgtyZJy3vKD1Nq6pMdHpp6K6czW3Q-D2GwggMMA@mail.gmail.com>
To: Joakim Erdfelt <joakim@intalio.com>, "hybi@ietf.org" <hybi@ietf.org>
Content-Type: multipart/alternative; boundary=047d7b6043c8e6ea2404dc1bbb68
X-Gm-Message-State: ALoCoQm7lKFSEJOftAUyubwxIvEz70mA9SKwxsqqHuMXRegeQgj8SwZ34nKLyCoux0PKUeu6mznG62PPsX1KQ6Bw6u0pVYg8VMSuEAA7x0CYWHrVVCLFw8w+b6jaAria8Rh8nYNybC2jPch1ZNs99RgHPzYq4i31lpigBBLjl1jCUTvUQvHFE8j9o6+yJ4TrvLzITvb6egnc
Subject: Re: [hybi] Reference Implementation of draft-ietf-hybi-permessage-compression-07 ?
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 07 May 2013 07:26:23 -0000

--047d7b6043c8e6ea2404dc1bbb68
Content-Type: text/plain; charset=ISO-8859-1

Hi,

Chrome Canary now supports permessage-deflate behind a flag.

You can turn it on by accessing chrome://flags, enabling "Enable
experimental WebSocket implementation" flags and then restarting the Chrome.

Chrome will send x-webkit-deflate-frame and permessage-deflate in this
order. We'll change it to send out permessage-deflate first this week.

Please tell me if you find any bug.

On Thu, Apr 25, 2013 at 11:16 PM, Takeshi Yoshino <tyoshino@google.com>wrote:

> Hi Joakim,
>
> permessage-compress is now implemented on echo_client.py of pywebsocket.
> You can do simple send/receive test using --use-permessage-deflate
> parameter.
>
>
Sorry. This is typo. s/permessage-compress/permessage-deflate/

--047d7b6043c8e6ea2404dc1bbb68
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hi,<div><br></div><div>Chrome Canary now supports permessa=
ge-deflate behind a flag.</div><div><br></div><div style>You can turn it on=
 by accessing chrome://flags, enabling &quot;Enable experimental WebSocket =
implementation&quot; flags and then restarting the Chrome.</div>

<div style><br></div><div style>Chrome will send x-webkit-deflate-frame and=
 permessage-deflate in this order. We&#39;ll change it to send out permessa=
ge-deflate first this week.</div><div style><br></div><div style>Please tel=
l me if you find any bug.</div>

<div class=3D"gmail_extra"><br><div class=3D"gmail_quote">On Thu, Apr 25, 2=
013 at 11:16 PM, Takeshi Yoshino <span dir=3D"ltr">&lt;<a href=3D"mailto:ty=
oshino@google.com" target=3D"_blank">tyoshino@google.com</a>&gt;</span> wro=
te:<br>

<blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-=
left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;p=
adding-left:1ex"><div dir=3D"ltr">Hi Joakim,<div><br></div><div>permessage-=
compress is now implemented on echo_client.py of pywebsocket. You can do si=
mple send/receive test using --use-permessage-deflate parameter.</div>

<div><br></div></div></blockquote><div><br></div><div style>Sorry. This is =
typo. s/permessage-compress/permessage-deflate/</div><div><br></div></div><=
/div></div>

--047d7b6043c8e6ea2404dc1bbb68--

From muehlbauer.marc@googlemail.com  Mon May 13 02:17:38 2013
Return-Path: <muehlbauer.marc@googlemail.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 93D5121F93E5 for <hybi@ietfa.amsl.com>; Mon, 13 May 2013 02:17:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.121
X-Spam-Level: 
X-Spam-Status: No, score=-0.121 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, MIME_8BIT_HEADER=0.3, RCVD_IN_DNSWL_LOW=-1, SARE_OBFU_AMP2B=2.555]
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 ECpE2b9X0YlA for <hybi@ietfa.amsl.com>; Mon, 13 May 2013 02:17:34 -0700 (PDT)
Received: from mail-lb0-f178.google.com (mail-lb0-f178.google.com [209.85.217.178]) by ietfa.amsl.com (Postfix) with ESMTP id 2764E21F93D7 for <hybi@ietf.org>; Mon, 13 May 2013 02:17:29 -0700 (PDT)
Received: by mail-lb0-f178.google.com with SMTP id p10so2436718lbv.37 for <hybi@ietf.org>; Mon, 13 May 2013 02:17:28 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=20120113; h=mime-version:x-received:date:message-id:subject:from:to :content-type; bh=GbZShKOM/1f+MzOuIY3f+0TL534smHcozXF0YMz0hw4=; b=0QKN2XID7wQjAd6T2DpLD9qvPcmdmXjpT2JHazK4Su0mErt8Zt2sULwm/2MjuGylPO +6xs+j8sUTeG7XC8bmiYyQea0xhItMPz14CR/a0H/suCA2Xr7YSC/7V+eEVTmJalNl1x U7QE95qtaGGhEKyUGr3ePkxwnl1y6SmDo1NtsP3rYDL2W5h9eSxBhb07bJb5XhYYWs4k 8/Qzeiat1uDgVucKIeaCbd+ZFPmAQPIj4cdWB7361JKZpy7O96HffbKzzWV9jTacXnBC KqSAYQASisfVH8bgZ8I1O9FfcI2x/1q3iJuzh0S5oyX37rwBEKOaYr7yUfiJN9zsAQtE zhEg==
MIME-Version: 1.0
X-Received: by 10.112.138.228 with SMTP id qt4mr12102430lbb.106.1368436648788;  Mon, 13 May 2013 02:17:28 -0700 (PDT)
Received: by 10.112.134.167 with HTTP; Mon, 13 May 2013 02:17:28 -0700 (PDT)
Date: Mon, 13 May 2013 11:17:28 +0200
Message-ID: <CAKeGy=cUO3yACsf8pKiNO+n9g7+j0V-B9yJjnaZ7FC4dwtNgqQ@mail.gmail.com>
From: =?ISO-8859-1?Q?Marc_M=FChlbauer?= <muehlbauer.marc@googlemail.com>
To: hybi@ietf.org
Content-Type: multipart/alternative; boundary=089e0112bfd05948d104dc95fc6a
Subject: [hybi] Client certificate authentication for WSS websockets using twisted and autobahn
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 13 May 2013 09:19:24 -0000

--089e0112bfd05948d104dc95fc6a
Content-Type: text/plain; charset=ISO-8859-1

Dear subscribers,

I'm currently facing an issue while implementing client certificate
authentication for the use of secure websockets using the twisted and
autobahn python libraries.

Following the instructions of Tobias Oberstein, the CEO of tavendo which
hosts the autobahn library, I managed to get the authentication working in
the IE 10 ( at
http://stackoverflow.com/questions/16234429/how-can-i-setup-an-autobahn-pub-sub-server-and-a-autobahn-webserver-listening-onyou
can see the conversation and part of my source code).

The problem seems to be, as far as I can tell, that the security context of
the SSL session is not used for the access to the websocket. In other
words: If I authenticate with a client certificate to retrieve the page
running on https://, Firefox and Chrome don't ask me again to authenticate
for the use of the secure websocket and also dont use the recent
authentication for that. This means, the client is not authenticated and
hence isnt allowed to use the websocket.

In IE 10, it works as intended.

Now I'm curious if this has something to do with the different browser
behaviour or the root cause of the problem can be found on server side.
Note that webserver and websocket are running on the same port. For test
purposes, the whole thing runs only locally on localhost:9000 and I'm using
self signed certificates.

Any advice would be really nice!

If you guys need further information, please let me know.

Thanks in advance!

Greetings, Marc

--089e0112bfd05948d104dc95fc6a
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div><div><div><div><div><div><div><div>Dear subscribers,<=
br><br></div>I&#39;m currently facing an issue while implementing client ce=
rtificate authentication for the use of secure websockets using the twisted=
 and autobahn python libraries.<br>
<br></div>Following the instructions of Tobias Oberstein, the CEO of tavend=
o which hosts the autobahn library, I managed to get the authentication wor=
king in the IE 10 ( at <a href=3D"http://stackoverflow.com/questions/162344=
29/how-can-i-setup-an-autobahn-pub-sub-server-and-a-autobahn-webserver-list=
ening-on">http://stackoverflow.com/questions/16234429/how-can-i-setup-an-au=
tobahn-pub-sub-server-and-a-autobahn-webserver-listening-on</a> you can see=
 the conversation and part of my source code).<br>
<br></div>The problem seems to be, as far as I can tell, that the security =
context of the SSL session is not used for the access to the websocket. In =
other words: If I authenticate with a client certificate to retrieve the pa=
ge running on https://, Firefox and Chrome don&#39;t ask me again to authen=
ticate for the use of the secure websocket and also dont use the recent aut=
hentication for that. This means, the client is not authenticated and hence=
 isnt allowed to use the websocket.<br>
<br></div>In IE 10, it works as intended.<br><br></div>Now I&#39;m curious =
if this has something to do with the different browser behaviour or the roo=
t cause of the problem can be found on server side. Note that webserver and=
 websocket are running on the same port. For test purposes, the whole thing=
 runs only locally on localhost:9000 and I&#39;m using self signed certific=
ates.<br>
<br></div>Any advice would be really nice!<br><br></div><div>If you guys ne=
ed further information, please let me know.<br></div><div><br></div>Thanks =
in advance!<br><br></div>Greetings, Marc<br></div>

--089e0112bfd05948d104dc95fc6a--

From tyoshino@google.com  Tue May 14 01:56:15 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A962B21F90AC for <hybi@ietfa.amsl.com>; Tue, 14 May 2013 01:56:15 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -100.121
X-Spam-Level: 
X-Spam-Status: No, score=-100.121 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, MIME_8BIT_HEADER=0.3, RCVD_IN_DNSWL_LOW=-1, SARE_OBFU_AMP2B=2.555, 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 VMJsJfNdEdU0 for <hybi@ietfa.amsl.com>; Tue, 14 May 2013 01:56:11 -0700 (PDT)
Received: from mail-ee0-f42.google.com (mail-ee0-f42.google.com [74.125.83.42]) by ietfa.amsl.com (Postfix) with ESMTP id 3DA5F21F90B3 for <hybi@ietf.org>; Tue, 14 May 2013 01:56:10 -0700 (PDT)
Received: by mail-ee0-f42.google.com with SMTP id c50so164277eek.1 for <hybi@ietf.org>; Tue, 14 May 2013 01:56:10 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type; bh=55Lub3Tzbugx0HDnbeXMsvDqkpBmZ3XePLBDGi7ZB3Y=; b=PMPt9Qqs6ef05tKqNRvg3AnZ/wE5M/uXj4BOyHokkhJXWGHxZMXSzruO26/WyjC2kh 7WTGWoqUF7zu/lKmuDkbZ9tiCaMvC5IVq93DdjbdGo9QaS23mnwKXxGfBPUxE7SEBKRk Z3qxohdzEgxtZ1JwCywyg6ut/tUCOS5pc+cxPVxaTsUGLUbv2Jr49Qj1yyjp89vt0N4m evUSBvd3Ts7u0QP3zPO1lUI7/Gy/ePwuV3RAMfGOtLKx0gh3ej3dF2ObXaSkqlwRiqOe l0Xc0Oyx022tDM4/Wfj4CGT2eA4ljXW191NAueAQmBK4ojd5MvNYM3l4woT49N/8WRwb 6DDQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type:x-gm-message-state; bh=55Lub3Tzbugx0HDnbeXMsvDqkpBmZ3XePLBDGi7ZB3Y=; b=i59epuni2Wd0GNxHg25u8dYBC0d3VfodiXa1cvlWFl2RdPPhZaVr0dCoWFgpGo7XM6 M14FEbPaejg3/n9D5IfguzXRUYoQTn79fllHnC5QqBOvjd4ZZHewJjTnfD/tE9sS7iPz II9755mbRnn5/t6sXyYcOH8ypgtrj2c37g6w6exHRar9Tc0MmZ82zMwoABt/axso0vuW qbLOVDhyTJNoez3ketFV2OZ5hwe6CO/NuiVK+JGvVw+whFxwm+RGOmWrfV38gIyBY3gU Ml6ggUqReOGXNC6QVayzkXQ/n0smJJe3eoNqU8oB/IZ2+GVaCHx67yg7tMpWCbbwhu3D d8ew==
X-Received: by 10.14.7.198 with SMTP id 46mr25552143eep.17.1368521770167; Tue, 14 May 2013 01:56:10 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.85.152 with HTTP; Tue, 14 May 2013 01:55:49 -0700 (PDT)
In-Reply-To: <CAKeGy=cUO3yACsf8pKiNO+n9g7+j0V-B9yJjnaZ7FC4dwtNgqQ@mail.gmail.com>
References: <CAKeGy=cUO3yACsf8pKiNO+n9g7+j0V-B9yJjnaZ7FC4dwtNgqQ@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Tue, 14 May 2013 17:55:49 +0900
Message-ID: <CAH9hSJZLqsL0ANUxcVSLeeLbAFpY9JfXQzgrSgwwiDB6YOVZ7w@mail.gmail.com>
To: =?ISO-8859-1?Q?Marc_M=FChlbauer?= <muehlbauer.marc@googlemail.com>
Content-Type: multipart/alternative; boundary=001a11c1b764faa63304dca9cdee
X-Gm-Message-State: ALoCoQk3nO1FPeXyaWIJoPfrtRj7vu4YZ93hWQL02dc6gCO2gP7GqA+PlkT1IZxZIIRJfqdBbb3d4sSAMn+y5sxgkzadXHWp7N4Il1D5rwcszPVSYRtv5GAIquUhO2Be/tQ94kbSsbURGXqdKMwS+NUp5vmlVfBqXZ+eMQcSaHzJEMGAlk/icONtc0eG1ACXGICiCjYUIS7w
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Client certificate authentication for WSS websockets using twisted and autobahn
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 14 May 2013 08:56:15 -0000

--001a11c1b764faa63304dca9cdee
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

On Mon, May 13, 2013 at 6:17 PM, Marc M=FChlbauer <
muehlbauer.marc@googlemail.com> wrote:

> I'm currently facing an issue while implementing client certificate
> authentication for the use of secure websockets using the twisted and
> autobahn python libraries.
>
> Following the instructions of Tobias Oberstein, the CEO of tavendo which
> hosts the autobahn library, I managed to get the authentication working i=
n
> the IE 10 ( at
> http://stackoverflow.com/questions/16234429/how-can-i-setup-an-autobahn-p=
ub-sub-server-and-a-autobahn-webserver-listening-onyou can see the conversa=
tion and part of my source code).
>
> The problem seems to be, as far as I can tell, that the security context
> of the SSL session is not used for the access to the websocket. In other
> words: If I authenticate with a client certificate to retrieve the page
> running on https://, Firefox and Chrome don't ask me again to
> authenticate for the use of the secure websocket and also dont use the
> recent authentication for that. This means, the client is not authenticat=
ed
> and hence isnt allowed to use the websocket.
>

Chrome does support wss client authentication using a certificate. We don't
show any prompt but just use the cached info for that host.

Can you get any debug information from your server about the TLS
negotiation? As it's not clear if it's related to the spec or just an issue
with Chrome, please file a bug at crbug.com and let's discuss there.


> In IE 10, it works as intended.
>
> Now I'm curious if this has something to do with the different browser
> behaviour or the root cause of the problem can be found on server side.
> Note that webserver and websocket are running on the same port. For test
> purposes, the whole thing runs only locally on localhost:9000 and I'm usi=
ng
> self signed certificates.
>

Using self signed one should be fine.

--001a11c1b764faa63304dca9cdee
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Mon, May 13, 2013 at 6:17 PM, Marc M=FChlbauer <span di=
r=3D"ltr">&lt;<a href=3D"mailto:muehlbauer.marc@googlemail.com" target=3D"_=
blank">muehlbauer.marc@googlemail.com</a>&gt;</span> wrote:<br><div class=
=3D"gmail_extra">


<div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margi=
n:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir=3D"ltr">=
<div><div><div><div><div><div><div><div>I&#39;m currently facing an issue w=
hile implementing client certificate authentication for the use of secure w=
ebsockets using the twisted and autobahn python libraries.<br>


</div>
<br></div>Following the instructions of Tobias Oberstein, the CEO of tavend=
o which hosts the autobahn library, I managed to get the authentication wor=
king in the IE 10 ( at <a href=3D"http://stackoverflow.com/questions/162344=
29/how-can-i-setup-an-autobahn-pub-sub-server-and-a-autobahn-webserver-list=
ening-on" target=3D"_blank">http://stackoverflow.com/questions/16234429/how=
-can-i-setup-an-autobahn-pub-sub-server-and-a-autobahn-webserver-listening-=
on</a> you can see the conversation and part of my source code).<br>



<br></div>The problem seems to be, as far as I can tell, that the security =
context of the SSL session is not used for the access to the websocket. In =
other words: If I authenticate with a client certificate to retrieve the pa=
ge running on https://, Firefox and Chrome don&#39;t ask me again to authen=
ticate for the use of the secure websocket and also dont use the recent aut=
hentication for that. This means, the client is not authenticated and hence=
 isnt allowed to use the websocket.<br>


</div></div></div></div></div></div></blockquote><div><br></div><div>Chrome=
 does support wss client authentication using a certificate. We don&#39;t s=
how any prompt but just use the cached info for that host.</div><div><br>

</div><div style>Can you get any debug information from your server about t=
he TLS negotiation? As it&#39;s not clear if it&#39;s related to the spec o=
r just an issue with Chrome, please file a bug at <a href=3D"http://crbug.c=
om">crbug.com</a> and let&#39;s discuss there.</div>

<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex">
<div dir=3D"ltr"><div><div><div><div><div></div>In IE 10, it works as inten=
ded.<br><br></div>Now I&#39;m curious if this has something to do with the =
different browser behaviour or the root cause of the problem can be found o=
n server side. Note that webserver and websocket are running on the same po=
rt. For test purposes, the whole thing runs only locally on localhost:9000 =
and I&#39;m using self signed certificates.<br>

</div></div></div></div></blockquote></div><br></div><div class=3D"gmail_ex=
tra" style>Using self signed one should be fine.</div><div class=3D"gmail_e=
xtra"><br></div></div>

--001a11c1b764faa63304dca9cdee--

From internet-drafts@ietf.org  Wed May 15 04:10:41 2013
Return-Path: <internet-drafts@ietf.org>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CE62221F89A5; Wed, 15 May 2013 04:10:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.476
X-Spam-Level: 
X-Spam-Status: No, score=-102.476 tagged_above=-999 required=5 tests=[AWL=0.124, BAYES_00=-2.599, NO_RELAYS=-0.001, 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 b7SwG9f6L1ch; Wed, 15 May 2013 04:10:40 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 19A4221F84B2; Wed, 15 May 2013 04:10:40 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From: internet-drafts@ietf.org
To: i-d-announce@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 4.45
Message-ID: <20130515111039.8252.96955.idtracker@ietfa.amsl.com>
Date: Wed, 15 May 2013 04:10:39 -0700
Cc: hybi@ietf.org
Subject: [hybi] I-D Action: draft-ietf-hybi-websocket-multiplexing-10.txt
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 15 May 2013 11:10:42 -0000

A New Internet-Draft is available from the on-line Internet-Drafts director=
ies.
 This draft is a work item of the BiDirectional or Server-Initiated HTTP Wo=
rking Group of the IETF.

	Title           : A Multiplexing Extension for WebSockets
	Author(s)       : John A. Tamplin
                          Takeshi Yoshino
	Filename        : draft-ietf-hybi-websocket-multiplexing-10.txt
	Pages           : 43
	Date            : 2013-05-15

Abstract:
   The WebSocket Protocol [RFC6455] requires a new transport connection
   for every WebSocket connection.  This presents a scalability problem
   when many clients connect to the same server, and is made worse by
   having multiple clients running in different tabs of the same user
   agent.  This extension provides a way for separate logical WebSocket
   connections to share an underlying transport connection.

   Please send feedback to the hybi@ietf.org mailing list.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-hybi-websocket-multiplexing

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-hybi-websocket-multiplexing-10

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-hybi-websocket-multiplexing-10


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


From tyoshino@google.com  Wed May 15 04:15:15 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2A13721F84E3 for <hybi@ietfa.amsl.com>; Wed, 15 May 2013 04:15:15 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.549
X-Spam-Level: 
X-Spam-Status: No, score=-101.549 tagged_above=-999 required=5 tests=[AWL=1.427, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, 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 0NXX7iXSf7oJ for <hybi@ietfa.amsl.com>; Wed, 15 May 2013 04:15:11 -0700 (PDT)
Received: from mail-ee0-f54.google.com (mail-ee0-f54.google.com [74.125.83.54]) by ietfa.amsl.com (Postfix) with ESMTP id 8FE0621F85DC for <hybi@ietf.org>; Wed, 15 May 2013 04:15:10 -0700 (PDT)
Received: by mail-ee0-f54.google.com with SMTP id e50so998990eek.41 for <hybi@ietf.org>; Wed, 15 May 2013 04:15:09 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:content-type; bh=3FDFfrht4OlaB0sDkuwtGn4ma76afw57ChYVTX68al8=; b=JZYCNOPfVGJ2GcwSq2bmX4h0xbBxD1v85GLdtKfJqK7/V0sHGqR/PEWOz3FDluR564 19KsOUYIL3ud62G00PMCMfOO8paBG9NMsfFJl5K/T3gqmRzRX/C+/Q4/cfkQWCW4yxNH mne0E9qKLkq5L7pxf2wK7JZ9aTq86EqUl0QfeQg/i+pTCQ48daxA15UBMzOqP1Mqv3ac H3uRZqhEvK/d9poHH6pZ51nO0rmfDsCr/imaXGSMCCr9IR2P3RnV5LaXz/7BYsbIntTY QARxsCTyYcOkFYpOUiDTU5quf2T/sXGdc42m58mC89zc04NsO0Bz6QIgFbZ4su8DPA2P MM5Q==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:content-type:x-gm-message-state; bh=3FDFfrht4OlaB0sDkuwtGn4ma76afw57ChYVTX68al8=; b=cRHLVdwPtpM3/fcSf1tHAZ0lYyU0t54NSBwlXTkAlyJpHPTco2oanzR8ZYjG3LfoVk zoYOmX5yCDZ6MoXHTxY71vF8xthtYFDTVKYQizRgpAQGDJhx28o1+oRy/JXIjScabyK6 z5WF/LPLR5FBjNw6GsXrfoAJsr+mlNXqSsRM+TEaUxusqUqHA0dlJKktmXvNKakUAC1r 9s/RCqawrb/Qam/d/MnLigbtO7DehdinE0UByNJXAoJTIuPHARxGs6NecDT3uz/sTipq +2fFO5x2kCSkYkbf4tBzJVsLIvatzpw1tyKXJT3x45BBRdemaQSElm04uUIiQFR/MLyB wIkw==
X-Received: by 10.14.179.133 with SMTP id h5mr102463708eem.34.1368616509511; Wed, 15 May 2013 04:15:09 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.85.152 with HTTP; Wed, 15 May 2013 04:14:49 -0700 (PDT)
In-Reply-To: <20130515111039.8252.96955.idtracker@ietfa.amsl.com>
References: <20130515111039.8252.96955.idtracker@ietfa.amsl.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Wed, 15 May 2013 20:14:49 +0900
Message-ID: <CAH9hSJYpixfZ9MJ-FPyeN_K=if86zQDn9AUdYKkEa5H4tpnzeA@mail.gmail.com>
To: "hybi@ietf.org" <hybi@ietf.org>
Content-Type: multipart/alternative; boundary=047d7b624308e25d4904dcbfdcad
X-Gm-Message-State: ALoCoQlVS1K/vNelqOmBx6XYYesErWhi4OmFxbXx5UcsOkwXe6SfYS3WTMxTGXcshzGIAP3HZ7ldIuclT0UyhxX78P6IZLzsT5F20xbtM61hlGHBhBUu0/3zLVinTdr80H9nJfLT4qeoWcyW1u72y3aCLv/ce0OKeWbXJu2002fDqae9LFJszO29D2lVFosk4/4Hz3BOjIU7
Subject: Re: [hybi] I-D Action: draft-ietf-hybi-websocket-multiplexing-10.txt
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 15 May 2013 11:15:15 -0000

--047d7b624308e25d4904dcbfdcad
Content-Type: text/plain; charset=ISO-8859-1

Just bumped up the version number to prevent expiration.

I had pending fix but I couldn't include them because of machine trouble.

Takeshi


On Wed, May 15, 2013 at 8:10 PM, <internet-drafts@ietf.org> wrote:

>
> A New Internet-Draft is available from the on-line Internet-Drafts
> directories.
>  This draft is a work item of the BiDirectional or Server-Initiated HTTP
> Working Group of the IETF.
>
>         Title           : A Multiplexing Extension for WebSockets
>         Author(s)       : John A. Tamplin
>                           Takeshi Yoshino
>         Filename        : draft-ietf-hybi-websocket-multiplexing-10.txt
>         Pages           : 43
>         Date            : 2013-05-15
>
> Abstract:
>    The WebSocket Protocol [RFC6455] requires a new transport connection
>    for every WebSocket connection.  This presents a scalability problem
>    when many clients connect to the same server, and is made worse by
>    having multiple clients running in different tabs of the same user
>    agent.  This extension provides a way for separate logical WebSocket
>    connections to share an underlying transport connection.
>
>    Please send feedback to the hybi@ietf.org mailing list.
>
>
> The IETF datatracker status page for this draft is:
> https://datatracker.ietf.org/doc/draft-ietf-hybi-websocket-multiplexing
>
> There's also a htmlized version available at:
> http://tools.ietf.org/html/draft-ietf-hybi-websocket-multiplexing-10
>
> A diff from the previous version is available at:
> http://www.ietf.org/rfcdiff?url2=draft-ietf-hybi-websocket-multiplexing-10
>
>
> Internet-Drafts are also available by anonymous FTP at:
> ftp://ftp.ietf.org/internet-drafts/
>
> _______________________________________________
> hybi mailing list
> hybi@ietf.org
> https://www.ietf.org/mailman/listinfo/hybi
>

--047d7b624308e25d4904dcbfdcad
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Just bumped up the version number to prevent expiration.<d=
iv class=3D"gmail_extra"><br></div><div class=3D"gmail_extra" style>I had p=
ending fix but I couldn&#39;t include them because of machine trouble.</div=
><div class=3D"gmail_extra">

<br clear=3D"all"><div>Takeshi</div>
<br><br><div class=3D"gmail_quote">On Wed, May 15, 2013 at 8:10 PM,  <span =
dir=3D"ltr">&lt;<a href=3D"mailto:internet-drafts@ietf.org" target=3D"_blan=
k">internet-drafts@ietf.org</a>&gt;</span> wrote:<br><blockquote class=3D"g=
mail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-l=
eft:1ex">

<br>
A New Internet-Draft is available from the on-line Internet-Drafts director=
ies.<br>
=A0This draft is a work item of the BiDirectional or Server-Initiated HTTP =
Working Group of the IETF.<br>
<br>
=A0 =A0 =A0 =A0 Title =A0 =A0 =A0 =A0 =A0 : A Multiplexing Extension for We=
bSockets<br>
=A0 =A0 =A0 =A0 Author(s) =A0 =A0 =A0 : John A. Tamplin<br>
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Takeshi Yoshino<br>
=A0 =A0 =A0 =A0 Filename =A0 =A0 =A0 =A0: draft-ietf-hybi-websocket-multipl=
exing-10.txt<br>
=A0 =A0 =A0 =A0 Pages =A0 =A0 =A0 =A0 =A0 : 43<br>
=A0 =A0 =A0 =A0 Date =A0 =A0 =A0 =A0 =A0 =A0: 2013-05-15<br>
<br>
Abstract:<br>
=A0 =A0The WebSocket Protocol [RFC6455] requires a new transport connection=
<br>
=A0 =A0for every WebSocket connection. =A0This presents a scalability probl=
em<br>
=A0 =A0when many clients connect to the same server, and is made worse by<b=
r>
=A0 =A0having multiple clients running in different tabs of the same user<b=
r>
=A0 =A0agent. =A0This extension provides a way for separate logical WebSock=
et<br>
=A0 =A0connections to share an underlying transport connection.<br>
<br>
=A0 =A0Please send feedback to the <a href=3D"mailto:hybi@ietf.org">hybi@ie=
tf.org</a> mailing list.<br>
<br>
<br>
The IETF datatracker status page for this draft is:<br>
<a href=3D"https://datatracker.ietf.org/doc/draft-ietf-hybi-websocket-multi=
plexing" target=3D"_blank">https://datatracker.ietf.org/doc/draft-ietf-hybi=
-websocket-multiplexing</a><br>
<br>
There&#39;s also a htmlized version available at:<br>
<a href=3D"http://tools.ietf.org/html/draft-ietf-hybi-websocket-multiplexin=
g-10" target=3D"_blank">http://tools.ietf.org/html/draft-ietf-hybi-websocke=
t-multiplexing-10</a><br>
<br>
A diff from the previous version is available at:<br>
<a href=3D"http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-hybi-websocket-mul=
tiplexing-10" target=3D"_blank">http://www.ietf.org/rfcdiff?url2=3Ddraft-ie=
tf-hybi-websocket-multiplexing-10</a><br>
<br>
<br>
Internet-Drafts are also available by anonymous FTP at:<br>
<a href=3D"ftp://ftp.ietf.org/internet-drafts/" target=3D"_blank">ftp://ftp=
.ietf.org/internet-drafts/</a><br>
<br>
_______________________________________________<br>
hybi mailing list<br>
<a href=3D"mailto:hybi@ietf.org">hybi@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/hybi" target=3D"_blank">ht=
tps://www.ietf.org/mailman/listinfo/hybi</a><br>
</blockquote></div><br></div></div>

--047d7b624308e25d4904dcbfdcad--

From tyoshino@google.com  Wed May 15 04:45:51 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C42CA21F8FA5 for <hybi@ietfa.amsl.com>; Wed, 15 May 2013 04:45:51 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.977
X-Spam-Level: 
X-Spam-Status: No, score=-101.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001, 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 5muhpFoOUa5q for <hybi@ietfa.amsl.com>; Wed, 15 May 2013 04:45:51 -0700 (PDT)
Received: from mail-ea0-x235.google.com (mail-ea0-x235.google.com [IPv6:2a00:1450:4013:c01::235]) by ietfa.amsl.com (Postfix) with ESMTP id 9D2F621F8F4A for <hybi@ietf.org>; Wed, 15 May 2013 04:45:50 -0700 (PDT)
Received: by mail-ea0-f181.google.com with SMTP id a11so1007260eae.40 for <hybi@ietf.org>; Wed, 15 May 2013 04:45:49 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:from:date:message-id:subject:to :content-type; bh=6AIFy3bVs9k8oVot0Xu3PV5lSd7+nZw6zvLCFwC5wIE=; b=gNQ1/lRaQz9mlreuxndVcnnzCqMeCM8HDzRM4SGj9XMGgeHKvgbwa4wK016vXg8AQo aFfCqjOA0KM5SRPisEaonf6910gB38Jemwyo8LyrdDOPNaRIXphAJFNzfGfMDja1kCKp LVdMlOPzaUIWcsghDsS01gbF0ZkZ4UKd+U38UfV4MtDKIvK2LfizUNebL5xv/H/rv5iE YLNjNZ1HBOCuX0Zjf/89sjCQXiG0VpdB8f4qeN7VyJ0CUdDB/wu46MI/Lb0zEZnFVCZm BjXNFw8LHrmZbqd+YkKaEt2HCVCs1LG64BUcU6ps+dyPxQgSyBc9qYcRikD0QvJQf0e/ b3tQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:from:date:message-id:subject:to :content-type:x-gm-message-state; bh=6AIFy3bVs9k8oVot0Xu3PV5lSd7+nZw6zvLCFwC5wIE=; b=CEuvDWmAS+iUIhSEHMTLONF7FUstXv2hJxQ5CmZKAYtNTxQ8CFJho/5iZP55a8hnk1 jyJljAJ51GE0qAULfCOYtBqI7OaqyhgAcE+tbMC/4NWTxLWEmunlblV+ujgGBkUcFqR1 jBjkdTYqsQ7SCwbtq7Ei+XjAs8LIe6xjLxfRAXW9UmPobvg44r3dtUEoGDNW8Bu40Vdf 5qN3OsqcNBkJJTbIkNpQfANVeD4ycsRZCTCPz0KirAtRZkPEkRciFmYoJlaWdZkD/ofq XTocDeFtYc/BAPfEFjiZb/xqpbOso5zWTeWVGR0B2E29LgUiMpAV5vNrkyTVwJaV8O+P X2Iw==
X-Received: by 10.14.7.198 with SMTP id 46mr39308718eep.17.1368618349658; Wed, 15 May 2013 04:45:49 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.85.152 with HTTP; Wed, 15 May 2013 04:45:28 -0700 (PDT)
From: Takeshi Yoshino <tyoshino@google.com>
Date: Wed, 15 May 2013 20:45:28 +0900
Message-ID: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com>
To: "hybi@ietf.org" <hybi@ietf.org>
Content-Type: multipart/alternative; boundary=001a11c1b76490b94604dcc04a24
X-Gm-Message-State: ALoCoQnox237D0h35gzb4MBOhDYzTZ3QfzRznhcZGPkmWWRCsLzbQ11exmIzh2Lr9czFIXX7jBQs0axsTSKbdPTJe4lQgPoUgQ3RnFzjqz2fE1urxSFcWg0N2Gx1abtdWLdsko72eRNsv2e7syXBTqV8tJkwfjmg8oSHtTQj1mOdIDllBnFsdlatJGV2UE9aQjn0SMx/7yk4
Subject: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 15 May 2013 11:45:51 -0000

--001a11c1b76490b94604dcc04a24
Content-Type: text/plain; charset=ISO-8859-1

Hi all,

(Note: This is not WG's official process)

I'd like to survey how many people are interested in standardization and
implementation of multiplexing dedicated for WebSocket again at this point.

* Multiplexing is still important for WebSocket

HyBi WG has been working on the WebSocket multiplexing spec (
http://tools.ietf.org/html/draft-ietf-hybi-websocket-multiplexing-10).
Multiplexing functionality is really important to make WebSocket scalable
enough to be used for large scale production services. It's still true.

* Rapid adoption of HTTP/2.0

While our work is delayed and delayed again, httpbis folks started
seriously working on next generation HTTP. Now it's actively standardized
as HTTP/2.0 based on Google's SPDY experiments and input from httpbis
members. Gradually but several web service providers already started
adopting HTTP/2.0.

* WebSocket/SPDY experiment

Chrome WebSocket team has been working on both WebSocket/SPDY and dedicated
multiplexing in parallel. You can find the WebSocket/SPDY proposal
incorporated into Roberto's unofficial draft (section 4) at
https://github.com/grmocg/SPDY-Specification/blob/gh-pages/draft-mbelshe-spdy-00.txt.
Its framing is not so efficient as the dedicated one, but can coexist with
HTTP traffic in one TCP connection and developers just need to understand
and implement HTTP/2.0 framing.

* Concentrate on the best one

Developing and maintaining two different multiplexing code will be burden
for vendors. It seems that we should stop and think which approach we
should focus on.

Thanks

--001a11c1b76490b94604dcc04a24
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hi all,
<div><br></div><div style>(Note: This is not WG&#39;s official process)</di=
v><div><br></div><div>I&#39;d like to survey how many people are interested=
 in standardization and implementation of multiplexing dedicated for WebSoc=
ket again at this point.<br>

</div><div><br></div><div style>* Multiplexing is still important for WebSo=
cket</div><div style><br></div><div style>HyBi WG has been working on the W=
ebSocket multiplexing spec (<a href=3D"http://tools.ietf.org/html/draft-iet=
f-hybi-websocket-multiplexing-10">http://tools.ietf.org/html/draft-ietf-hyb=
i-websocket-multiplexing-10</a>). Multiplexing functionality is really impo=
rtant to make WebSocket scalable enough to be used for large scale producti=
on services. It&#39;s still true.</div>

<div><br></div><div style>* Rapid adoption of HTTP/2.0</div><div style><br>=
</div><div>While our work is delayed and delayed again, httpbis folks start=
ed seriously working on next generation HTTP. Now it&#39;s actively standar=
dized as HTTP/2.0 based on Google&#39;s SPDY experiments and input from htt=
pbis members. Gradually but several web service providers already started a=
dopting HTTP/2.0.<br>

</div><div style><br></div><div style>* WebSocket/SPDY experiment</div><div=
 style><br></div><div style>Chrome WebSocket team has been working on both =
WebSocket/SPDY and dedicated multiplexing in parallel. You can find the Web=
Socket/SPDY proposal incorporated into Roberto&#39;s unofficial draft (sect=
ion 4) at=A0<a href=3D"https://github.com/grmocg/SPDY-Specification/blob/gh=
-pages/draft-mbelshe-spdy-00.txt">https://github.com/grmocg/SPDY-Specificat=
ion/blob/gh-pages/draft-mbelshe-spdy-00.txt</a>. Its framing is not so effi=
cient as the dedicated one, but can coexist with HTTP traffic in one TCP co=
nnection and developers just need to understand and implement HTTP/2.0 fram=
ing.</div>

<div style><br></div><div style>* Concentrate on the best one</div><div sty=
le><br></div><div style>Developing and maintaining two different multiplexi=
ng code will be burden for vendors. It seems that we should stop and think =
which approach we should focus on.</div>

<div style><br></div><div style>Thanks</div></div>

--001a11c1b76490b94604dcc04a24--

From tyoshino@google.com  Wed May 15 04:59:47 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0AC6A21F8C69 for <hybi@ietfa.amsl.com>; Wed, 15 May 2013 04:59:47 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.262
X-Spam-Level: 
X-Spam-Status: No, score=-102.262 tagged_above=-999 required=5 tests=[AWL=0.714, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, 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 RJha-kv+X7vw for <hybi@ietfa.amsl.com>; Wed, 15 May 2013 04:59:42 -0700 (PDT)
Received: from mail-ee0-f41.google.com (mail-ee0-f41.google.com [74.125.83.41]) by ietfa.amsl.com (Postfix) with ESMTP id 3CB6021F882A for <hybi@ietf.org>; Wed, 15 May 2013 04:59:37 -0700 (PDT)
Received: by mail-ee0-f41.google.com with SMTP id d4so986347eek.14 for <hybi@ietf.org>; Wed, 15 May 2013 04:59:36 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:content-type; bh=ZzZ0sqV+svlc/LKXPwSytArKf9nlInwewpIKWNXhJbE=; b=HFq7/NYEnGYgb53xR0ol6abfEiubeYqFo4ta10nFMdRXCalzs5f8538D/pa3CLv8be lqHJtzC/RF6ufMfIDqcHDpSbxivEt1gAJsSBsfq6MqfgN7UVAxEM612P+3G/hA3FI7Gz wfOCMOzk5CV9w3LCMTO04kjj+Tlc2OFMCqFYhzDwz3R82pHCqPCfKGMA0Viu3ZHMhlHt 3ZADSur7ykoX0bY8/Yv815aUNeU46uE5aXDdkYswF64Ff227srcqtlpaqRwHsmfmmjxO f/3BeZHAXOhPi4svFOQ0/SaUMyQEMuFQa2Vz6H6GMRJzYeaQ2ArB6pmsN4B4aBPUBYXD nXZw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:content-type:x-gm-message-state; bh=ZzZ0sqV+svlc/LKXPwSytArKf9nlInwewpIKWNXhJbE=; b=Ja5BhKvPKt3mPluM5FE/zlLiy1KoKPtltTB3vCMigHpFxU7F4FmS2xcHdpzaVYN8cC bZLTlW2Gb3CccZuSzrpAACOhUtHlqiANDR8xNfiv4BN8VQp0UrQd08AL/AwS9u6ozWCv IXpjJN+2Db8J/w1QEI92fe+0WpsB4IhO8wQlSOBtCXeThTsbT6fxmsePOdV/S7Et0Xon SJLzAAoO0lVH+MsG60KENk4PRiVTgHP9NAF5IRiIOBov183Aj73316RmrJBlzsnn+NE8 BN2o5QiwSq/GrWn3ytH4UbhPcXHH9cFslvSjLvZx9e2unYTuwSiQzmnwBGleUVQ7s9eu U4Lw==
X-Received: by 10.14.3.137 with SMTP id 9mr103400298eeh.0.1368619176278; Wed, 15 May 2013 04:59:36 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.85.152 with HTTP; Wed, 15 May 2013 04:59:16 -0700 (PDT)
In-Reply-To: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Wed, 15 May 2013 20:59:16 +0900
Message-ID: <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com>
To: "hybi@ietf.org" <hybi@ietf.org>
Content-Type: multipart/alternative; boundary=047d7b6043c8d60ad604dcc07b48
X-Gm-Message-State: ALoCoQnbr52CNDdJNb8OhHHEFbN+LG4JHoW81dIOqE8sZUeKr6ka8kqSUlO7RmvS58mruLQYZ6WbVDfDKT63JCzkzbDDdJ7QFBTC1ilD7Yz6U8RfbvlPJLW8gDCWq8LxksTwZoNXhKBnNM8AEy2KdyPQW64gYXx2RRnE+ky5o/XaGLKERXnjoZVHMLtWsb5z0zF3G+E1tw1j
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 15 May 2013 11:59:47 -0000

--047d7b6043c8d60ad604dcc07b48
Content-Type: text/plain; charset=ISO-8859-1

On Wed, May 15, 2013 at 8:45 PM, Takeshi Yoshino <tyoshino@google.com>wrote:

> * Concentrate on the best one
>
> Developing and maintaining two different multiplexing code will be burden
> for vendors. It seems that we should stop and think which approach we
> should focus on.
>

I'm not saying we must abandon one of them.

If it turned out both are still considered useful and supported by certain
number of people, we can continue both.

--047d7b6043c8d60ad604dcc07b48
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Wed, May 15, 2013 at 8:45 PM, Takeshi Yoshino <span dir=
=3D"ltr">&lt;<a href=3D"mailto:tyoshino@google.com" target=3D"_blank">tyosh=
ino@google.com</a>&gt;</span> wrote:<br><div class=3D"gmail_extra"><div cla=
ss=3D"gmail_quote">

<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div dir=3D"ltr"><div>* Concentrate on the b=
est one<br></div><div><br></div><div>Developing and maintaining two differe=
nt multiplexing code will be burden for vendors. It seems that we should st=
op and think which approach we should focus on.</div>

</div></blockquote><div><br></div><div style>I&#39;m not saying we must aba=
ndon one of them.</div><div style><br></div><div style>If it turned out bot=
h are still considered useful and supported by certain number of people, we=
 can continue both.</div>

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

--047d7b6043c8d60ad604dcc07b48--

From tyoshino@google.com  Sat May 18 21:31:02 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B3E4721F8756 for <hybi@ietfa.amsl.com>; Sat, 18 May 2013 21:31:02 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.5
X-Spam-Level: 
X-Spam-Status: No, score=-102.5 tagged_above=-999 required=5 tests=[AWL=0.476,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, 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 pvAsGQ9o8+uR for <hybi@ietfa.amsl.com>; Sat, 18 May 2013 21:30:57 -0700 (PDT)
Received: from mail-ee0-f53.google.com (mail-ee0-f53.google.com [74.125.83.53]) by ietfa.amsl.com (Postfix) with ESMTP id 5B27921F869A for <hybi@ietf.org>; Sat, 18 May 2013 21:30:57 -0700 (PDT)
Received: by mail-ee0-f53.google.com with SMTP id c1so3027526eek.40 for <hybi@ietf.org>; Sat, 18 May 2013 21:30:56 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:content-type; bh=+KX3y/jZxTVxUSYjrlxC6iqPjZlGP6O5736IXb+pBXY=; b=Pq6PgyonT1Wf5WPQWfthdOKb42En7R7Cd/o0ul01b4oK2bknyrPR4eLdwAYDL2Ookh Z0AW4ZdsoivPtmqSogAV+vRY6VIEdxTKVajp4NMbNhs2IsQI1lDUfQGN52YV7wRDO6YS j79D1Xq60b4h4MxTrKTipUM7Qar8x3YJ2QGKc/IYeTURbpEkpNJlIWCvKiijqWQ3hBPl DejzsMwKC6K0/UZQGx1wHbk9KAbMNxzrolyI9dGY9age3f3fpKDglGWin9k3ZsIG4OI9 8Gg1Zt+uiMDn3ZRLQSw/OHProFE/WTFSGk5NU/aK2yLwqjljajDUDRgomO7kyuyGqg12 +znw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:content-type:x-gm-message-state; bh=+KX3y/jZxTVxUSYjrlxC6iqPjZlGP6O5736IXb+pBXY=; b=J8YT7StLnn7QZUNUGzQxuzujYa1QD5ykvVrw9OnEk4pXEqkqB3Vn3bgEbnGk5PKUHr 8z/KAUsq+VEHBgySn2ynCqerPkPtHBoKzvpzocvw8wVj6cB51RA0UgUK+sT2fBcnuzo0 zMeS0t8NYWlcgp6xlSPpWtZy4k12u3qmhWeen13Jmp/ukIgOhECGlIbd6l9CkWieXhFC zUKURF9CbdWci7KE5N9BQMjfwxNf8IRZDkPK6uSucDnLi3B3u8KefjdxKNnuSf7GICh/ jyiq8rLaP4DWMMF5YYa858g8yUODb7Hg8NzbZLKAZU7pQ+KBejZVgpawsKz11ugriEqw yaOw==
X-Received: by 10.14.172.195 with SMTP id t43mr7786808eel.34.1368937856258; Sat, 18 May 2013 21:30:56 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.85.152 with HTTP; Sat, 18 May 2013 21:30:36 -0700 (PDT)
In-Reply-To: <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Sun, 19 May 2013 13:30:36 +0900
Message-ID: <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com>
To: "hybi@ietf.org" <hybi@ietf.org>
Content-Type: multipart/alternative; boundary=047d7b621ee6a48a1904dd0aae14
X-Gm-Message-State: ALoCoQkNxL5651Vr9ZYZgSZzawezjrf2pTitLh5Q0EEZdmKoD9eS5COmMnko5IafzPYepZjhEIOoNN+jFBnhsnbMSXkQQFEQfaSkf9pR7eOyVl4R82nYDanlOv3VEwASrLX4r/hNd3ZpNlA+fYkYv6VIpWB9kPrptewzIhGEt37AvNfGEo/QtvWV9CZI8oubngGR45caflx2
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 19 May 2013 04:31:02 -0000

--047d7b621ee6a48a1904dd0aae14
Content-Type: text/plain; charset=ISO-8859-1

I got one comment about this.

He/she suggested that we finish the WebSocket dedicated multiplexing
quickly even by simplifying the spec much e.g. cutting some functionality
in it that would take much time to get consensus, and when HTTP/2.0 is
matured, switch to that.

--047d7b621ee6a48a1904dd0aae14
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">I got one comment about this.<div><br></div><div>He/she su=
ggested that we finish the WebSocket dedicated multiplexing quickly even by=
 simplifying the spec much e.g. cutting some functionality in it that would=
 take much time to get consensus, and when HTTP/2.0 is matured, switch to t=
hat.<div>



=A0</div></div></div>

--047d7b621ee6a48a1904dd0aae14--

From simone.bordet@gmail.com  Sun May 19 11:37:14 2013
Return-Path: <simone.bordet@gmail.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id ABE9321F87C5 for <hybi@ietfa.amsl.com>; Sun, 19 May 2013 11:37:14 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.978
X-Spam-Level: 
X-Spam-Status: No, score=-1.978 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, NO_RELAYS=-0.001]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PDnXB1j3RNCA for <hybi@ietfa.amsl.com>; Sun, 19 May 2013 11:37:13 -0700 (PDT)
Received: from mail-pb0-x22c.google.com (mail-pb0-x22c.google.com [IPv6:2607:f8b0:400e:c01::22c]) by ietfa.amsl.com (Postfix) with ESMTP id C38E121F8733 for <hybi@ietf.org>; Sun, 19 May 2013 11:37:13 -0700 (PDT)
Received: by mail-pb0-f44.google.com with SMTP id wz12so459145pbc.3 for <hybi@ietf.org>; Sun, 19 May 2013 11:37:13 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=usSgSCHsCOzVIdRRtQy+KNFE5dN4msDVSs31qrH+hsw=; b=poMyP6lPMtNvZOC/nUH724u9W3PQ07vOMhl5qBirHVapw1XkXPYha6YLFN9LgeL7SZ dHMXEl6mqI1xLePtrNv81fSu58MMcts3jIp3N0La1avI6X8eNWXbdJqQeXUo0qMoyKLg VGUqGlZHH3/SsmqI/OOdYG+vYNg2hnmw7AuU3d+eQfBYwX0tG7KOZ3lLF2hUtu4gSvff fG9R8TfOkdQGcHITMv2NvrwVZFFplKMkEuK88U6d9I1RO7BdXy72fsqxNo7CtKC6Vk2x 9RB9mSruKX/R5t46NK3ja+2xpjNCbWCGawean2/X8gAlX7HruB38V8XM3Sy3nbJSgf73 hdJQ==
MIME-Version: 1.0
X-Received: by 10.66.235.69 with SMTP id uk5mr39703480pac.110.1368988633248; Sun, 19 May 2013 11:37:13 -0700 (PDT)
Sender: simone.bordet@gmail.com
Received: by 10.68.103.129 with HTTP; Sun, 19 May 2013 11:37:13 -0700 (PDT)
In-Reply-To: <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com>
Date: Sun, 19 May 2013 20:37:13 +0200
X-Google-Sender-Auth: BahVMXytH0B0Xfuz92okx77Z_5k
Message-ID: <CAFWmRJ0naLw=Om70ag3p2xJ8mYvgNHLL6Z_DtpiOjVWZ3iS5Dg@mail.gmail.com>
From: Simone Bordet <sbordet@intalio.com>
To: Takeshi Yoshino <tyoshino@google.com>
Content-Type: text/plain; charset=UTF-8
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 19 May 2013 18:37:14 -0000

Hi,

On Sun, May 19, 2013 at 6:30 AM, Takeshi Yoshino <tyoshino@google.com> wrote:
> I got one comment about this.
>
> He/she suggested that we finish the WebSocket dedicated multiplexing quickly
> even by simplifying the spec much e.g. cutting some functionality in it that
> would take much time to get consensus, and when HTTP/2.0 is matured, switch
> to that.

My impression is that I have not seen much request for WebSocket multiplexing.
Given the complexities related to defining and implementing ws mux
correctly, and considering that HTTP/2.0 comes with mux built-in,
building on the SPDY experience, my preference would be to
deprioritize ws mux in favour of HTTP/2.0 integration.
It is true that there may be systems that may rely on a pure-ws
communication (and therefore do not need a full HTTP/2.0
implementation) and as such would benefit from ws mux, but my guess is
that those system will be a minority or can get by without ws mux. I
may be biased on this view though, so comments welcome.

Thanks !

--
Simone Bordet
----
http://cometd.org
http://webtide.com
http://intalio.com
Developer advice, training, services and support
from the Jetty & CometD experts.
Intalio, the modern way to build business applications.

From tyoshino@google.com  Sun May 19 21:02:14 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E5F3121F8F9E for <hybi@ietfa.amsl.com>; Sun, 19 May 2013 21:02:13 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.619
X-Spam-Level: 
X-Spam-Status: No, score=-102.619 tagged_above=-999 required=5 tests=[AWL=0.357, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, 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 gXUeh+TIsMaW for <hybi@ietfa.amsl.com>; Sun, 19 May 2013 21:02:08 -0700 (PDT)
Received: from mail-ee0-f51.google.com (mail-ee0-f51.google.com [74.125.83.51]) by ietfa.amsl.com (Postfix) with ESMTP id E9C7921F8FA5 for <hybi@ietf.org>; Sun, 19 May 2013 21:02:07 -0700 (PDT)
Received: by mail-ee0-f51.google.com with SMTP id e51so3635370eek.38 for <hybi@ietf.org>; Sun, 19 May 2013 21:02:07 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type; bh=EQ1/L4HGo71r/C1/OZ6Kpza240UA7gd72NDqZKTikY0=; b=io+zfG9ZPpK6p7sRKpVKmqM8Cox/rDY12kp/QVnn/9c6zCREwZDXbOlPlkB+ollK7h sNKirwp0Vjey4tjfRjmkSdlwdUtY7BxHP1U3opYM46JBtda1beC4NZ4J6ahC1GyPwkrV F0e+S5uM1xRIaJ7JrRYITHFD9UmmLhVbnozVoimQlPJqq8V3ptm+TFf8hASo+/xwQUBd 3v1DN1HXAaTtJYASsQLyDGUMdhJY4yq7f41D3IqlOMqE25R9ZRU3hFHfvC6Wb8Dfh+o0 lYIDKsMu5tAA175jp+Vt0AmLYxZG80NTPC6t5d3cp22hmWqWae/ssbNmOrtsIWI+F/Xk puSw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=x-received:mime-version:in-reply-to:references:from:date:message-id :subject:to:cc:content-type:x-gm-message-state; bh=EQ1/L4HGo71r/C1/OZ6Kpza240UA7gd72NDqZKTikY0=; b=F5OL4qhkf29s3Gdiappbm+QMgOexUXpVoJV5QeDPRTGajX+DKuXdWdMvnvpuSxAvO/ tdnaVcYHN+AYzmCYjOUU6AAGyfp2lJw/C7c9hp10V9QdjgHnBfaUtuBel7Uy0t3aZoHl Cst5mNBfPJVmxaU+LCpDV/dgAOByLu7eeckXWn1PFYS3yUUxX+gX5quqiw7wsD0sytE2 wTUqytE0NOKVghBwpEV5HAqss9ADq9DKTk7cVU9jEDAAQe9JtjiOaA2thmUAfm7zIu6w BYTbjB4B/8JUM7QkxxCh0scq/Fx4pljPV/72DBqeJKhAVQ7z9E5S94zYwcAFCRsNbeV4 4YtA==
X-Received: by 10.15.107.77 with SMTP id ca53mr32891212eeb.40.1369022527048; Sun, 19 May 2013 21:02:07 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.86.67 with HTTP; Sun, 19 May 2013 21:01:46 -0700 (PDT)
In-Reply-To: <CAFWmRJ0naLw=Om70ag3p2xJ8mYvgNHLL6Z_DtpiOjVWZ3iS5Dg@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <CAFWmRJ0naLw=Om70ag3p2xJ8mYvgNHLL6Z_DtpiOjVWZ3iS5Dg@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Mon, 20 May 2013 13:01:46 +0900
Message-ID: <CAH9hSJYQP4HOmQKxY2oRczy4B_Xyo+kh7AyQgavWZjeOqz+xaA@mail.gmail.com>
To: Simone Bordet <sbordet@intalio.com>
Content-Type: multipart/alternative; boundary=089e0162830c6a562004dd1e6597
X-Gm-Message-State: ALoCoQneMiQrooEuijmUbunK+y60OdHtl8wn7cixU1pxIAXknnOCji6UOsvKegjEHDbhI0URVPukdP3A/FFEAqiTwRyo0DmEKC+E1qouQ2WWT+pYdVPSdwv0P1A33PxitgeZwAMSxaCtP548LxDJTJPCPH0o+2iBvQp2H06MoybREZJnVWqpa/H6WBKRAeooFNQvR+i7aEKL
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 20 May 2013 04:02:14 -0000

--089e0162830c6a562004dd1e6597
Content-Type: text/plain; charset=ISO-8859-1

Thanks for comment, Simone.

On Mon, May 20, 2013 at 3:37 AM, Simone Bordet <sbordet@intalio.com> wrote:

> Hi,
>
> On Sun, May 19, 2013 at 6:30 AM, Takeshi Yoshino <tyoshino@google.com>
> wrote:
> > I got one comment about this.
> >
> > He/she suggested that we finish the WebSocket dedicated multiplexing
> quickly
> > even by simplifying the spec much e.g. cutting some functionality in it
> that
> > would take much time to get consensus, and when HTTP/2.0 is matured,
> switch
> > to that.
>
> My impression is that I have not seen much request for WebSocket
> multiplexing.
> Given the complexities related to defining and implementing ws mux
> correctly, and considering that HTTP/2.0 comes with mux built-in,
> building on the SPDY experience, my preference would be to
> deprioritize ws mux in favour of HTTP/2.0 integration.
> It is true that there may be systems that may rely on a pure-ws
> communication (and therefore do not need a full HTTP/2.0
> implementation) and as such would benefit from ws mux, but my guess is
> that those system will be a minority or can get by without ws mux. I
> may be biased on this view though, so comments welcome.
>

Yeah, more comments from those who are using / interested in pure-ws are
appreciated.

--089e0162830c6a562004dd1e6597
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div><div>Thanks for comment, Simone.</div></div><div><br>=
</div>On Mon, May 20, 2013 at 3:37 AM, Simone Bordet <span dir=3D"ltr">&lt;=
<a href=3D"mailto:sbordet@intalio.com" target=3D"_blank">sbordet@intalio.co=
m</a>&gt;</span> wrote:<br>

<div class=3D"gmail_extra"><div class=3D"gmail_quote"><blockquote class=3D"=
gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;border=
-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">Hi,<=
br>

<div><div class=3D"h5"><br>
On Sun, May 19, 2013 at 6:30 AM, Takeshi Yoshino &lt;<a href=3D"mailto:tyos=
hino@google.com">tyoshino@google.com</a>&gt; wrote:<br>
&gt; I got one comment about this.<br>
&gt;<br>
&gt; He/she suggested that we finish the WebSocket dedicated multiplexing q=
uickly<br>
&gt; even by simplifying the spec much e.g. cutting some functionality in i=
t that<br>
&gt; would take much time to get consensus, and when HTTP/2.0 is matured, s=
witch<br>
&gt; to that.<br>
<br>
</div></div>My impression is that I have not seen much request for WebSocke=
t multiplexing.<br>
Given the complexities related to defining and implementing ws mux<br>
correctly, and considering that HTTP/2.0 comes with mux built-in,<br>
building on the SPDY experience, my preference would be to<br>
deprioritize ws mux in favour of HTTP/2.0 integration.<br>
It is true that there may be systems that may rely on a pure-ws<br>
communication (and therefore do not need a full HTTP/2.0<br>
implementation) and as such would benefit from ws mux, but my guess is<br>
that those system will be a minority or can get by without ws mux. I<br>
may be biased on this view though, so comments welcome.<br></blockquote><di=
v style><br></div><div style>Yeah, more comments from those who are using /=
 interested in pure-ws are appreciated.</div><div style>=A0<br></div></div>

</div></div>

--089e0162830c6a562004dd1e6597--

From bruce@callenish.com  Tue May 21 16:03:12 2013
Return-Path: <bruce@callenish.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 88A2011E80EC for <hybi@ietfa.amsl.com>; Tue, 21 May 2013 16:03:12 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id h0iDHAAQCV2K for <hybi@ietfa.amsl.com>; Tue, 21 May 2013 16:03:05 -0700 (PDT)
Received: from biz82.inmotionhosting.com (biz82.inmotionhosting.com [66.117.4.147]) by ietfa.amsl.com (Postfix) with ESMTP id CECFD11E80E4 for <hybi@ietf.org>; Tue, 21 May 2013 16:03:03 -0700 (PDT)
Received: from [96.50.69.82] (port=63091 helo=[192.168.145.118]) by biz82.inmotionhosting.com with esmtpsa (TLSv1:DHE-RSA-AES256-SHA:256) (Exim 4.80) (envelope-from <bruce@callenish.com>) id 1Ueva8-0001Ul-L7; Tue, 21 May 2013 16:03:01 -0700
Message-ID: <519BFD21.70300@callenish.com>
Date: Tue, 21 May 2013 16:02:57 -0700
From: Bruce Atherton <bruce@callenish.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: Simone Bordet <sbordet@intalio.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <CAFWmRJ0naLw=Om70ag3p2xJ8mYvgNHLL6Z_DtpiOjVWZ3iS5Dg@mail.gmail.com>
In-Reply-To: <CAFWmRJ0naLw=Om70ag3p2xJ8mYvgNHLL6Z_DtpiOjVWZ3iS5Dg@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-OutGoing-Spam-Status: No, score=-2.9
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - biz82.inmotionhosting.com
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - callenish.com
X-Get-Message-Sender-Via: biz82.inmotionhosting.com: authenticated_id: bruce+callenish.com/only user confirmed/virtual account not confirmed
X-Source: 
X-Source-Args: 
X-Source-Dir: 
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 21 May 2013 23:03:12 -0000

I agree. During the planning stages of the protocol there was a great 
deal of interest expressed in stand-alone WebSocket servers that were 
independent of any HTTP server. That has not been the real world 
deployment behaviour, at least in my experience. So long as HTTP/2.0 
doesn't introduce so much added complexity that it becomes too costly to 
include a server for it when you are deploying WebSockets, I don't see 
much benefit to a WebSockets-only mux definition any more, much though I 
appreciate the effort that Takeshi and John have put into it.

On 19/05/2013 11:37 AM, Simone Bordet wrote:
> My impression is that I have not seen much request for WebSocket multiplexing.
> Given the complexities related to defining and implementing ws mux
> correctly, and considering that HTTP/2.0 comes with mux built-in,
> building on the SPDY experience, my preference would be to
> deprioritize ws mux in favour of HTTP/2.0 integration.
> It is true that there may be systems that may rely on a pure-ws
> communication (and therefore do not need a full HTTP/2.0
> implementation) and as such would benefit from ws mux, but my guess is
> that those system will be a minority or can get by without ws mux. I
> may be biased on this view though, so comments welcome.
>
>


From tobias.oberstein@tavendo.de  Wed May 22 03:01:30 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 227F221F919D for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 03:01:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pLXJx2ddmeao for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 03:01:25 -0700 (PDT)
Received: from EXHUB020-3.exch020.serverdata.net (exhub020-3.exch020.serverdata.net [206.225.164.30]) by ietfa.amsl.com (Postfix) with ESMTP id F336121F8F33 for <hybi@ietf.org>; Wed, 22 May 2013 03:01:18 -0700 (PDT)
Received: from EXVMBX020-12.exch020.serverdata.net ([169.254.3.33]) by EXHUB020-3.exch020.serverdata.net ([206.225.164.30]) with mapi; Wed, 22 May 2013 03:01:17 -0700
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
To: Takeshi Yoshino <tyoshino@google.com>, "hybi@ietf.org" <hybi@ietf.org>
Date: Wed, 22 May 2013 03:01:13 -0700
Thread-Topic: [hybi] Call for interest: multiplexing dedicated for WebSocket
Thread-Index: Ac5RYcSRQA6rDIMOQDKLCNyQXVyfwAFcN/pQ
Message-ID: <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com>
In-Reply-To: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: de-DE, en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 10:01:30 -0000

Am 15.05.2013 13:45, schrieb Takeshi Yoshino:> Hi all,
>=20
> (Note: This is not WG's official process)
>=20
> I'd like to survey how many people are interested in standardization and=
=20
> implementation of multiplexing dedicated for WebSocket again at this poin=
t.

+1 for having WS mux and not depend on HTTP 2.0 / SPDY.

>=20
> * Multiplexing is still important for WebSocket
>=20
> HyBi WG has been working on the WebSocket multiplexing spec=20
> (http://tools.ietf.org/html/draft-ietf-hybi-websocket-multiplexing-10).=20
> Multiplexing functionality is really important to make WebSocket=20
> scalable enough to be used for large scale production services. It's=20
> still true.

Could you eloborate a bit on this? I'd like to better understand "the probl=
em". As far as I can see, the current RFC text contains the following motiv=
ation for multiplexing:

"""
The WebSocket Protocol [RFC6455] requires a new transport connection
for every WebSocket connection.  This presents a scalability problem
when many clients connect to the same server, and is made worse by
having multiple clients running in different tabs of the same user
agent.
"""

What exactly is the scalability problem? Is it: scalability of a single ser=
ver box is limited by the number of concurrently active TCP connections the=
 server can handle? I.e. one is be able to handle 200k TCPs on a single box=
, but not 20Mio ..

If so, is the thinking that a) one needs some kind of "network concentrator=
" front-end servers that aggregate the many TCPs onto a few, and b) for doi=
ng that kind of concentration, WS multiplexing is one option?

However, concentration could also be done at the application level, e.g. st=
ateless frontend server speaking WS with some app defined WS subprotocol to=
 clients, and native database protocol to backend servers, multiplexing req=
uests contained in the WS-based app protocol to requests on the few databas=
e protocol connections to backend servers.

That can also solve the scalability problem without requiring multiplexing =
at the WS level. So why do we need it at the WS level?

Sure, multiplexing at the WS level would allow to create generic concentrat=
ors, but a generic concentrator could not map incoming connections to backe=
nd boxes based on app data like UserID etc for sharding etc.

Another aspect is client-server transport efficiency.

Most efficient: exactly 1 TCP connection between given client device and se=
rver for any communication between the two.

The use of only 1 TCP connection between a given client device and a given =
server allows mechanisms like TCP slow start etc work efficiently.

A browser with multiple tabs opened to the same origin might be a common si=
tuation which could be handled by multiplexing in the networking code insid=
e the browser.

Yet another option with browser could be a shared web worker opening 1 WS t=
o the server, and have all tabs communicate only via that web worker.

However, when running _different_ apps on the same device connecting to the=
 same origin, the multiplexing needed to happen cross-app, and hence the mu=
ltiplexing needed to happen in kernel space or in a user space library used=
 by all apps on the client device platform.

=3D=3D

So my current thinking is:

- the "1 TCP for efficiency" problem can not be solved at the app layer, bu=
t needs multiplexing at WS (or within SPDY), which then solves the problem =
- but only partially (not across different apps, at least not without kerne=
l-space implementation). The problem can be solved however with browsers an=
d shared web workers also.

- the scalability problem can be solved both at the app layer or the WS/SPD=
Y layer.

>=20
> * Rapid adoption of HTTP/2.0
>=20
> While our work is delayed and delayed again, httpbis folks started=20
> seriously working on next generation HTTP. Now it's actively=20
> standardized as HTTP/2.0 based on Google's SPDY experiments and input=20
> from httpbis members. Gradually but several web service providers=20
> already started adopting HTTP/2.0.
>=20
> * WebSocket/SPDY experiment
>=20
> Chrome WebSocket team has been working on both WebSocket/SPDY and=20
> dedicated multiplexing in parallel. You can find the WebSocket/SPDY=20
> proposal incorporated into Roberto's unofficial draft (section 4) at=20
> https://github.com/grmocg/SPDY-Specification/blob/gh-pages/draft-mbelshe-=
spdy-00.txt.=20
> Its framing is not so efficient as the dedicated one, but can coexist=20
> with HTTP traffic in one TCP connection and developers just need to=20
> understand and implement HTTP/2.0 framing.

"""
4.2.1. Frame mapping

   One WebSocket frame is mapped into one proceeding SPDY HEADERS frame
   and following plural SPDY data frames. The SPDY HEADERS frame must
   contain WebSocket frame fields and following SPDY data frames contain
   payload data. These data frames are free to be reframed.

   This plan doesn't care about overheads.
"""

This seems somehow straightforward, but introduces inefficiencies: the over=
head per WS message is significantly increased and an additional framing la=
yer that allows reframing costs CPU cycles.

Btw: assuming SPDY runs over TLS exclusively, is there any need for WS fram=
e masking anymore?

>=20
> * Concentrate on the best one
>=20
> Developing and maintaining two different multiplexing code will be=20
> burden for vendors. It seems that we should stop and think which=20
> approach we should focus on.

Yes, but WS over SPDY is not ready today. WS is widely implemented and work=
s today.

As an app developer, the only thing I care about is having a message-based,=
 reliable, ordered, full-duplex communication with browsers that is both ef=
ficient and widely implemented.

Personally I don't care a lot about improving HTTP, since with single-page =
web apps (and manifests), app assets would be in browser cache anyway. So I=
 care about pure WS solutions.

Stacking framing onto framing layers or WS on SPDY in general does not seem=
 to improve efficiency or implementation availability in browsers. It's a s=
tep backward from my POV.

So yes, I do care about WS multiplexing.

>=20
> Thanks


From SRS0+1ae53a578e5209af=PH=noemax.com=arman@noemax.com  Wed May 22 06:40:38 2013
Return-Path: <SRS0+1ae53a578e5209af=PH=noemax.com=arman@noemax.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 310D421F8FE8 for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 06:40:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.598
X-Spam-Level: 
X-Spam-Status: No, score=-2.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id E+OwxgHVOpeu for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 06:40:33 -0700 (PDT)
Received: from mail.noemax.com (mail.noemax.com [74.208.113.37]) by ietfa.amsl.com (Postfix) with ESMTP id 9EB9421F8529 for <hybi@ietf.org>; Wed, 22 May 2013 06:40:33 -0700 (PDT)
DKIM-Signature: a=rsa-sha1; t=1369229371; x=1369834171; s=m2048; d=noemax.com; c=relaxed/relaxed; v=1; bh=doPMA3VSQOM4Rvb7xdMGH9zJ8qg=; h=From:Subject:Date:Message-ID:To:MIME-Version:Content-Type:In-Reply-To:References; b=A5zBuz+EHr1MPwrakeZlhmDuXhuIYU+rFdQhMh1GTJcyFtKfBbawXFYxA1K06F4DWEqDF2su1SGMPGwc09+cFmKp7lShLk04d2lFvGncuZAgiP5WiSBiEOn+9ccET4UNCBUqJnt2A0rGjO/sHdHQf0Vd94U6348s5uYmmgfa5iK+6AknfGPpVb7g4oitqH0tUs6TlUf5U0og8VtwZqSf5atGdPtHeDkxaqSH4ZWfoCnDp+VJGpGXG6MARpw3EPKkruSJwixkRlwMo9ViZAhvjBi6brjU/zpFshtH4hFdd+MLSDfYP/ML2ZUuqkRBPqdkRO7skplPXbAwl0qLCFkHsw==
Received: from mail.noemax.com by mail.noemax.com (Noemax Mail Server) with ASMTP (SSL) id 201305221629301123; Wed, 22 May 2013 16:29:30 +0300
From: "Arman Djusupov" <arman@noemax.com>
To: "'Takeshi Yoshino'" <tyoshino@google.com>, <hybi@ietf.org>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com>	<CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com>
In-Reply-To: <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com>
Date: Wed, 22 May 2013 16:29:33 +0300
Message-ID: <007501ce56f0$67f74080$37e5c180$@noemax.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_NextPart_000_0076_01CE5709.8D44EDB0"
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQGJbPE/yAcYUaa2Hw/8qx+KPIY2GAJfnsrjAhgTLgGZd0jZEA==
Content-Language: en-us
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 13:52:13 -0000

This is a multipart message in MIME format.

------=_NextPart_000_0076_01CE5709.8D44EDB0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit

Sorry, just realized that my message was sent to Takeshi privately by
mistake a few days ago.. (wrong button). It was actually intended for the
list so I am repeating it publicly now:

 

 

Obviously continuing to develop a hybi-mux spec for another year or two
doesn't make sense. Taking into account that new specifications are coming
and that HTTP2.0 will probably provide multiplexing for both WebSocket and
HTTP traffic, at  this stage it would only be worth investing development
time in a "simple and ready to go" hybi-mux specification. Maybe if we
simplify the spec to rock bottom simplicity, even sacrificing some features
so that we can conclude it very fast, then maybe vendors might pick it up.
This way we will give to WebSocket its own mux for now, while the HTTPbis
draft might take much longer to form into the final specification. In any
case, hybi-mux must be reviewed and finalized in a very short time, plus it
should be simpler and easier to implement than both SDPY and HTTPbis based
multiplexing.

 

With best regards,

Arman

 

From: hybi-bounces@ietf.org [mailto:hybi-bounces@ietf.org] On Behalf Of
Takeshi Yoshino
Sent: Sunday, May 19, 2013 7:31 AM
To: hybi@ietf.org
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket

 

I got one comment about this.

 

He/she suggested that we finish the WebSocket dedicated multiplexing quickly
even by simplifying the spec much e.g. cutting some functionality in it that
would take much time to get consensus, and when HTTP/2.0 is matured, switch
to that.

 


------=_NextPart_000_0076_01CE5709.8D44EDB0
Content-Type: text/html;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" =
xmlns:o=3D"urn:schemas-microsoft-com:office:office" =
xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" =
xmlns=3D"http://www.w3.org/TR/REC-html40"><head><META =
HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii"><meta name=3DGenerator content=3D"Microsoft Word 14 =
(filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue =
vlink=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Sorry, just realized that my message was sent to Takeshi privately by =
mistake a few days ago.. (wrong button). It was actually intended for =
the list so I am repeating it publicly now:<o:p></o:p></span></p><p =
class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Obviously continuing to develop a hybi-mux spec for another year or =
two doesn&#8217;t make sense. Taking into account that new =
specifications are coming and that HTTP2.0 will probably provide =
multiplexing for both WebSocket and HTTP traffic, at&nbsp; this stage it =
would only be worth investing development time in a &#8220;simple and =
ready to go&#8221; hybi-mux specification. Maybe if we simplify the spec =
to rock bottom simplicity, even sacrificing some features so that we can =
conclude it very fast, then maybe vendors might pick it up. This way we =
will give to WebSocket its own mux for now, while the HTTPbis draft =
might take much longer to form into the final specification. In any =
case, hybi-mux must be reviewed and finalized in a very short time, plus =
it should be simpler and easier to implement than both SDPY and HTTPbis =
based multiplexing.<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>With best regards,<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Arman<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><b><span =
style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span>=
</b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'> =
hybi-bounces@ietf.org [mailto:hybi-bounces@ietf.org] <b>On Behalf Of =
</b>Takeshi Yoshino<br><b>Sent:</b> Sunday, May 19, 2013 7:31 =
AM<br><b>To:</b> hybi@ietf.org<br><b>Subject:</b> Re: [hybi] Call for =
interest: multiplexing dedicated for WebSocket<o:p></o:p></span></p><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p><div><p class=3DMsoNormal>I got =
one comment about this.<o:p></o:p></p><div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p =
class=3DMsoNormal>He/she suggested that we finish the WebSocket =
dedicated multiplexing quickly even by simplifying the spec much e.g. =
cutting some functionality in it that would take much time to get =
consensus, and when HTTP/2.0 is matured, switch to =
that.<o:p></o:p></p><div><p =
class=3DMsoNormal>&nbsp;<o:p></o:p></p></div></div></div></div></body></h=
tml>
------=_NextPart_000_0076_01CE5709.8D44EDB0--


From salvatore.loreto@ericsson.com  Wed May 22 07:31:06 2013
Return-Path: <salvatore.loreto@ericsson.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 69D7621F90CD for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 07:31:06 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -106.248
X-Spam-Level: 
X-Spam-Status: No, score=-106.248 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HELO_EQ_SE=0.35, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4, 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 qmLaHE9-d7uz for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 07:31:00 -0700 (PDT)
Received: from mailgw2.ericsson.se (mailgw2.ericsson.se [193.180.251.37]) by ietfa.amsl.com (Postfix) with ESMTP id 190C021F9080 for <hybi@ietf.org>; Wed, 22 May 2013 07:30:59 -0700 (PDT)
X-AuditID: c1b4fb25-b7efb6d000007c26-fc-519cd6a2114c
Received: from esessmw0237.eemea.ericsson.se (Unknown_Domain [153.88.253.125]) by mailgw2.ericsson.se (Symantec Mail Security) with SMTP id BD.68.31782.2A6DC915; Wed, 22 May 2013 16:30:59 +0200 (CEST)
Received: from mail.lmf.ericsson.se (153.88.115.8) by esessmw0237.eemea.ericsson.se (153.88.115.91) with Microsoft SMTP Server id 8.3.279.1; Wed, 22 May 2013 16:30:58 +0200
Received: from nomadiclab.lmf.ericsson.se (nomadiclab.lmf.ericsson.se [131.160.33.3])	by mail.lmf.ericsson.se (Postfix) with ESMTP id 8F21E240A	for <hybi@ietf.org>; Wed, 22 May 2013 17:30:58 +0300 (EEST)
Received: from nomadiclab.lmf.ericsson.se (localhost [127.0.0.1])	by nomadiclab.lmf.ericsson.se (Postfix) with ESMTP id 3C738550E2	for <hybi@ietf.org>; Wed, 22 May 2013 17:30:57 +0300 (EEST)
Received: from Salvatore-Loretos-MacBook-Pro.local (localhost [127.0.0.1])	by nomadiclab.lmf.ericsson.se (Postfix) with ESMTP id E9AD855091	for <hybi@ietf.org>; Wed, 22 May 2013 17:30:56 +0300 (EEST)
Message-ID: <519CD6A1.7080708@ericsson.com>
Date: Wed, 22 May 2013 17:30:57 +0300
From: Salvatore Loreto <salvatore.loreto@ericsson.com>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: hybi@ietf.org
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com>	<CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com>
In-Reply-To: <007501ce56f0$67f74080$37e5c180$@noemax.com>
Content-Type: multipart/alternative; boundary="------------090503060001090806000308"
X-Virus-Scanned: ClamAV using ClamSMTP
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFtrJLMWRmVeSWpSXmKPExsUyM+Jvre7ia3MCDWavZ7Z4/3IbkwOjx5Il P5kCGKO4bFJSczLLUov07RK4Mma84C144FJx9MU+5gbG22ZdjBwcEgImEh+XRXQxcgKZYhIX 7q1n62Lk4hASOMUosbpzHTuEs4FRon1RI1TmPKPE9fbtzBDOIUaJju17WSGc04wSJ2ZMZwMZ xiugLfHv7iVGEJtFQFVi1q5VzCA2m4CZxPOHW8BsUYFkiclTV7BA1AtKnJz5BMwWAbK7t64B qxEW8JZoajnPArGgnUlix6ZbYEM5BSwlVvdcYQexmQXCJHaunc0I8YWaxNVzm8CahQS0JHrP djJNYBSehWTHLCQtELatxIU516Hi8hLb385hhrB1JS78n4IivoCRbRUje25iZk56udEmRmDw H9zyW3UH451zIocYpTlYlMR5e7WnBgoJpCeWpGanphakFsUXleakFh9iZOLglGpg1Hv/lWVS io/Csz1KaWbLTY1LlB5fOT5llkblVu+YzyHTHtkURmkl6JWtOn8tKX6d04Ey/2/zBARvbP0/ PcFVrWrSpTDuUL76ryIHKg++qr/n099xYlFs1y1fdqNl+a/LXixOTXHMPO5z3dMuKvq+cmvm 9wj+W8bK5nYmK2173GfJX3PgXLVCiaU4I9FQi7moOBEAggfAs0wCAAA=
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 14:31:06 -0000

--------------090503060001090806000308
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
Content-Transfer-Encoding: 7bit

<as individual>
I tend to agree with the fact that if we want standardize mux in 
WebSocket we should speed up simplifying the spec.

/Salvatore

On 5/22/13 4:29 PM, Arman Djusupov wrote:
>
> Sorry, just realized that my message was sent to Takeshi privately by 
> mistake a few days ago.. (wrong button). It was actually intended for 
> the list so I am repeating it publicly now:
>
> Obviously continuing to develop a hybi-mux spec for another year or 
> two doesn't make sense. Taking into account that new specifications 
> are coming and that HTTP2.0 will probably provide multiplexing for 
> both WebSocket and HTTP traffic, at  this stage it would only be worth 
> investing development time in a "simple and ready to go" hybi-mux 
> specification. Maybe if we simplify the spec to rock bottom 
> simplicity, even sacrificing some features so that we can conclude it 
> very fast, then maybe vendors might pick it up. This way we will give 
> to WebSocket its own mux for now, while the HTTPbis draft might take 
> much longer to form into the final specification. In any case, 
> hybi-mux must be reviewed and finalized in a very short time, plus it 
> should be simpler and easier to implement than both SDPY and HTTPbis 
> based multiplexing.
>
> With best regards,
>
> Arman
>
> *From:*hybi-bounces@ietf.org [mailto:hybi-bounces@ietf.org] *On Behalf 
> Of *Takeshi Yoshino
> *Sent:* Sunday, May 19, 2013 7:31 AM
> *To:* hybi@ietf.org
> *Subject:* Re: [hybi] Call for interest: multiplexing dedicated for 
> WebSocket
>
> I got one comment about this.
>
> He/she suggested that we finish the WebSocket dedicated multiplexing 
> quickly even by simplifying the spec much e.g. cutting some 
> functionality in it that would take much time to get consensus, and 
> when HTTP/2.0 is matured, switch to that.
>
>
>
> _______________________________________________
> hybi mailing list
> hybi@ietf.org
> https://www.ietf.org/mailman/listinfo/hybi



--------------090503060001090806000308
Content-Type: text/html; charset="ISO-8859-1"
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">&lt;as individual&gt;<br>
      I tend to agree with the fact that if we want standardize mux in
      WebSocket we should speed up simplifying the spec.<br>
      <br>
      /Salvatore<br>
      <br>
      On 5/22/13 4:29 PM, Arman Djusupov wrote:<br>
    </div>
    <blockquote cite="mid:007501ce56f0$67f74080$37e5c180$@noemax.com"
      type="cite">
      <meta http-equiv="Content-Type" content="text/html;
        charset=ISO-8859-1">
      <meta name="Generator" content="Microsoft Word 14 (filtered
        medium)">
      <style><!--
/* Font Definitions */
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
      <div class="WordSection1">
        <p class="MsoNormal"><span
style="font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D">Sorry,
            just realized that my message was sent to Takeshi privately
            by mistake a few days ago.. (wrong button). It was actually
            intended for the list so I am repeating it publicly now:<o:p></o:p></span></p>
        <p class="MsoNormal"><span
style="font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span></p>
        <p class="MsoNormal"><span
style="font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span></p>
        <p class="MsoNormal"><span
style="font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D">Obviously
            continuing to develop a hybi-mux spec for another year or
            two doesn&#8217;t make sense. Taking into account that new
            specifications are coming and that HTTP2.0 will probably
            provide multiplexing for both WebSocket and HTTP traffic,
            at&nbsp; this stage it would only be worth investing development
            time in a &#8220;simple and ready to go&#8221; hybi-mux specification.
            Maybe if we simplify the spec to rock bottom simplicity,
            even sacrificing some features so that we can conclude it
            very fast, then maybe vendors might pick it up. This way we
            will give to WebSocket its own mux for now, while the
            HTTPbis draft might take much longer to form into the final
            specification. In any case, hybi-mux must be reviewed and
            finalized in a very short time, plus it should be simpler
            and easier to implement than both SDPY and HTTPbis based
            multiplexing.<o:p></o:p></span></p>
        <p class="MsoNormal"><span
style="font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span></p>
        <p class="MsoNormal"><span
style="font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D">With
            best regards,<o:p></o:p></span></p>
        <p class="MsoNormal"><span
style="font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D">Arman<o:p></o:p></span></p>
        <p class="MsoNormal"><span
style="font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;sans-serif&quot;;color:#1F497D"><o:p>&nbsp;</o:p></span></p>
        <p class="MsoNormal"><b><span
style="font-size:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span
style="font-size:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;">
            <a class="moz-txt-link-abbreviated" href="mailto:hybi-bounces@ietf.org">hybi-bounces@ietf.org</a> [<a class="moz-txt-link-freetext" href="mailto:hybi-bounces@ietf.org">mailto:hybi-bounces@ietf.org</a>] <b>On
              Behalf Of </b>Takeshi Yoshino<br>
            <b>Sent:</b> Sunday, May 19, 2013 7:31 AM<br>
            <b>To:</b> <a class="moz-txt-link-abbreviated" href="mailto:hybi@ietf.org">hybi@ietf.org</a><br>
            <b>Subject:</b> Re: [hybi] Call for interest: multiplexing
            dedicated for WebSocket<o:p></o:p></span></p>
        <p class="MsoNormal"><o:p>&nbsp;</o:p></p>
        <div>
          <p class="MsoNormal">I got one comment about this.<o:p></o:p></p>
          <div>
            <p class="MsoNormal"><o:p>&nbsp;</o:p></p>
          </div>
          <div>
            <p class="MsoNormal">He/she suggested that we finish the
              WebSocket dedicated multiplexing quickly even by
              simplifying the spec much e.g. cutting some functionality
              in it that would take much time to get consensus, and when
              HTTP/2.0 is matured, switch to that.<o:p></o:p></p>
            <div>
              <p class="MsoNormal">&nbsp;<o:p></o:p></p>
            </div>
          </div>
        </div>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
hybi mailing list
<a class="moz-txt-link-abbreviated" href="mailto:hybi@ietf.org">hybi@ietf.org</a>
<a class="moz-txt-link-freetext" href="https://www.ietf.org/mailman/listinfo/hybi">https://www.ietf.org/mailman/listinfo/hybi</a>
</pre>
    </blockquote>
    <br>
    <br>
  </body>
</html>

--------------090503060001090806000308--

From tobias.oberstein@tavendo.de  Wed May 22 08:13:53 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4943721F8E87 for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 08:13:53 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1v21J1rNnnVD for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 08:13:48 -0700 (PDT)
Received: from EXHUB020-4.exch020.serverdata.net (exhub020-4.exch020.serverdata.net [206.225.164.31]) by ietfa.amsl.com (Postfix) with ESMTP id BFF7B21F8607 for <hybi@ietf.org>; Wed, 22 May 2013 08:13:45 -0700 (PDT)
Received: from [10.157.82.16] (89.204.138.16) by smtpx20.serverdata.net (206.225.164.36) with Microsoft SMTP Server (TLS) id 8.3.264.0; Wed, 22 May 2013 08:13:44 -0700
Message-ID: <519CE075.4000106@tavendo.de>
Date: Wed, 22 May 2013 17:12:53 +0200
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: Salvatore Loreto <salvatore.loreto@ericsson.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com>
In-Reply-To: <519CD6A1.7080708@ericsson.com>
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 15:13:53 -0000

Am 22.05.2013 16:30, schrieb Salvatore Loreto:
> <as individual>
> I tend to agree with the fact that if we want standardize mux in
> WebSocket we should speed up simplifying the spec.

+1 for the simplest possible mux that does the core job.

However, I currently think that having a per-logical channel flow 
control is required to make it useful (cannot be dropped for the sake of 
simplification). I might be wrong.

So the practical question: in what areas could we simplify the current 
mux spec?

>> much longer to form into the final specification. In any case,
>> hybi-mux must be reviewed and finalized in a very short time, plus it
>> should be simpler and easier to implement than both SDPY and HTTPbis
>> based multiplexing.

Yep, I agree.

>>
>> With best regards,
>>
>> Arman
>>
>> *From:*hybi-bounces@ietf.org [mailto:hybi-bounces@ietf.org] *On Behalf
>> Of *Takeshi Yoshino
>> *Sent:* Sunday, May 19, 2013 7:31 AM
>> *To:* hybi@ietf.org
>> *Subject:* Re: [hybi] Call for interest: multiplexing dedicated for
>> WebSocket
>>
>> I got one comment about this.
>>
>> He/she suggested that we finish the WebSocket dedicated multiplexing
>> quickly even by simplifying the spec much e.g. cutting some
>> functionality in it that would take much time to get consensus, and
>> when HTTP/2.0 is matured, switch to that.
>>
>>
>>
>> _______________________________________________
>> hybi mailing list
>> hybi@ietf.org
>> https://www.ietf.org/mailman/listinfo/hybi
>
>


From salvatore.loreto@ericsson.com  Wed May 22 08:23:52 2013
Return-Path: <salvatore.loreto@ericsson.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D4D4221F8FE8 for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 08:23:52 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -106.249
X-Spam-Level: 
X-Spam-Status: No, score=-106.249 tagged_above=-999 required=5 tests=[AWL=0.000, BAYES_00=-2.599, HELO_EQ_SE=0.35, RCVD_IN_DNSWL_MED=-4, 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 F-PN3Lyy8qbo for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 08:23:47 -0700 (PDT)
Received: from mailgw1.ericsson.se (mailgw1.ericsson.se [193.180.251.45]) by ietfa.amsl.com (Postfix) with ESMTP id F278321F8FA5 for <hybi@ietf.org>; Wed, 22 May 2013 08:23:46 -0700 (PDT)
X-AuditID: c1b4fb2d-b7fe36d000007102-cf-519ce301f965
Received: from esessmw0247.eemea.ericsson.se (Unknown_Domain [153.88.253.125]) by mailgw1.ericsson.se (Symantec Mail Security) with SMTP id C9.79.28930.103EC915; Wed, 22 May 2013 17:23:45 +0200 (CEST)
Received: from mail.lmf.ericsson.se (153.88.115.8) by esessmw0247.eemea.ericsson.se (153.88.115.94) with Microsoft SMTP Server id 8.3.279.1; Wed, 22 May 2013 17:23:45 +0200
Received: from nomadiclab.lmf.ericsson.se (nomadiclab.lmf.ericsson.se [131.160.33.3])	by mail.lmf.ericsson.se (Postfix) with ESMTP id 50B03240A; Wed, 22 May 2013 18:23:45 +0300 (EEST)
Received: from nomadiclab.lmf.ericsson.se (localhost [127.0.0.1])	by nomadiclab.lmf.ericsson.se (Postfix) with ESMTP id EEAA9550E4; Wed, 22 May 2013 18:23:43 +0300 (EEST)
Received: from Salvatore-Loretos-MacBook-Pro.local (localhost [127.0.0.1])	by nomadiclab.lmf.ericsson.se (Postfix) with ESMTP id 949ED53704; Wed, 22 May 2013 18:23:43 +0300 (EEST)
Message-ID: <519CE300.40107@ericsson.com>
Date: Wed, 22 May 2013 18:23:44 +0300
From: Salvatore Loreto <salvatore.loreto@ericsson.com>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de>
In-Reply-To: <519CE075.4000106@tavendo.de>
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: ClamAV using ClamSMTP
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFtrOLMWRmVeSWpSXmKPExsUyM+JvrS7j4zmBBi2rTC3ev9zGZDFvUQ+L A5PHkiU/mTyuTJ3AEsAUxWWTkpqTWZZapG+XwJWx9NkPpoJPfBX/2ttYGhj/c3cxcnJICJhI bDnwkhHCFpO4cG89WxcjF4eQwClGiV+P1zNCOBsYJfYt62GCcHYzShx58RDKWcco8Wrqfqie 7YwSjX2L2EGG8QpoSqw8+YwJxGYRUJVoufIOLM4mYCbx/OEWZhBbVCBZYvLUFSwQ9YISJ2c+ AbNFBIwkpu9dBGYzCyhLXD0GUSMs4C3R1HKeBWLZUSaJBV/3gQ3lFNCSuLRwPpDNAdRgL/Fg axlEr7zE9rdzmCGeU5O4em4TmC0EVN57tpNpAqPoLCSrZyF0z0LSvYCReRUje25iZk56ueEm RmDgH9zyW3cH46lzIocYpTlYlMR5e7WnBgoJpCeWpGanphakFsUXleakFh9iZOLglGpgdLT8 3Rv/4K/U4rLVqbKnDZ85tbWEhjOE/a8/mLHh3KmjUzd5nVH1n9cT6Ru8s+Dy0/DD3R8edv2+ cYJlpqDf3BnOL8r3L690DX72Y+vMF7POCGw+XGI4uUP+Hs/vf4umNBjUN1csvz3j2tMDLNqc s/9nvHZ9bH5xbYr73qa09AhRZqml+XGcYUosxRmJhlrMRcWJANC33YdKAgAA
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 15:23:53 -0000

<as individual>

my concerns are just about the importance of WebSocket mux once HTTP/2.0 
will be finalized
also considering the fact that the HTTP/2.0 standardization is 
proceeding quite fast and it is supposed
to be finalized in one year and something

/Salvatore

On 5/22/13 6:12 PM, Tobias Oberstein wrote:
> Am 22.05.2013 16:30, schrieb Salvatore Loreto:
>> <as individual>
>> I tend to agree with the fact that if we want standardize mux in
>> WebSocket we should speed up simplifying the spec.
>
> +1 for the simplest possible mux that does the core job.
>
> However, I currently think that having a per-logical channel flow 
> control is required to make it useful (cannot be dropped for the sake 
> of simplification). I might be wrong.
>
> So the practical question: in what areas could we simplify the current 
> mux spec?
>
>>> much longer to form into the final specification. In any case,
>>> hybi-mux must be reviewed and finalized in a very short time, plus it
>>> should be simpler and easier to implement than both SDPY and HTTPbis
>>> based multiplexing.
>
> Yep, I agree.
>
>>>
>>> With best regards,
>>>
>>> Arman
>>>
>>> *From:*hybi-bounces@ietf.org [mailto:hybi-bounces@ietf.org] *On Behalf
>>> Of *Takeshi Yoshino
>>> *Sent:* Sunday, May 19, 2013 7:31 AM
>>> *To:* hybi@ietf.org
>>> *Subject:* Re: [hybi] Call for interest: multiplexing dedicated for
>>> WebSocket
>>>
>>> I got one comment about this.
>>>
>>> He/she suggested that we finish the WebSocket dedicated multiplexing
>>> quickly even by simplifying the spec much e.g. cutting some
>>> functionality in it that would take much time to get consensus, and
>>> when HTTP/2.0 is matured, switch to that.
>>>
>>>
>>>
>>> _______________________________________________
>>> hybi mailing list
>>> hybi@ietf.org
>>> https://www.ietf.org/mailman/listinfo/hybi
>>
>>


From tobias.oberstein@tavendo.de  Wed May 22 08:32:05 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1ACD221F9051 for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 08:32:05 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pGztLh-JMgm0 for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 08:31:59 -0700 (PDT)
Received: from EXHUB020-3.exch020.serverdata.net (exhub020-3.exch020.serverdata.net [206.225.164.30]) by ietfa.amsl.com (Postfix) with ESMTP id E640E21F92E7 for <hybi@ietf.org>; Wed, 22 May 2013 08:31:58 -0700 (PDT)
Received: from [10.157.82.16] (89.204.138.16) by smtpx20.serverdata.net (206.225.164.35) with Microsoft SMTP Server (TLS) id 8.3.264.0; Wed, 22 May 2013 08:31:57 -0700
Message-ID: <519CE4EC.30902@tavendo.de>
Date: Wed, 22 May 2013 17:31:56 +0200
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: Salvatore Loreto <salvatore.loreto@ericsson.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <519CE300.40107@ericsson.com>
In-Reply-To: <519CE300.40107@ericsson.com>
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 15:32:05 -0000

Am 22.05.2013 17:23, schrieb Salvatore Loreto:
> <as individual>
>
> my concerns are just about the importance of WebSocket mux once HTTP/2.0
> will be finalized
> also considering the fact that the HTTP/2.0 standardization is
> proceeding quite fast and it is supposed
> to be finalized in one year and something

I would prefer a pure WS protocol (including mux and compression) _not_ 
layered on top of HTTP/2.0 as a long-term, widely implemented solution.

This is driven by our primary use case: have single-page web apps (with 
cached assets) that talk to backends exclusively over WS.

Layering WS on top of HTTP/2.0 does add additional overhead without 
gain, and for retrieving HTML etc assets that are persistently cached in 
browser (Web app manifests etc), HTTP/2.0 does not seem to add a lot 
over HTTP/1.1 also.

So personally I'd be fine with just WS and HTTP/1.1.

/Tobias

From joakim@intalio.com  Wed May 22 08:33:21 2013
Return-Path: <joakim@intalio.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3707121F84A6 for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 08:33:21 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.977
X-Spam-Level: 
X-Spam-Status: No, score=-1.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 qJReH3zJdlig for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 08:33:20 -0700 (PDT)
Received: from mail-ea0-x232.google.com (mail-ea0-x232.google.com [IPv6:2a00:1450:4013:c01::232]) by ietfa.amsl.com (Postfix) with ESMTP id 4709321F9118 for <hybi@ietf.org>; Wed, 22 May 2013 08:33:15 -0700 (PDT)
Received: by mail-ea0-f178.google.com with SMTP id q16so1232195ead.23 for <hybi@ietf.org>; Wed, 22 May 2013 08:33:15 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:x-gm-message-state; bh=nX64Lb458TnlAwWlTWgJApfAYlnUabKnaT9PRv00dN8=; b=f2iiybVFjv6AAdI5WHGgUCNehvYBwdpVj8h2flhOnaBDqRKUIxtta137UnKLSNKXJ3 x/huAixA3jDCGDSGJc+3STAv3euqfz62btAvYCg7jRz6U76g8Ia7Vp5CDOuTHTDCqSd/ rjDc1HLEa57HxP9GpJvtDCa5k8aDGohjL4GheMvDc0GN5OBNQ0o6DARekDrCZs3vkqsR IwV4q/6KiSwDWKB9reZith8b9I7fzViQlC8X+mx3BiqdxBF9jhFl0ZawT/CSabvvGGyS 0wXRs9BiCJwr0Y4upLezJ3eGftabGmmpcZbQZnXjF1faDLskMnLL4gkhI8M/brN4MlP5 j/Cw==
MIME-Version: 1.0
X-Received: by 10.15.107.77 with SMTP id ca53mr19743217eeb.40.1369236795074; Wed, 22 May 2013 08:33:15 -0700 (PDT)
Received: by 10.15.32.8 with HTTP; Wed, 22 May 2013 08:33:14 -0700 (PDT)
In-Reply-To: <007501ce56f0$67f74080$37e5c180$@noemax.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com>
Date: Wed, 22 May 2013 08:33:14 -0700
Message-ID: <CAG4zZZBpL3=szF6R80SHJeSTYig-bKe8cDPeJS=r2j4Z5oJ_NQ@mail.gmail.com>
From: Joakim Erdfelt <joakim@intalio.com>
To: Arman Djusupov <arman@noemax.com>
Content-Type: multipart/alternative; boundary=089e0162830cc8d94404dd504832
X-Gm-Message-State: ALoCoQmbH8Qw59WO5OUrtE7AVG61O2fElCLJ4zCFFwf7GyfpLe24NDxrFDLwi+MIF402rOZQuSLH
Cc: hybi@ietf.org
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 15:33:21 -0000

--089e0162830cc8d94404dd504832
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

+1 for simplification.

I'll toss some topics onto the table now:
 * AddChannelRequest - eliminate header deltas
 * AddChannelResponse - eliminate header deltas
 * NewChannelSlot - eliminate the "channel slots" management entirely from
the spec, go back to the earlier, easier to implement, concepts from past
drafts.
 * Simplify the extension rules.
     Extensions after initial physical connection are applied only to first
logical channel.
     in AddChannelRequest, extensions provided after the mux extension are
used for logical channels.



--
Joakim Erdfelt <joakim@intalio.com>
webtide.com <http://www.webtide.com/>
Developer advice, services and support
from the Jetty & CometD experts
eclipse.org/jetty - cometd.org


On Wed, May 22, 2013 at 6:29 AM, Arman Djusupov <arman@noemax.com> wrote:

> Sorry, just realized that my message was sent to Takeshi privately by
> mistake a few days ago.. (wrong button). It was actually intended for the
> list so I am repeating it publicly now:****
>
> ** **
>
> ** **
>
> Obviously continuing to develop a hybi-mux spec for another year or two
> doesn=92t make sense. Taking into account that new specifications are com=
ing
> and that HTTP2.0 will probably provide multiplexing for both WebSocket an=
d
> HTTP traffic, at  this stage it would only be worth investing development
> time in a =93simple and ready to go=94 hybi-mux specification. Maybe if w=
e
> simplify the spec to rock bottom simplicity, even sacrificing some featur=
es
> so that we can conclude it very fast, then maybe vendors might pick it up=
.
> This way we will give to WebSocket its own mux for now, while the HTTPbis
> draft might take much longer to form into the final specification. In any
> case, hybi-mux must be reviewed and finalized in a very short time, plus =
it
> should be simpler and easier to implement than both SDPY and HTTPbis base=
d
> multiplexing.****
>
> ** **
>
> With best regards,****
>
> Arman****
>
> ** **
>
> *From:* hybi-bounces@ietf.org [mailto:hybi-bounces@ietf.org] *On Behalf
> Of *Takeshi Yoshino
> *Sent:* Sunday, May 19, 2013 7:31 AM
> *To:* hybi@ietf.org
> *Subject:* Re: [hybi] Call for interest: multiplexing dedicated for
> WebSocket****
>
> ** **
>
> I got one comment about this.****
>
> ** **
>
> He/she suggested that we finish the WebSocket dedicated multiplexing
> quickly even by simplifying the spec much e.g. cutting some functionality
> in it that would take much time to get consensus, and when HTTP/2.0 is
> matured, switch to that.****
>
>  ****
>
> _______________________________________________
> hybi mailing list
> hybi@ietf.org
> https://www.ietf.org/mailman/listinfo/hybi
>
>

--089e0162830cc8d94404dd504832
Content-Type: text/html; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">+1 for simplification.<div><br></div><div style>I&#39;ll t=
oss some topics onto the table now:</div><div style>=A0* AddChannelRequest =
- eliminate header deltas</div><div style>=A0* AddChannelResponse - elimina=
te header deltas</div>
<div style>=A0* NewChannelSlot - eliminate the &quot;channel slots&quot; ma=
nagement entirely from the spec, go back to the earlier, easier to implemen=
t, concepts from past drafts.</div><div style>=A0* Simplify the extension r=
ules. =A0</div>
<div style>=A0 =A0 =A0Extensions after initial physical connection are appl=
ied only to first logical channel.</div><div style>=A0 =A0 =A0in AddChannel=
Request, extensions provided after the mux extension are used for logical c=
hannels.</div>
<div style><br></div><div style><br></div></div><div class=3D"gmail_extra">=
<br clear=3D"all"><div><div>--</div><div>Joakim Erdfelt &lt;<a href=3D"mail=
to:joakim@intalio.com" target=3D"_blank">joakim@intalio.com</a>&gt;</div><d=
iv><a href=3D"http://www.webtide.com/" target=3D"_blank">webtide.com</a></d=
iv>
<div><span>Developer advice, services and support</span><br><span>from the =
Jetty &amp; CometD experts</span></div><div><a href=3D"http://eclipse.org/j=
etty/" target=3D"_blank">eclipse.org/jetty</a>=A0-=A0<a href=3D"http://come=
td.org/" target=3D"_blank">cometd.org</a></div>
</div>
<br><br><div class=3D"gmail_quote">On Wed, May 22, 2013 at 6:29 AM, Arman D=
jusupov <span dir=3D"ltr">&lt;<a href=3D"mailto:arman@noemax.com" target=3D=
"_blank">arman@noemax.com</a>&gt;</span> wrote:<br><blockquote class=3D"gma=
il_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-lef=
t:1ex">
<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple"><div><p class=3D"MsoNorm=
al"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,&quot;s=
ans-serif&quot;;color:#1f497d">Sorry, just realized that my message was sen=
t to Takeshi privately by mistake a few days ago.. (wrong button). It was a=
ctually intended for the list so I am repeating it publicly now:<u></u><u><=
/u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></spa=
n></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Obviously continuing to d=
evelop a hybi-mux spec for another year or two doesn=92t make sense. Taking=
 into account that new specifications are coming and that HTTP2.0 will prob=
ably provide multiplexing for both WebSocket and HTTP traffic, at=A0 this s=
tage it would only be worth investing development time in a =93simple and r=
eady to go=94 hybi-mux specification. Maybe if we simplify the spec to rock=
 bottom simplicity, even sacrificing some features so that we can conclude =
it very fast, then maybe vendors might pick it up. This way we will give to=
 WebSocket its own mux for now, while the HTTPbis draft might take much lon=
ger to form into the final specification. In any case, hybi-mux must be rev=
iewed and finalized in a very short time, plus it should be simpler and eas=
ier to implement than both SDPY and HTTPbis based multiplexing.<u></u><u></=
u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></span><=
/p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d">With best regards,<u><=
/u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,&quot;sans-serif&quot;;color:#1f497d">Arman<u></u><u></u></span=
></p><p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&qu=
ot;Calibri&quot;,&quot;sans-serif&quot;;color:#1f497d"><u></u>=A0<u></u></s=
pan></p>
<p class=3D"MsoNormal"><b><span style=3D"font-size:10.0pt;font-family:&quot=
;Tahoma&quot;,&quot;sans-serif&quot;">From:</span></b><span style=3D"font-s=
ize:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> <a href=
=3D"mailto:hybi-bounces@ietf.org" target=3D"_blank">hybi-bounces@ietf.org</=
a> [mailto:<a href=3D"mailto:hybi-bounces@ietf.org" target=3D"_blank">hybi-=
bounces@ietf.org</a>] <b>On Behalf Of </b>Takeshi Yoshino<br>
<b>Sent:</b> Sunday, May 19, 2013 7:31 AM<br><b>To:</b> <a href=3D"mailto:h=
ybi@ietf.org" target=3D"_blank">hybi@ietf.org</a><br><b>Subject:</b> Re: [h=
ybi] Call for interest: multiplexing dedicated for WebSocket<u></u><u></u><=
/span></p>
<div class=3D"im"><p class=3D"MsoNormal"><u></u>=A0<u></u></p><div><p class=
=3D"MsoNormal">I got one comment about this.<u></u><u></u></p><div><p class=
=3D"MsoNormal"><u></u>=A0<u></u></p></div><div><p class=3D"MsoNormal">He/sh=
e suggested that we finish the WebSocket dedicated multiplexing quickly eve=
n by simplifying the spec much e.g. cutting some functionality in it that w=
ould take much time to get consensus, and when HTTP/2.0 is matured, switch =
to that.<u></u><u></u></p>
<div><p class=3D"MsoNormal">=A0<u></u><u></u></p></div></div></div></div></=
div></div><br>_______________________________________________<br>
hybi mailing list<br>
<a href=3D"mailto:hybi@ietf.org">hybi@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/hybi" target=3D"_blank">ht=
tps://www.ietf.org/mailman/listinfo/hybi</a><br>
<br></blockquote></div><br></div>

--089e0162830cc8d94404dd504832--

From jat@jaet.org  Wed May 22 09:27:21 2013
Return-Path: <jat@jaet.org>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CF73921F970F for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 09:27:21 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.425
X-Spam-Level: 
X-Spam-Status: No, score=-0.425 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FH_RELAY_NODNS=1.451, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, RDNS_NONE=0.1]
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 XRwWJdS9MEKD for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 09:27:16 -0700 (PDT)
Received: from mail-ie0-x22e.google.com (mail-ie0-x22e.google.com [IPv6:2607:f8b0:4001:c03::22e]) by ietfa.amsl.com (Postfix) with ESMTP id 7245221F96C9 for <hybi@ietf.org>; Wed, 22 May 2013 09:27:16 -0700 (PDT)
Received: by mail-ie0-f174.google.com with SMTP id 10so5483884ied.5 for <hybi@ietf.org>; Wed, 22 May 2013 09:27:11 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:x-originating-ip:in-reply-to:references:from:date :message-id:subject:to:cc:content-type:x-gm-message-state; bh=Zke0Ol3t1HsnFpbgWAZwMv/CbOJy0xlGzoHVZdZglsA=; b=lgnlBf6d7uRgl1wt/vsU8IKMZB4MW5IH7QqUfFUU3L8OdcwjyGJ1nrB4zGp3q2dzP0 IxQFTn6Brhob2+Qg/H8OBawo5RIgrCh5x7BmpXBa+P2abF9V5PliTgC7rgWk4oA/sJBM 9m8/8ii8dM0dr/tVkzR5lfCJkHg3foeAhTZHdK1lL/Dr+UAlvslSSSZlAx1OXU73GtVW hPCaF+kKzX3ORT8TJxTqfPugpQTYxwIgSpJ+b9bnglZ/HcOLvaNcqXOgA9BntmFLafZx QS5gCdmTvYNbRCGPsRX8p3s37eAXzZJR0MXVYmPfn1mgcLaaNFuZXHHzr+rEkY4/FpFE kWtg==
X-Received: by 10.42.70.147 with SMTP id f19mr6630007icj.30.1369240031284; Wed, 22 May 2013 09:27:11 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.50.37.236 with HTTP; Wed, 22 May 2013 09:26:51 -0700 (PDT)
X-Originating-IP: [207.67.64.142]
In-Reply-To: <519CE075.4000106@tavendo.de>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de>
From: "John A. Tamplin" <jat@jaet.org>
Date: Wed, 22 May 2013 12:26:51 -0400
Message-ID: <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com>
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
Content-Type: multipart/alternative; boundary=90e6ba1effeead6a7304dd510989
X-Gm-Message-State: ALoCoQlNWOZui8MMLc65F4/n0+aW32mE+2ROTq4YKNjIjeWTgIuDRMgyw7viafPdbRH+DIpPrV/n
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 16:27:21 -0000

--90e6ba1effeead6a7304dd510989
Content-Type: text/plain; charset=UTF-8

On Wed, May 22, 2013 at 11:12 AM, Tobias Oberstein <
tobias.oberstein@tavendo.de> wrote:

> +1 for the simplest possible mux that does the core job.
>
> However, I currently think that having a per-logical channel flow control
> is required to make it useful (cannot be dropped for the sake of
> simplification). I might be wrong.
>

I agree -- without flow control, there is very little benefit, and almost
nothing that an app couldn't easily do themselves by just adding a simple
framing layer inside WS messages (perhaps just the first byte is a logical
channel number).

-- 
John A. Tamplin

--90e6ba1effeead6a7304dd510989
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Wed, May 22, 2013 at 11:12 AM, Tobias Oberstein <span d=
ir=3D"ltr">&lt;<a href=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_bl=
ank" class=3D"cremed">tobias.oberstein@tavendo.de</a>&gt;</span> wrote:<br>=
<div class=3D"gmail_extra">

<div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margi=
n:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">+1 for the simple=
st possible mux that does the core job.<br>
<br>
However, I currently think that having a per-logical channel flow control i=
s required to make it useful (cannot be dropped for the sake of simplificat=
ion). I might be wrong.<br></blockquote><div><br></div><div style>I agree -=
- without flow control, there is very little benefit, and almost nothing th=
at an app couldn&#39;t easily do themselves by just adding a simple framing=
 layer inside WS messages (perhaps just the first byte is a logical channel=
 number).</div>

</div><div><br></div>-- <br>John A. Tamplin
</div></div>

--90e6ba1effeead6a7304dd510989--

From simone.bordet@gmail.com  Wed May 22 09:50:22 2013
Return-Path: <simone.bordet@gmail.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B883211E80FF for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 09:50:22 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.977
X-Spam-Level: 
X-Spam-Status: No, score=-1.977 tagged_above=-999 required=5 tests=[AWL=-0.001, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001]
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 y72N2ZWamfcA for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 09:50:17 -0700 (PDT)
Received: from mail-pd0-f180.google.com (mail-pd0-f180.google.com [209.85.192.180]) by ietfa.amsl.com (Postfix) with ESMTP id 4B19721F96F9 for <hybi@ietf.org>; Wed, 22 May 2013 09:48:44 -0700 (PDT)
Received: by mail-pd0-f180.google.com with SMTP id 10so1897505pdc.39 for <hybi@ietf.org>; Wed, 22 May 2013 09:48:43 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=I77T8lcEMHDJjlnXn1BIXcWqAWchRv7xfZ2a6Sl4mns=; b=lpQtT50ES9T7P7EuQnGYTGqSvIZhLK1BQEtU6aK/KosTka3/AlR2snsYrqGbtyL5nw uzTQ9RZkUVV0/wahw6JAuZoBolIH4EGi/3X7/cVg+MG4q1Tphq5HTGCK++uu+5ogNgZX 5ixVX8NRm7DBfUXByBywrQkhHXIxMIOdbBbbqlhEyNtk+bvy+RW9yZEV2DyO/JIYJMh1 UQEvWRMt+uVCU6Gz2Vx/e48Ui19i//gjBrBxrNL/ANh7lYSa/2oOcn3eyR+orNJj7u37 Hg8GYfXuhKq6tfACePPi5vl+ipA2RO+uA3YxEd2hbpGeh9ygmNlPgLCN+hKojHcw3gmk Zsew==
MIME-Version: 1.0
X-Received: by 10.66.235.69 with SMTP id uk5mr9136729pac.110.1369241323615; Wed, 22 May 2013 09:48:43 -0700 (PDT)
Sender: simone.bordet@gmail.com
Received: by 10.68.103.129 with HTTP; Wed, 22 May 2013 09:48:43 -0700 (PDT)
In-Reply-To: <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com>
Date: Wed, 22 May 2013 18:48:43 +0200
X-Google-Sender-Auth: ki20rsALVOkWXLS-bw324CYNQOI
Message-ID: <CAFWmRJ2Hbe0x5FeV2T7Gkp3WEsxQHe2=YPBTgvHYLcus3A4rBQ@mail.gmail.com>
From: Simone Bordet <sbordet@intalio.com>
To: "John A. Tamplin" <jat@jaet.org>
Content-Type: multipart/alternative; boundary=047d7b111e83b4c10904dd5156c5
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 16:50:22 -0000

--047d7b111e83b4c10904dd5156c5
Content-Type: text/plain; charset=UTF-8

Hi,

On Wed, May 22, 2013 at 6:26 PM, John A. Tamplin <jat@jaet.org> wrote:
> I agree -- without flow control, there is very little benefit, and almost
> nothing that an app couldn't easily do themselves by just adding a simple
> framing layer inside WS messages (perhaps just the first byte is a logical
> channel number).

In SPDY, the flow control had an initial window of 64k. This proved way too
small for uploads/downloads, which resulted in a 3x or more slowdown.
So there needs to be in ws a mechanism to configure this window (which SPDY
has).
Turns out that now every client sends a huge initial window, so downloads
are now fast.
This of course impacts memory, since flow control is about memory, but
pretty much solved the flow control problem.

The next problem that arose was that now a stream/channel can take the
whole bandwidth and squat others channels.
So there is the need for a further flow control mechanism to prioritize
channels, that must be dynamic (i.e. changed during the transmissions of
frames).
SPDY is still discussing this, I am guessing it's also being discussed in
HTTP 2.0.

So "simple mux" looks like an oxymoron to me, and makes me wonder what
useful could be a ws mux solution that does not do dynamic channel
prioritization.
A typical use case Joakim presented to me was to tune video download rate
in a websocket server: a big priority initially to download faster than the
client video player plays, to make some buffer, and then reducing the
priority to a steady state with the video player's speed, or reducing
priority to almost zero if the user pauses the play.

--
Simone Bordet
----
http://cometd.org
http://webtide.com
http://intalio.com
Developer advice, training, services and support
from the Jetty & CometD experts.
Intalio, the modern way to build business applications.

--047d7b111e83b4c10904dd5156c5
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hi,<br><br>On Wed, May 22, 2013 at 6:26 PM, John A. Tampli=
n &lt;<a href=3D"mailto:jat@jaet.org">jat@jaet.org</a>&gt; wrote:<br>&gt; I=
 agree -- without flow control, there is very little benefit, and almost<br=
>
&gt; nothing that an app couldn&#39;t easily do themselves by just adding a=
 simple<br>&gt; framing layer inside WS messages (perhaps just the first by=
te is a logical<br>&gt; channel number).<br><br>In SPDY, the flow control h=
ad an initial window of 64k. This proved way too small for uploads/download=
s, which resulted in a 3x or more slowdown.<br>
So there needs to be in ws a mechanism to configure this window (which SPDY=
 has).<br>Turns out that now every client sends a huge initial window, so d=
ownloads are now fast.<br>This of course impacts memory, since flow control=
 is about memory, but pretty much solved the flow control problem.<br>
<br>The next problem that arose was that now a stream/channel can take the =
whole bandwidth and squat others channels.<br>So there is the need for a fu=
rther flow control mechanism to prioritize channels, that must be dynamic (=
i.e. changed during the transmissions of frames).<br>
SPDY is still discussing this, I am guessing it&#39;s also being discussed =
in HTTP 2.0.<br><br>So &quot;simple mux&quot; looks like an oxymoron to me,=
 and makes me wonder what useful could be a ws mux solution that does not d=
o dynamic channel prioritization.<br>
A typical use case Joakim presented to me was to tune video download rate i=
n a websocket server: a big priority initially to download faster than the =
client video player plays, to make some buffer, and then reducing the prior=
ity to a steady state with the video player&#39;s speed, or reducing priori=
ty to almost zero if the user pauses the play.<br>
<br><span class=3D"sew2g0eoo9eutzc"></span><span class=3D"sew2g0eoo9eutzc">=
</span>--<br>Simone Bordet<br>----<br><a href=3D"http://cometd.org">http://=
cometd.org</a><br><a href=3D"http://webtide.com">http://webtide.com</a><br>=
<a href=3D"http://intalio.com">http://intalio.com</a><br>
Developer advice, training, services and support<br>from the Jetty &amp; Co=
metD experts.<br>Intalio, the modern way to build business applications.</d=
iv>

--047d7b111e83b4c10904dd5156c5--

From zhong.j.yu@gmail.com  Wed May 22 10:13:48 2013
Return-Path: <zhong.j.yu@gmail.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 677FC21F96EF for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 10:13:48 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -7.098
X-Spam-Level: 
X-Spam-Status: No, score=-7.098 tagged_above=-999 required=5 tests=[AWL=-3.500, 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 N9UFNKDRNxcq for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 10:13:43 -0700 (PDT)
Received: from mail-oa0-f49.google.com (mail-oa0-f49.google.com [209.85.219.49]) by ietfa.amsl.com (Postfix) with ESMTP id 8B15121F8E8E for <hybi@ietf.org>; Wed, 22 May 2013 10:13:39 -0700 (PDT)
Received: by mail-oa0-f49.google.com with SMTP id k14so3071832oag.36 for <hybi@ietf.org>; Wed, 22 May 2013 10:13:39 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=w/yR3er5NX44sNyPnF3idfoMK9VYkfTXcp7daO8lmZ0=; b=lHfqtLIJnp60EOZVz5sL/Ldg4iuOWlLhnMPGZOCp5CGvUchMgjTTHh/l68T9ZR53cL 8SFnGO8r6ik8N8wIaU26V5rYsxHMJ8C24A66bYJXZu5OqlbPGHRsyGd4rjamiGcw0gIl Mq5cRc03mgSG/4zhKn5nYYsCeyBelSYwzcDk4rozaoF6+AxDTiHjpgMnJ0nmKGTL6nGM sCpcVo5bFdZcoyeZJYptu1GNx0bTPNm/84MM/Z+ub7Dh6MCQeXBlwvsq4CLJ2Rp/qyz8 Sw7YxtQ77Fj1uEJBofZbmA7+HBIFrRp3GEMna5wDvirNeUm8RQuYw1mOkbjsi8MDWat/ NkpQ==
MIME-Version: 1.0
X-Received: by 10.182.138.4 with SMTP id qm4mr5652267obb.101.1369242819179; Wed, 22 May 2013 10:13:39 -0700 (PDT)
Received: by 10.76.98.227 with HTTP; Wed, 22 May 2013 10:13:39 -0700 (PDT)
In-Reply-To: <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com>
Date: Wed, 22 May 2013 12:13:39 -0500
Message-ID: <CACuKZqEXghX9ONECBGTEB1cD5Pt5cakKih9unjSZAmFnHZ1f2A@mail.gmail.com>
From: Zhong Yu <zhong.j.yu@gmail.com>
To: "John A. Tamplin" <jat@jaet.org>
Content-Type: multipart/alternative; boundary=e89a8ff1cd0ed93bd404dd51af4f
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 17:13:48 -0000

--e89a8ff1cd0ed93bd404dd51af4f
Content-Type: text/plain; charset=ISO-8859-1

On Wed, May 22, 2013 at 11:26 AM, John A. Tamplin <jat@jaet.org> wrote:

> On Wed, May 22, 2013 at 11:12 AM, Tobias Oberstein <
> tobias.oberstein@tavendo.de> wrote:
>
>> +1 for the simplest possible mux that does the core job.
>>
>> However, I currently think that having a per-logical channel flow control
>> is required to make it useful (cannot be dropped for the sake of
>> simplification). I might be wrong.
>>
>
> I agree -- without flow control, there is very little benefit, and almost
> nothing that an app couldn't easily do themselves by just adding a simple
> framing layer inside WS messages (perhaps just the first byte is a logical
> channel number).
>
>
In traditional socket programming model, the application chooses the time
and the speed to perform read(). If the application is not reading fast
enough, the receive buffer is full and more incoming data are not welcome.

Contrary to that, most WebSocket API are event based - all incoming message
data are buffered without any blocking, then the complete message is sent
to the application callback. In this model, per channel flow control
doesn't seems to exist.

If per channel flow control is removed from the spec, what use cases would
suffer?

--
And I have a stupid question - what's the motivation to do mux on
WebSocket? To save local port numbers? To avoid slow start? If we drop mux
all together, what use cases would suffer?

Zhong Yu

--e89a8ff1cd0ed93bd404dd51af4f
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><br><div class=3D"gmail_extra"><br><br><div class=3D"gmail=
_quote">On Wed, May 22, 2013 at 11:26 AM, John A. Tamplin <span dir=3D"ltr"=
>&lt;<a href=3D"mailto:jat@jaet.org" target=3D"_blank">jat@jaet.org</a>&gt;=
</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div dir=3D"ltr"><div class=3D"im">On Wed, M=
ay 22, 2013 at 11:12 AM, Tobias Oberstein <span dir=3D"ltr">&lt;<a href=3D"=
mailto:tobias.oberstein@tavendo.de" target=3D"_blank">tobias.oberstein@tave=
ndo.de</a>&gt;</span> wrote:<br>
</div><div class=3D"gmail_extra">

<div class=3D"gmail_quote"><div class=3D"im"><blockquote class=3D"gmail_quo=
te" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"=
>+1 for the simplest possible mux that does the core job.<br>
<br>
However, I currently think that having a per-logical channel flow control i=
s required to make it useful (cannot be dropped for the sake of simplificat=
ion). I might be wrong.<br></blockquote><div><br></div></div><div>I agree -=
- without flow control, there is very little benefit, and almost nothing th=
at an app couldn&#39;t easily do themselves by just adding a simple framing=
 layer inside WS messages (perhaps just the first byte is a logical channel=
 number).</div>
<span class=3D"HOEnZb"><font color=3D"#888888">

</font></span></div><span class=3D"HOEnZb"></span><br></div></div></blockqu=
ote></div><br></div><div class=3D"gmail_extra">In traditional socket progra=
mming model, the application chooses the time and the speed to perform read=
(). If the application is not reading fast enough, the receive buffer is fu=
ll and more incoming data are not welcome.<br>
</div><div class=3D"gmail_extra"><br></div><div class=3D"gmail_extra">Contr=
ary to that, most WebSocket API are event based - all incoming message data=
 are buffered without any blocking, then the complete message is sent to th=
e application callback. In this model, per channel flow control doesn&#39;t=
 seems to exist.<br>
<br></div><div class=3D"gmail_extra">If per channel flow control is removed=
 from the spec, what use cases would suffer?<br><br>--<br></div><div class=
=3D"gmail_extra">And I have a stupid question - what&#39;s the motivation t=
o do mux on WebSocket? To save local port numbers? To avoid slow start? If =
we drop mux all together, what use cases would suffer?<br>
</div><div class=3D"gmail_extra"><br></div><div class=3D"gmail_extra">Zhong=
 Yu<br></div><div class=3D"gmail_extra"><br></div></div>

--e89a8ff1cd0ed93bd404dd51af4f--

From tobias.oberstein@tavendo.de  Wed May 22 10:40:02 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2189D21F923B for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 10:40:02 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id aWjGm1Sp0R7h for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 10:39:56 -0700 (PDT)
Received: from EXHUB020-2.exch020.serverdata.net (exhub020-2.exch020.serverdata.net [206.225.164.29]) by ietfa.amsl.com (Postfix) with ESMTP id 9822D21F915C for <hybi@ietf.org>; Wed, 22 May 2013 10:39:56 -0700 (PDT)
Received: from [10.157.82.16] (89.204.138.16) by smtpx20.serverdata.net (206.225.164.34) with Microsoft SMTP Server (TLS) id 8.3.264.0; Wed, 22 May 2013 10:39:53 -0700
Message-ID: <519D02E7.6040009@tavendo.de>
Date: Wed, 22 May 2013 19:39:51 +0200
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: "John A. Tamplin" <jat@jaet.org>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com>
In-Reply-To: <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 17:40:02 -0000

Am 22.05.2013 18:26, schrieb John A. Tamplin:
> On Wed, May 22, 2013 at 11:12 AM, Tobias Oberstein
> <tobias.oberstein@tavendo.de <mailto:tobias.oberstein@tavendo.de>> wrote:
>
>     +1 for the simplest possible mux that does the core job.
>
>     However, I currently think that having a per-logical channel flow
>     control is required to make it useful (cannot be dropped for the
>     sake of simplification). I might be wrong.
>
>
> I agree -- without flow control, there is very little benefit, and
> almost nothing that an app couldn't easily do themselves by just adding
> a simple framing layer inside WS messages (perhaps just the first byte
> is a logical channel number).

Regarding implementing flow control above TCP (and below app), I found this:

"Multiplexing on top of TCP in overall fails to deliver the advantages 
it is assumed to provide."

Source: http://www.250bpm.com/multiplexing

Do you have an opinion on that?

More important: what problem exactly does multiplexing solve? The 
current mux RFC mentions "scalability", but doesn't go into detail ..

/Tobias

>
> --
> John A. Tamplin


From jat@jaet.org  Wed May 22 11:32:47 2013
Return-Path: <jat@jaet.org>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id ECC5011E812E for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 11:32:47 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.425
X-Spam-Level: 
X-Spam-Status: No, score=-0.425 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FH_RELAY_NODNS=1.451, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, RDNS_NONE=0.1]
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 8IyEqniBBLqM for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 11:32:43 -0700 (PDT)
Received: from mail-ie0-x234.google.com (mail-ie0-x234.google.com [IPv6:2607:f8b0:4001:c03::234]) by ietfa.amsl.com (Postfix) with ESMTP id 7483611E812F for <hybi@ietf.org>; Wed, 22 May 2013 11:32:43 -0700 (PDT)
Received: by mail-ie0-f180.google.com with SMTP id ar20so5876468iec.39 for <hybi@ietf.org>; Wed, 22 May 2013 11:32:43 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:x-originating-ip:in-reply-to:references:from:date :message-id:subject:to:cc:content-type:x-gm-message-state; bh=Ko8m7Fn5a7sS+ethsduBqC7LixQCBl69UHTC/XwixPA=; b=b/t1RXODnCxGMfgrofe/tJdd1bcYA/WcFNDBjuCZY6KcBecmC3XxD0RwTcZkrqnm1S hq2cEf693Ttpz1R0OuH2dzb12eNgHeZrS4V6thaFODtW4th4qh3UEV+W8JheNsWJVGr3 QxIlsKwQx8VogCaEU9l64mmTN175UjYM0A5l77DrcoEyWceU9DQKJuWHatLMtilrYzno nJC0POof9kdJ49QMI5qgeL9QmkkY5XMfOwymTbQGAgx99TESRfEAzlqRWbGVfxT1Jy07 f3CpTFVXCrMQXtCEZLehRAFrMdl1FZzhIkdKsw2xbhyyRtIHyG+BNTJD3TZ7yE/tHJeF PeJg==
X-Received: by 10.42.84.201 with SMTP id n9mr7733587icl.47.1369247562895; Wed, 22 May 2013 11:32:42 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.50.37.236 with HTTP; Wed, 22 May 2013 11:32:22 -0700 (PDT)
X-Originating-IP: [207.67.64.142]
In-Reply-To: <519D02E7.6040009@tavendo.de>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <519D02E7.6040009@tavendo.de>
From: "John A. Tamplin" <jat@jaet.org>
Date: Wed, 22 May 2013 14:32:22 -0400
Message-ID: <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com>
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
Content-Type: multipart/alternative; boundary=20cf301cc49298aa2904dd52ca2e
X-Gm-Message-State: ALoCoQklRrmCGhMKv6TXp82haBV6t8CW7egjUT8d1ir9YKqYwQG7CYAdOgta1zz8lj9L3hb6k9D4
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 18:32:48 -0000

--20cf301cc49298aa2904dd52ca2e
Content-Type: text/plain; charset=UTF-8

On Wed, May 22, 2013 at 1:39 PM, Tobias Oberstein <
tobias.oberstein@tavendo.de> wrote:

> Regarding implementing flow control above TCP (and below app), I found
> this:
>
> "Multiplexing on top of TCP in overall fails to deliver the advantages it
> is assumed to provide."
>
> Source: http://www.250bpm.com/**multiplexing<http://www.250bpm.com/multiplexing>
>
> Do you have an opinion on that?
>

One thing he doesn't address are resources used to keep lots of connections
open, both at the OS level and in intermediate network nodes (such as load
balancers -- try running tons of long-lived connections through an F5 for
example).  Also, he doesn't address the interaction of TCP slow start with
lots of new connections.  I also don't believe the equivalent of a full TCP
handshake is required to allocate buffer space.


> More important: what problem exactly does multiplexing solve? The current
> mux RFC mentions "scalability", but doesn't go into detail ..
>

Here is a use-case I envision:  Imagine GMail, G+ sandbar, etc all use
WebSockets heavily.  Many users may have multiple copies of such apps
running in tabs or windows in the browser, and some of these are going to
be separate (such as GMail and the G+ sandbar probably are separate apps
running on the same page and maintain their own connections to their
backends).  Currently, that would mean you have a ton of TCP connections
open to the server.  Servers don't handle tons of long-lived connections
very well (or intermediate load balancers), so you would greatly prefer to
have one connection per browser.  WS MUX lets you do that, by having the
browser aggregate connections to the same backend.  You could do some of it
at the app level, but it is going to be complicated and in many cases these
apps won't all be from the same vendor (for example, some widget on the
page that uses the GData API) and will have separate deploy schedules.

-- 
John A. Tamplin

--20cf301cc49298aa2904dd52ca2e
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Wed, May 22, 2013 at 1:39 PM, Tobias Oberstein <span di=
r=3D"ltr">&lt;<a href=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_bla=
nk" class=3D"cremed">tobias.oberstein@tavendo.de</a>&gt;</span> wrote:<br><=
div class=3D"gmail_extra">

<div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margi=
n:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Regarding impleme=
nting flow control above TCP (and below app), I found this:<br>
<br>
&quot;Multiplexing on top of TCP in overall fails to deliver the advantages=
 it is assumed to provide.&quot;<br>
<br>
Source: <a href=3D"http://www.250bpm.com/multiplexing" target=3D"_blank" cl=
ass=3D"cremed">http://www.250bpm.com/<u></u>multiplexing</a><br>
<br>
Do you have an opinion on that?<br></blockquote><div><br></div><div style>O=
ne thing he doesn&#39;t address are resources used to keep lots of connecti=
ons open, both at the OS level and in intermediate network nodes (such as l=
oad balancers -- try running tons of long-lived connections through an F5 f=
or example). =C2=A0Also, he doesn&#39;t address the interaction of TCP slow=
 start with lots of new connections. =C2=A0I also don&#39;t believe the equ=
ivalent of a full TCP handshake is required to allocate buffer space.</div>

<div>=C2=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8=
ex;border-left:1px #ccc solid;padding-left:1ex">
More important: what problem exactly does multiplexing solve? The current m=
ux RFC mentions &quot;scalability&quot;, but doesn&#39;t go into detail ..<=
span class=3D"HOEnZb"><font color=3D"#888888"><br></font></span></blockquot=
e>

</div><div class=3D"gmail_extra"><br></div>Here is a use-case I envision: =
=C2=A0Imagine GMail, G+ sandbar, etc all use WebSockets heavily. =C2=A0Many=
 users may have multiple copies of such apps running in tabs or windows in =
the browser, and some of these are going to be separate (such as GMail and =
the G+ sandbar probably are separate apps running on the same page and main=
tain their own connections to their backends). =C2=A0Currently, that would =
mean you have a ton of TCP connections open to the server. =C2=A0Servers do=
n&#39;t handle tons of long-lived connections very well (or intermediate lo=
ad balancers), so you would greatly prefer to have one connection per brows=
er. =C2=A0WS MUX lets you do that, by having the browser aggregate connecti=
ons to the same backend. =C2=A0You could do some of it at the app level, bu=
t it is going to be complicated and in many cases these apps won&#39;t all =
be from the same vendor (for example, some widget on the page that uses the=
 GData API) and will have separate deploy schedules.</div>

<div class=3D"gmail_extra"><br></div><div class=3D"gmail_extra">-- <br>John=
 A. Tamplin
</div></div>

--20cf301cc49298aa2904dd52ca2e--

From zhong.j.yu@gmail.com  Wed May 22 12:00:19 2013
Return-Path: <zhong.j.yu@gmail.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1384821F9673 for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 12:00:19 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 hl5Z8xca9LBa for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 12:00:17 -0700 (PDT)
Received: from mail-ob0-x230.google.com (mail-ob0-x230.google.com [IPv6:2607:f8b0:4003:c01::230]) by ietfa.amsl.com (Postfix) with ESMTP id 61F4C21F9021 for <hybi@ietf.org>; Wed, 22 May 2013 12:00:15 -0700 (PDT)
Received: by mail-ob0-f176.google.com with SMTP id wp18so2797914obc.7 for <hybi@ietf.org>; Wed, 22 May 2013 12:00:10 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=pAL1c8grR94qbXnhDx2iec+AverY8DdTuFGpQg8rRyo=; b=xMXGK5f/+V60FGJ4CaLTjfoxi1zlMyQ8YNuQzYWy1Deq5KaM688HsKwC1ByBKUZIE7 0PgGmowyfmbjz18FlXg2X4E3JqZBHbcaQOdwT6+Qr3OgtmS7saUq8fmtM9dOBemKayAP GokQUQXR32gVMa7HIkpdSJikWX9j2EFRCcgcYBfHaWtkY7oYqpCyDYfJ4d1N/XJGzwQ6 S812H6H7ZxknRVYQgOBImNBQXqp0G6ziOGFuoE9fHqVQe0d8fr8samVeD19wFPrCK+YW mfntrDJuYe3EIyZaQv0HZrnuJuPAry+dbUql4XRD6G+zQU1CykVOSYSaTD8AFAfWX03v XMWA==
MIME-Version: 1.0
X-Received: by 10.60.79.165 with SMTP id k5mr5820671oex.108.1369249209939; Wed, 22 May 2013 12:00:09 -0700 (PDT)
Received: by 10.76.98.227 with HTTP; Wed, 22 May 2013 12:00:09 -0700 (PDT)
In-Reply-To: <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <519D02E7.6040009@tavendo.de> <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com>
Date: Wed, 22 May 2013 14:00:09 -0500
Message-ID: <CACuKZqHYmF+-zECFrGsHqt0i91jHAzk9Az72=tva_kn=ra2pKA@mail.gmail.com>
From: Zhong Yu <zhong.j.yu@gmail.com>
To: "John A. Tamplin" <jat@jaet.org>
Content-Type: multipart/alternative; boundary=089e012281b6c47fc704dd532c93
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 19:00:19 -0000

--089e012281b6c47fc704dd532c93
Content-Type: text/plain; charset=ISO-8859-1

On Wed, May 22, 2013 at 1:32 PM, John A. Tamplin <jat@jaet.org> wrote:

> On Wed, May 22, 2013 at 1:39 PM, Tobias Oberstein <
> tobias.oberstein@tavendo.de> wrote:
>
>> Regarding implementing flow control above TCP (and below app), I found
>> this:
>>
>> "Multiplexing on top of TCP in overall fails to deliver the advantages it
>> is assumed to provide."
>>
>> Source: http://www.250bpm.com/**multiplexing<http://www.250bpm.com/multiplexing>
>>
>> Do you have an opinion on that?
>>
>
> One thing he doesn't address are resources used to keep lots of
> connections open, both at the OS level and in intermediate network nodes
> (such as load balancers -- try running tons of long-lived connections
> through an F5 for example).  Also, he doesn't address the interaction of
> TCP slow start with lots of new connections.  I also don't believe the
> equivalent of a full TCP handshake is required to allocate buffer space.
>
>
>> More important: what problem exactly does multiplexing solve? The current
>> mux RFC mentions "scalability", but doesn't go into detail ..
>>
>
> Here is a use-case I envision:  Imagine GMail, G+ sandbar, etc all use
> WebSockets heavily.  Many users may have multiple copies of such apps
> running in tabs or windows in the browser, and some of these are going to
> be separate (such as GMail and the G+ sandbar probably are separate apps
> running on the same page and maintain their own connections to their
> backends).  Currently, that would mean you have a ton of TCP connections
> open to the server.
>


> Servers don't handle tons of long-lived connections very well
>

Is that still true today? Recent server implementations all boast they can
handle tons of connections.



> (or intermediate load balancers), so you would greatly prefer to have one
> connection per browser.  WS MUX lets you do that, by having the browser
> aggregate connections to the same backend.  You could do some of it at the
> app level, but it is going to be complicated and in many cases these apps
> won't all be from the same vendor (for example, some widget on the page
> that uses the GData API) and will have separate deploy schedules.
>
> --
> John A. Tamplin
>
> _______________________________________________
> hybi mailing list
> hybi@ietf.org
> https://www.ietf.org/mailman/listinfo/hybi
>
>

--089e012281b6c47fc704dd532c93
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><br><div class=3D"gmail_extra"><br><br><div class=3D"gmail=
_quote">On Wed, May 22, 2013 at 1:32 PM, John A. Tamplin <span dir=3D"ltr">=
&lt;<a href=3D"mailto:jat@jaet.org" target=3D"_blank">jat@jaet.org</a>&gt;<=
/span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div dir=3D"ltr"><div class=3D"im">On Wed, M=
ay 22, 2013 at 1:39 PM, Tobias Oberstein <span dir=3D"ltr">&lt;<a href=3D"m=
ailto:tobias.oberstein@tavendo.de" target=3D"_blank">tobias.oberstein@taven=
do.de</a>&gt;</span> wrote:<br>
</div><div class=3D"gmail_extra">

<div class=3D"gmail_quote"><div class=3D"im"><blockquote class=3D"gmail_quo=
te" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"=
>Regarding implementing flow control above TCP (and below app), I found thi=
s:<br>

<br>
&quot;Multiplexing on top of TCP in overall fails to deliver the advantages=
 it is assumed to provide.&quot;<br>
<br>
Source: <a href=3D"http://www.250bpm.com/multiplexing" target=3D"_blank">ht=
tp://www.250bpm.com/<u></u>multiplexing</a><br>
<br>
Do you have an opinion on that?<br></blockquote><div><br></div></div><div>O=
ne thing he doesn&#39;t address are resources used to keep lots of connecti=
ons open, both at the OS level and in intermediate network nodes (such as l=
oad balancers -- try running tons of long-lived connections through an F5 f=
or example). =A0Also, he doesn&#39;t address the interaction of TCP slow st=
art with lots of new connections. =A0I also don&#39;t believe the equivalen=
t of a full TCP handshake is required to allocate buffer space.</div>
<div class=3D"im">

<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex">
More important: what problem exactly does multiplexing solve? The current m=
ux RFC mentions &quot;scalability&quot;, but doesn&#39;t go into detail ..<=
span><font color=3D"#888888"><br></font></span></blockquote>

</div></div><div class=3D"gmail_extra"><br></div>Here is a use-case I envis=
ion: =A0Imagine GMail, G+ sandbar, etc all use WebSockets heavily. =A0Many =
users may have multiple copies of such apps running in tabs or windows in t=
he browser, and some of these are going to be separate (such as GMail and t=
he G+ sandbar probably are separate apps running on the same page and maint=
ain their own connections to their backends). =A0Currently, that would mean=
 you have a ton of TCP connections open to the server. =A0</div>
</div></blockquote><div>=A0</div><blockquote class=3D"gmail_quote" style=3D=
"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir=3D=
"ltr"><div>Servers don&#39;t handle tons of long-lived connections very wel=
l</div>
</div></blockquote><div><br></div><div>Is that still true today? Recent ser=
ver implementations all boast they can handle tons of connections.<br><br>=
=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;borde=
r-left:1px #ccc solid;padding-left:1ex">
<div dir=3D"ltr"><div class=3D"gmail_extra">(or intermediate load balancers=
), so you would greatly prefer to have one connection per browser. =A0WS MU=
X lets you do that, by having the browser aggregate connections to the same=
 backend. =A0You could do some of it at the app level, but it is going to b=
e complicated and in many cases these apps won&#39;t all be from the same v=
endor (for example, some widget on the page that uses the GData API) and wi=
ll have separate deploy schedules.</div>
<span class=3D"HOEnZb"><font color=3D"#888888">

<div class=3D"gmail_extra"><br></div><div class=3D"gmail_extra">-- <br>John=
 A. Tamplin
</div></font></span></div>
<br>_______________________________________________<br>
hybi mailing list<br>
<a href=3D"mailto:hybi@ietf.org">hybi@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/hybi" target=3D"_blank">ht=
tps://www.ietf.org/mailman/listinfo/hybi</a><br>
<br></blockquote></div><br></div></div>

--089e012281b6c47fc704dd532c93--

From bruce@callenish.com  Wed May 22 12:31:43 2013
Return-Path: <bruce@callenish.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1AAF111E8151 for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 12:31:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[AWL=-0.001, BAYES_00=-2.599, HTML_MESSAGE=0.001]
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 oRWJhB0lrVnq for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 12:31:37 -0700 (PDT)
Received: from biz82.inmotionhosting.com (biz82.inmotionhosting.com [66.117.4.147]) by ietfa.amsl.com (Postfix) with ESMTP id 4FB4811E8152 for <hybi@ietf.org>; Wed, 22 May 2013 12:31:36 -0700 (PDT)
Received: from [96.50.69.82] (port=50564 helo=[192.168.145.118]) by biz82.inmotionhosting.com with esmtpsa (TLSv1:DHE-RSA-AES256-SHA:256) (Exim 4.80) (envelope-from <bruce@callenish.com>) id 1UfEl3-0000xb-OB; Wed, 22 May 2013 12:31:34 -0700
Message-ID: <519D1D0E.5000905@callenish.com>
Date: Wed, 22 May 2013 12:31:26 -0700
From: Bruce Atherton <bruce@callenish.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: Zhong Yu <zhong.j.yu@gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <519D02E7.6040009@tavendo.de> <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com> <CACuKZqHYmF+-zECFrGsHqt0i91jHAzk9Az72=tva_kn=ra2pKA@mail.gmail.com>
In-Reply-To: <CACuKZqHYmF+-zECFrGsHqt0i91jHAzk9Az72=tva_kn=ra2pKA@mail.gmail.com>
Content-Type: multipart/alternative; boundary="------------010507050402080309080607"
X-OutGoing-Spam-Status: No, score=-2.9
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - biz82.inmotionhosting.com
X-AntiAbuse: Original Domain - ietf.org
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - callenish.com
X-Get-Message-Sender-Via: biz82.inmotionhosting.com: authenticated_id: bruce+callenish.com/only user confirmed/virtual account not confirmed
X-Source: 
X-Source-Args: 
X-Source-Dir: 
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 19:31:43 -0000

This is a multi-part message in MIME format.
--------------010507050402080309080607
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

On 22/05/2013 12:00 PM, Zhong Yu wrote:
>
> On Wed, May 22, 2013 at 1:32 PM, John A. Tamplin <jat@jaet.org 
> <mailto:jat@jaet.org>> wrote:
>
>     Servers don't handle tons of long-lived connections very well
>
>
> Is that still true today? Recent server implementations all boast they 
> can handle tons of connections.
>

Tons of short, HTTP Request/Response type connections. John is talking 
about long-lived WebSockets connections, the kind that might hang around 
for days, weeks, even months. It is a different problem.

Think of it this way. Say your server can handle 10K simultaneous 
connections to serve web pages and it takes about 330ms to serve them. 
If every one of those opens a websocket connection, you'll need 30K 
WebSocket connections just to handle a single second of traffic. If on 
average the page is kept open for only an hour, that is 108M connections.


--------------010507050402080309080607
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <div class="moz-cite-prefix">On 22/05/2013 12:00 PM, Zhong Yu wrote:<br>
    </div>
    <blockquote
cite="mid:CACuKZqHYmF+-zECFrGsHqt0i91jHAzk9Az72=tva_kn=ra2pKA@mail.gmail.com"
      type="cite">
      <div dir="ltr"><br>
        <div class="gmail_extra">
          <div class="gmail_quote">On Wed, May 22, 2013 at 1:32 PM, John
            A. Tamplin <span dir="ltr">&lt;<a moz-do-not-send="true"
                href="mailto:jat@jaet.org" target="_blank">jat@jaet.org</a>&gt;</span>
            wrote:<br>
            <div>&nbsp;</div>
            <blockquote class="gmail_quote" style="margin:0 0 0
              .8ex;border-left:1px #ccc solid;padding-left:1ex">
              <div dir="ltr">
                <div>Servers don't handle tons of long-lived connections
                  very well</div>
              </div>
            </blockquote>
            <div><br>
            </div>
            <div>Is that still true today? Recent server implementations
              all boast they can handle tons of connections. <br>
            </div>
            <br>
          </div>
        </div>
      </div>
    </blockquote>
    <br>
    Tons of short, HTTP Request/Response type connections. John is
    talking about long-lived WebSockets connections, the kind that might
    hang around for days, weeks, even months. It is a different problem.<br>
    <br>
    Think of it this way. Say your server can handle 10K simultaneous
    connections to serve web pages and it takes about 330ms to serve
    them. If every one of those opens a websocket connection, you'll
    need 30K WebSocket connections just to handle a single second of
    traffic. If on average the page is kept open for only an hour, that
    is 108M connections.<br>
    <br>
  </body>
</html>

--------------010507050402080309080607--

From simone.bordet@gmail.com  Wed May 22 13:41:35 2013
Return-Path: <simone.bordet@gmail.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 11F9721F961B for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 13:41:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.978
X-Spam-Level: 
X-Spam-Status: No, score=-1.978 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, NO_RELAYS=-0.001]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XbLNELK8vHPh for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 13:41:34 -0700 (PDT)
Received: from mail-pb0-x22b.google.com (mail-pb0-x22b.google.com [IPv6:2607:f8b0:400e:c01::22b]) by ietfa.amsl.com (Postfix) with ESMTP id 0FF9821F8CE9 for <hybi@ietf.org>; Wed, 22 May 2013 13:41:33 -0700 (PDT)
Received: by mail-pb0-f43.google.com with SMTP id ma3so2106943pbc.2 for <hybi@ietf.org>; Wed, 22 May 2013 13:41:29 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=dUSXo1P2AwKj17gKzwAtxd3x1dD+nPUoJerk6Jup1HU=; b=yKrPtfRhrgbl5kKdlOb97tqw5bRSfwuIAPCOq7oybOSarwuak94s3VfCyKhEyDzwNW 41UxtnvfFje2DEGZjoF7W+1JvtozCCVlqdpXCHNy/eBCIOAQXCQ3rsYJq5KaSJUrNXIg 0gakz8ApLffIMfukUZEborwNbaYGsWoYUAGUkIq0JPLu9t/nSAPnazlUdLvr6lBDGWjD 2rXDXnhe3HQb+CeILMnYA3ZcTgA4Z/biQ9Gk1Gr5+GAkGobPa8djMdHpwSApSgW08Uvu 55MmpXvO5nD8L/Yhr86LtCI3h8hrIV++NVGM1HWh8iT7SHHBBz9kgYD8T//mvJKAlSbb 73lA==
MIME-Version: 1.0
X-Received: by 10.66.165.73 with SMTP id yw9mr10200531pab.25.1369255288965; Wed, 22 May 2013 13:41:28 -0700 (PDT)
Sender: simone.bordet@gmail.com
Received: by 10.68.103.129 with HTTP; Wed, 22 May 2013 13:41:28 -0700 (PDT)
In-Reply-To: <519D1D0E.5000905@callenish.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <519D02E7.6040009@tavendo.de> <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com> <CACuKZqHYmF+-zECFrGsHqt0i91jHAzk9Az72=tva_kn=ra2pKA@mail.gmail.com> <519D1D0E.5000905@callenish.com>
Date: Wed, 22 May 2013 22:41:28 +0200
X-Google-Sender-Auth: 0PismNcTlzNM2n3UEYOSyoD1PiQ
Message-ID: <CAFWmRJ3-04eUqsme1mS-KTUir-KJagXnTDg7QWFzW6jquYuzmQ@mail.gmail.com>
From: Simone Bordet <sbordet@intalio.com>
To: Bruce Atherton <bruce@callenish.com>
Content-Type: text/plain; charset=UTF-8
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 20:41:35 -0000

Hi,

>>> On Wed, May 22, 2013 at 1:32 PM, John A. Tamplin <jat@jaet.org> wrote:
>>> Servers don't handle tons of long-lived connections very well
>>
>> On 22/05/2013 12:00 PM, Zhong Yu wrote:
>> Is that still true today? Recent server implementations all boast they can handle tons of connections.
>
>On Wed, May 22, 2013 at 9:31 PM, Bruce Atherton <bruce@callenish.com> wrote:
> Tons of short, HTTP Request/Response type connections. John is talking about long-lived WebSockets connections, the kind that might hang around for days, weeks, even months. It is a different problem.

Well I am not sure.
I have seen the TCP Linux stack collapsing due to usage of HTTP 1.0 by
a proxy. Opening and closing connections at a high rate it's a killer.
Long lived connections, on the other hand, not so problematic, at
least in my experience, so Zhong Yu's question still stand for me.

Another point not addressed by the article linked by Tobias is the
type of usage of the protocol.
The HTTP protocol has a typical 1 request followed by 10-200 requests
in a burst, that can be performed concurrently.
In a non-mux protocol (like HTTP/1.1) browsers open 6 connections, so
to serve the burst you need multiple "passes" at using those 6
connections, which leads to the well known ziggurat profile: 6 out,
pause, 6 out, pause, etc.
In SPDY, which is muxed, the 100 requests go out at once, with a flat profile.
Roberto Peon and us (Jetty) have showed demos where this effect is
*eye visible*. SPDY is faster than HTTP just because of mux
(http://webtide.intalio.com/2012/10/spdy-push-demo-from-javaone-2012/).

Now, HTTP has a semantic which benefits clearly from multiplexing.
Does this hold true for WebSocket, which does not have semantic, but
just framing ? I think it depends.
If you use WebSocket to transport HTTP (i.e. the opposite that
HTTP/2.0 is working on) then perhaps mux would be very important.
But if you use WebSocket for chat messaging, perhaps that's not that
important to have mux.
The difference here is between a request/response model versus a
messaging model.
The model changes the need for mux.

Since WebSocket does not define a model (just framing), perhaps mux is
less important.
Application protocols may implement mux if they need to.

> Think of it this way. Say your server can handle 10K simultaneous connections to serve web pages and it takes about 330ms to serve them.
> If every one of those opens a websocket connection, you'll need 30K WebSocket connections just to handle a single second of traffic.
> If on average the page is kept open for only an hour, that is 108M connections.

That's under the assumption that WebSocket connections are closed
after a "request", which is wrong.
WebSocket connections are persistent.
If you want to serve HTTP resources via WebSocket, you better open 1
connection per client (like SPDY does), and your server will only be
able to handle 10k clients at any time.
But then you'll need mux, or you go back to HTTP/1.1, just with a
different framing (WebSocket).

--
Simone Bordet
----
http://cometd.org
http://webtide.com
http://intalio.com
Developer advice, training, services and support
from the Jetty & CometD experts.
Intalio, the modern way to build business applications.

From tobias.oberstein@tavendo.de  Wed May 22 13:44:38 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9A4F811E8106 for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 13:44:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.284
X-Spam-Level: 
X-Spam-Status: No, score=-2.284 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, SARE_MILLIONSOF=0.315]
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 JAm6HKdKRuWK for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 13:44:33 -0700 (PDT)
Received: from EXHUB020-1.exch020.serverdata.net (exhub020-1.exch020.serverdata.net [206.225.164.28]) by ietfa.amsl.com (Postfix) with ESMTP id 57ABA21F961B for <hybi@ietf.org>; Wed, 22 May 2013 13:44:33 -0700 (PDT)
Received: from [192.168.178.39] (87.79.136.230) by smtpx20.serverdata.net (206.225.164.33) with Microsoft SMTP Server (TLS) id 8.3.264.0; Wed, 22 May 2013 13:44:32 -0700
Message-ID: <519D2E2D.4080809@tavendo.de>
Date: Wed, 22 May 2013 22:44:29 +0200
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: "John A. Tamplin" <jat@jaet.org>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <519D02E7.6040009@tavendo.de> <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com>
In-Reply-To: <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 20:44:38 -0000

Am 22.05.2013 20:32, schrieb John A. Tamplin:
> On Wed, May 22, 2013 at 1:39 PM, Tobias Oberstein
> <tobias.oberstein@tavendo.de <mailto:tobias.oberstein@tavendo.de>> wrote:
>
>     Regarding implementing flow control above TCP (and below app), I
>     found this:
>
>     "Multiplexing on top of TCP in overall fails to deliver the
>     advantages it is assumed to provide."
>
>     Source: http://www.250bpm.com/__multiplexing
>     <http://www.250bpm.com/multiplexing>
>
>     Do you have an opinion on that?
>
>
> One thing he doesn't address are resources used to keep lots of
> connections open, both at the OS level and in intermediate network nodes
> (such as load balancers -- try running tons of long-lived connections
> through an F5 for example).  Also, he doesn't address the interaction of

He does not address intermediaries, yes. But he does address memory 
consumption at endpoints: when running multiple TCPs between a given 
pair of endpoints, send/receive buffers reside in kernel space, whereas 
multiplexing over a single TCP requires send/receive buffers (per 
stream) in user space (to avoid TCP pushback). Hence he calls it a draw.

However, thinking about it: having send/receive buffers in kernel space 
requires one to decide how big buffers should be for the complete 
machine (TCP send / receive buffer size is usually a kernel tunable).

Having those buffers in user space (in app) allows one to tune the 
buffer size _per stream_. Thus, we can have large buffers for streams 
that carry mass data, and small buffers for chatty streams. Thats a real 
advantage.

Rgd load balancers: yep, L7 LBs will have a problem with massive numbers 
of connections. But L4 LBs that e.g. distribute connections to backend 
cluster nodes based on consistent hash of source IP not .. since they 
dont need to maintain state per connection. A L4 LB should be able to 
scale to millions of TCPs.

> TCP slow start with lots of new connections.  I also don't believe the
> equivalent of a full TCP handshake is required to allocate buffer space.

So to make the proposed advantage explicit: the per-stream handshaking 
that happens with mux consumes less resources than full TCP handshakes.

>
>     More important: what problem exactly does multiplexing solve? The
>     current mux RFC mentions "scalability", but doesn't go into detail ..
>
>
> Here is a use-case I envision:  Imagine GMail, G+ sandbar, etc all use
> WebSockets heavily.  Many users may have multiple copies of such apps
> running in tabs or windows in the browser, and some of these are going
> to be separate (such as GMail and the G+ sandbar probably are separate
> apps running on the same page and maintain their own connections to
> their backends).  Currently, that would mean you have a ton of TCP
> connections open to the server.  Servers don't handle tons of long-lived

How about having a shared Web worker in the background that handles all 
messaging for the app over a single WS?

> connections very well (or intermediate load balancers), so you would

Async servers running on top of modern APIs like epoll/kqueue/IOCP have 
no issue with massive numbers of long-lived connections. I have run 180k 
WebSocket connections on a single-threaded server running on a 2 core 
4GB Ram VM (with small TCP send/receive buffers configured).

> greatly prefer to have one connection per browser.  WS MUX lets you do
> that, by having the browser aggregate connections to the same backend.
>   You could do some of it at the app level, but it is going to be
> complicated and in many cases these apps won't all be from the same
> vendor (for example, some widget on the page that uses the GData API)
> and will have separate deploy schedules.

I agree, MUX at app level is probably less than ideal.

>
> --
> John A. Tamplin


From tobias.oberstein@tavendo.de  Wed May 22 13:49:09 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9877C11E80A2 for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 13:49:09 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.141
X-Spam-Level: 
X-Spam-Status: No, score=-2.141 tagged_above=-999 required=5 tests=[AWL=-0.142, BAYES_00=-2.599, J_CHICKENPOX_44=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 zkSHVLDvTOxu for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 13:49:05 -0700 (PDT)
Received: from EXHUB020-5.exch020.serverdata.net (exhub020-5.exch020.serverdata.net [206.225.164.32]) by ietfa.amsl.com (Postfix) with ESMTP id DBA4621F9195 for <hybi@ietf.org>; Wed, 22 May 2013 13:49:04 -0700 (PDT)
Received: from [192.168.178.39] (87.79.136.230) by smtpx20.serverdata.net (206.225.164.37) with Microsoft SMTP Server (TLS) id 8.3.264.0; Wed, 22 May 2013 13:49:03 -0700
Message-ID: <519D2F3E.3000609@tavendo.de>
Date: Wed, 22 May 2013 22:49:02 +0200
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: Simone Bordet <sbordet@intalio.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com>	<519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <CAFWmRJ2Hbe0x5FeV2T7Gkp3WEsxQHe2=YPBTgvHYLcus3A4rBQ@mail.gmail.com>
In-Reply-To: <CAFWmRJ2Hbe0x5FeV2T7Gkp3WEsxQHe2=YPBTgvHYLcus3A4rBQ@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 20:49:09 -0000

> So "simple mux" looks like an oxymoron to me, and makes me wonder what
> useful could be a ws mux solution that does not do dynamic channel
> prioritization.
> A typical use case Joakim presented to me was to tune video download
> rate in a websocket server: a big priority initially to download faster
> than the client video player plays, to make some buffer, and then
> reducing the priority to a steady state with the video player's speed,
> or reducing priority to almost zero if the user pauses the play.

Interesting. And yep, being able to dynamically tune the priority of a 
stream from the app seem desirable. The buffer/window size in this use 
case would probably be relatively large and fixed.

May I add another user experience centric use case where an app might 
want to tune both priority and buffer size?

Say I wanna run something like chat while also doing file transfers over 
WS. Different QoS. I don't want to increase the latency on the chatty 
stream by running a mass data stream. However, this probably requires 
both flow control _and_ message/frame priorization.

The current mux RFC reads:

" The sender MUST use a fair mechanism for selecting which logical
channel's data to send in the next WebSocket message.  Simple
implementations may choose a round-robin scheduler, while more
advanced implementations may adjust priority based on the amount
or frequency of data sent by each logical channel.
"

This seems to imply that the app has no control over QoS per stream 
(i.e. "chatty" vs "mass data").

IMHO the app should be able to control that (ultimately effecting buffer 
sizes and priorization/scheduling). If the app has no control chances 
are that people will invent something at the app level. IOW: this is 
something where MUX could really deliver added value (better UX), even 
if you are not a big G and concerned about infrastructure resource costs 
and world scale apps.

Do you think the current MUX addresses a chat+file transfer use case?

What about the proposed SPDY/HTTP2.0 multiplexing?

Are there even any plans for exposing tunables to apps?

/Tobias

>
> --
> Simone Bordet
> ----
> http://cometd.org
> http://webtide.com
> http://intalio.com
> Developer advice, training, services and support
> from the Jetty & CometD experts.
> Intalio, the modern way to build business applications.


From zhong.j.yu@gmail.com  Wed May 22 15:02:16 2013
Return-Path: <zhong.j.yu@gmail.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1E9C511E814C for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 15:02:16 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 HyNZrzxwXw5i for <hybi@ietfa.amsl.com>; Wed, 22 May 2013 15:02:14 -0700 (PDT)
Received: from mail-ob0-x230.google.com (mail-ob0-x230.google.com [IPv6:2607:f8b0:4003:c01::230]) by ietfa.amsl.com (Postfix) with ESMTP id BB25211E814B for <hybi@ietf.org>; Wed, 22 May 2013 15:02:14 -0700 (PDT)
Received: by mail-ob0-f176.google.com with SMTP id wp18so3080013obc.21 for <hybi@ietf.org>; Wed, 22 May 2013 15:02:14 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=16QBlhjjoxzTjlviujpEs6cVs2T4Bc8k2UQ2LWXiyMQ=; b=U2ITR7CehNeO8cqVqclgX0vKhmFl/c7iguddXTScnMzrzLexWVpsMiiPFObDRvojwl bIvgDKxZBz+fucLUaP5LCvx8yKWrcIN2jHM3e2E9RZv51uhBTVHQGY9YmDg1F+zROhIz g2pd1hiGgEbSFtQTzWbRLulMVjI/ucHBdKFnWAzajQNQarG3BvkOhhpJ4wIyO7M9K2OA omITBvJZRFw5FHZx3EpdxT510CVckEG/QU1jQDdxO/7KB71FP0D8vwn/8VEPCvZZQZys WcmyLH4TV1gC9Sz3CLCOVvtYRdkIkTXaRrTxQiNc5l4Tizp/wHEWO5yPZFKLv24xmgKC svXg==
MIME-Version: 1.0
X-Received: by 10.60.47.200 with SMTP id f8mr2949372oen.33.1369260134316; Wed, 22 May 2013 15:02:14 -0700 (PDT)
Received: by 10.76.98.227 with HTTP; Wed, 22 May 2013 15:02:14 -0700 (PDT)
In-Reply-To: <CAFWmRJ3-04eUqsme1mS-KTUir-KJagXnTDg7QWFzW6jquYuzmQ@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <519D02E7.6040009@tavendo.de> <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com> <CACuKZqHYmF+-zECFrGsHqt0i91jHAzk9Az72=tva_kn=ra2pKA@mail.gmail.com> <519D1D0E.5000905@callenish.com> <CAFWmRJ3-04eUqsme1mS-KTUir-KJagXnTDg7QWFzW6jquYuzmQ@mail.gmail.com>
Date: Wed, 22 May 2013 17:02:14 -0500
Message-ID: <CACuKZqEO7b4bgKV9pVktDhyuF=shrZqp8nrTy=KBQCOHFXZ+sw@mail.gmail.com>
From: Zhong Yu <zhong.j.yu@gmail.com>
To: Simone Bordet <sbordet@intalio.com>
Content-Type: multipart/alternative; boundary=001a11c2d682e944b804dd55b714
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 22 May 2013 22:02:16 -0000

--001a11c2d682e944b804dd55b714
Content-Type: text/plain; charset=ISO-8859-1

On Wed, May 22, 2013 at 3:41 PM, Simone Bordet <sbordet@intalio.com> wrote:

> Hi,
>
> >>> On Wed, May 22, 2013 at 1:32 PM, John A. Tamplin <jat@jaet.org> wrote:
> >>> Servers don't handle tons of long-lived connections very well
> >>
> >> On 22/05/2013 12:00 PM, Zhong Yu wrote:
> >> Is that still true today? Recent server implementations all boast they
> can handle tons of connections.
> >
> >On Wed, May 22, 2013 at 9:31 PM, Bruce Atherton <bruce@callenish.com>
> wrote:
> > Tons of short, HTTP Request/Response type connections. John is talking
> about long-lived WebSockets connections, the kind that might hang around
> for days, weeks, even months. It is a different problem.
>
> Well I am not sure.
> I have seen the TCP Linux stack collapsing due to usage of HTTP 1.0 by
> a proxy. Opening and closing connections at a high rate it's a killer.
> Long lived connections, on the other hand, not so problematic, at
> least in my experience, so Zhong Yu's question still stand for me.
>
> Another point not addressed by the article linked by Tobias is the
> type of usage of the protocol.
> The HTTP protocol has a typical 1 request followed by 10-200 requests
> in a burst, that can be performed concurrently.
> In a non-mux protocol (like HTTP/1.1) browsers open 6 connections, so
> to serve the burst you need multiple "passes" at using those 6
> connections, which leads to the well known ziggurat profile: 6 out,
> pause, 6 out, pause, etc.
> In SPDY, which is muxed, the 100 requests go out at once, with a flat
> profile.
> Roberto Peon and us (Jetty) have showed demos where this effect is
> *eye visible*. SPDY is faster than HTTP just because of mux
> (http://webtide.intalio.com/2012/10/spdy-push-demo-from-javaone-2012/).
>
> Now, HTTP has a semantic which benefits clearly from multiplexing.
> Does this hold true for WebSocket, which does not have semantic, but
> just framing ? I think it depends.
> If you use WebSocket to transport HTTP (i.e. the opposite that
> HTTP/2.0 is working on) then perhaps mux would be very important.
> But if you use WebSocket for chat messaging, perhaps that's not that
> important to have mux.
> The difference here is between a request/response model versus a
> messaging model.
> The model changes the need for mux.
>

I think the main difference is that, a WebSocket channel is a meaningful
identity that cannot be reduced, so there must be an overhead associated
with each live channel; mux cannot change that fact. On the other hand,
HTTP messages have no logical relationship to HTTP connections, it doesn't
matter which connection a message is delivered on, therefore it's possible
to reduce HTTP connections to remove overhead.

Zhong Yu


>
> Since WebSocket does not define a model (just framing), perhaps mux is
> less important.
> Application protocols may implement mux if they need to.
>
> > Think of it this way. Say your server can handle 10K simultaneous
> connections to serve web pages and it takes about 330ms to serve them.
> > If every one of those opens a websocket connection, you'll need 30K
> WebSocket connections just to handle a single second of traffic.
> > If on average the page is kept open for only an hour, that is 108M
> connections.
>
> That's under the assumption that WebSocket connections are closed
> after a "request", which is wrong.
> WebSocket connections are persistent.
> If you want to serve HTTP resources via WebSocket, you better open 1
> connection per client (like SPDY does), and your server will only be
> able to handle 10k clients at any time.
> But then you'll need mux, or you go back to HTTP/1.1, just with a
> different framing (WebSocket).
>
> --
> Simone Bordet
> ----
> http://cometd.org
> http://webtide.com
> http://intalio.com
> Developer advice, training, services and support
> from the Jetty & CometD experts.
> Intalio, the modern way to build business applications.
>

--001a11c2d682e944b804dd55b714
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><br><div class=3D"gmail_extra"><br><br><div class=3D"gmail=
_quote">On Wed, May 22, 2013 at 3:41 PM, Simone Bordet <span dir=3D"ltr">&l=
t;<a href=3D"mailto:sbordet@intalio.com" target=3D"_blank">sbordet@intalio.=
com</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">Hi,<br>
<div class=3D"im"><br>
&gt;&gt;&gt; On Wed, May 22, 2013 at 1:32 PM, John A. Tamplin &lt;<a href=
=3D"mailto:jat@jaet.org">jat@jaet.org</a>&gt; wrote:<br>
&gt;&gt;&gt; Servers don&#39;t handle tons of long-lived connections very w=
ell<br>
&gt;&gt;<br>
&gt;&gt; On 22/05/2013 12:00 PM, Zhong Yu wrote:<br>
&gt;&gt; Is that still true today? Recent server implementations all boast =
they can handle tons of connections.<br>
&gt;<br>
</div><div class=3D"im">&gt;On Wed, May 22, 2013 at 9:31 PM, Bruce Atherton=
 &lt;<a href=3D"mailto:bruce@callenish.com">bruce@callenish.com</a>&gt; wro=
te:<br>
&gt; Tons of short, HTTP Request/Response type connections. John is talking=
 about long-lived WebSockets connections, the kind that might hang around f=
or days, weeks, even months. It is a different problem.<br>
<br>
</div>Well I am not sure.<br>
I have seen the TCP Linux stack collapsing due to usage of HTTP 1.0 by<br>
a proxy. Opening and closing connections at a high rate it&#39;s a killer.<=
br>
Long lived connections, on the other hand, not so problematic, at<br>
least in my experience, so Zhong Yu&#39;s question still stand for me.<br>
<br>
Another point not addressed by the article linked by Tobias is the<br>
type of usage of the protocol.<br>
The HTTP protocol has a typical 1 request followed by 10-200 requests<br>
in a burst, that can be performed concurrently.<br>
In a non-mux protocol (like HTTP/1.1) browsers open 6 connections, so<br>
to serve the burst you need multiple &quot;passes&quot; at using those 6<br=
>
connections, which leads to the well known ziggurat profile: 6 out,<br>
pause, 6 out, pause, etc.<br>
In SPDY, which is muxed, the 100 requests go out at once, with a flat profi=
le.<br>
Roberto Peon and us (Jetty) have showed demos where this effect is<br>
*eye visible*. SPDY is faster than HTTP just because of mux<br>
(<a href=3D"http://webtide.intalio.com/2012/10/spdy-push-demo-from-javaone-=
2012/" target=3D"_blank">http://webtide.intalio.com/2012/10/spdy-push-demo-=
from-javaone-2012/</a>).<br>
<br>
Now, HTTP has a semantic which benefits clearly from multiplexing.<br>
Does this hold true for WebSocket, which does not have semantic, but<br>
just framing ? I think it depends.<br>
If you use WebSocket to transport HTTP (i.e. the opposite that<br>
HTTP/2.0 is working on) then perhaps mux would be very important.<br>
But if you use WebSocket for chat messaging, perhaps that&#39;s not that<br=
>
important to have mux.<br>
The difference here is between a request/response model versus a<br>
messaging model.<br>
The model changes the need for mux.<br></blockquote><div><br></div><div>I t=
hink the main difference is that, a WebSocket channel is a meaningful ident=
ity that cannot be reduced, so there must be an overhead associated with ea=
ch live channel; mux cannot change that fact. On the other hand, HTTP messa=
ges have no logical relationship to HTTP connections, it doesn&#39;t matter=
 which connection a message is delivered on, therefore it&#39;s possible to=
 reduce HTTP connections to remove overhead.<br>
<br></div><div>Zhong Yu<br></div><div>=A0</div><blockquote class=3D"gmail_q=
uote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1e=
x">
<br>
Since WebSocket does not define a model (just framing), perhaps mux is<br>
less important.<br>
Application protocols may implement mux if they need to.<br>
<div class=3D"im"><br>
&gt; Think of it this way. Say your server can handle 10K simultaneous conn=
ections to serve web pages and it takes about 330ms to serve them.<br>
&gt; If every one of those opens a websocket connection, you&#39;ll need 30=
K WebSocket connections just to handle a single second of traffic.<br>
&gt; If on average the page is kept open for only an hour, that is 108M con=
nections.<br>
<br>
</div>That&#39;s under the assumption that WebSocket connections are closed=
<br>
after a &quot;request&quot;, which is wrong.<br>
WebSocket connections are persistent.<br>
If you want to serve HTTP resources via WebSocket, you better open 1<br>
connection per client (like SPDY does), and your server will only be<br>
able to handle 10k clients at any time.<br>
But then you&#39;ll need mux, or you go back to HTTP/1.1, just with a<br>
different framing (WebSocket).<br>
<div class=3D"HOEnZb"><div class=3D"h5"><br>
--<br>
Simone Bordet<br>
----<br>
<a href=3D"http://cometd.org" target=3D"_blank">http://cometd.org</a><br>
<a href=3D"http://webtide.com" target=3D"_blank">http://webtide.com</a><br>
<a href=3D"http://intalio.com" target=3D"_blank">http://intalio.com</a><br>
Developer advice, training, services and support<br>
from the Jetty &amp; CometD experts.<br>
Intalio, the modern way to build business applications.<br>
</div></div></blockquote></div><br></div></div>

--001a11c2d682e944b804dd55b714--

From tobias.oberstein@tavendo.de  Thu May 23 01:29:20 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2710921F96A6 for <hybi@ietfa.amsl.com>; Thu, 23 May 2013 01:29:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[AWL=-0.001, BAYES_00=-2.599, HTML_MESSAGE=0.001]
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 DcVt05FHAS9k for <hybi@ietfa.amsl.com>; Thu, 23 May 2013 01:29:15 -0700 (PDT)
Received: from EXHUB020-2.exch020.serverdata.net (exhub020-2.exch020.serverdata.net [206.225.164.29]) by ietfa.amsl.com (Postfix) with ESMTP id 55AFF21F93FC for <hybi@ietf.org>; Thu, 23 May 2013 01:29:15 -0700 (PDT)
Received: from EXVMBX020-12.exch020.serverdata.net ([169.254.3.33]) by EXHUB020-2.exch020.serverdata.net ([206.225.164.29]) with mapi; Thu, 23 May 2013 01:29:14 -0700
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
To: "hybi@ietf.org" <hybi@ietf.org>
Date: Thu, 23 May 2013 01:29:12 -0700
Thread-Topic: WebSockets in Penetration Testing
Thread-Index: Ac5Xj5p3fZ4vuhWOQBqINOGZg8sc5Q==
Message-ID: <634914A010D0B943A035D226786325D4422C3D9D7B@EXVMBX020-12.exch020.serverdata.net>
References: <519DBB9C.3010400@loggia.at>
In-Reply-To: <519DBB9C.3010400@loggia.at>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: de-DE, en-US
Content-Type: multipart/alternative; boundary="_000_634914A010D0B943A035D226786325D4422C3D9D7BEXVMBX02012ex_"
MIME-Version: 1.0
Cc: Robert Koch <robert.koch@loggia.at>, "autobahnws@googlegroups.com" <autobahnws@googlegroups.com>
Subject: [hybi] WG: WebSockets in Penetration Testing
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 May 2013 08:29:20 -0000

--_000_634914A010D0B943A035D226786325D4422C3D9D7BEXVMBX02012ex_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

RT for protocol testing nerds: a master thesis and GSoC project about WS pe=
netration testing.

There is also an interesting chapter about WS usage and adoption in Android=
 apps ..


Von: Robert Koch [mailto:robert.koch@loggia.at]
Gesendet: Donnerstag, 23. Mai 2013 08:48
An: tobias.oberstein@gmail.com
Betreff: WebSockets in Penetration Testing

Hello Tobias,

my master's thesis is now available online:

 *   Info-Page: http://permalink.obvsg.at/AC07815487
 *   PDF: http://media.obvsg.at/p-AC07815487-2001
Feel free to share it as you like. Thanks for helping me out to understand =
the term "TCP-clean". My master's thesis is based on my Google Summer of Co=
de Project 2012 (http://code.google.com/p/zaproxy/).

Best wishes,
Robert

--_000_634914A010D0B943A035D226786325D4422C3D9D7BEXVMBX02012ex_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr=
osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:=
//www.w3.org/TR/REC-html40"><head><meta http-equiv=3DContent-Type content=
=3D"text/html; charset=3Dus-ascii"><meta name=3DGenerator content=3D"Micros=
oft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:Wingdings;
	panose-1:5 0 0 0 0 0 0 0 0 0;}
@font-face
	{font-family:Wingdings;
	panose-1:5 0 0 0 0 0 0 0 0 0;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0cm;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";
	color:black;}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
p.MsoAcetate, li.MsoAcetate, div.MsoAcetate
	{mso-style-priority:99;
	mso-style-link:"Sprechblasentext Zchn";
	margin:0cm;
	margin-bottom:.0001pt;
	font-size:8.0pt;
	font-family:"Tahoma","sans-serif";
	color:black;}
span.exldetailsdisplayval
	{mso-style-name:exldetailsdisplayval;}
span.E-MailFormatvorlage18
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
span.SprechblasentextZchn
	{mso-style-name:"Sprechblasentext Zchn";
	mso-style-priority:99;
	mso-style-link:Sprechblasentext;
	font-family:"Tahoma","sans-serif";
	color:black;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-size:10.0pt;}
@page WordSection1
	{size:612.0pt 792.0pt;
	margin:70.85pt 70.85pt 2.0cm 70.85pt;}
div.WordSection1
	{page:WordSection1;}
/* List Definitions */
@list l0
	{mso-list-id:662468840;
	mso-list-template-ids:2037945250;}
@list l0:level1
	{mso-level-number-format:bullet;
	mso-level-text:\F0B7;
	mso-level-tab-stop:36.0pt;
	mso-level-number-position:left;
	text-indent:-18.0pt;
	mso-ansi-font-size:10.0pt;
	font-family:Symbol;}
@list l0:level2
	{mso-level-number-format:bullet;
	mso-level-text:o;
	mso-level-tab-stop:72.0pt;
	mso-level-number-position:left;
	text-indent:-18.0pt;
	mso-ansi-font-size:10.0pt;
	font-family:"Courier New";
	mso-bidi-font-family:"Times New Roman";}
@list l0:level3
	{mso-level-number-format:bullet;
	mso-level-text:\F0A7;
	mso-level-tab-stop:108.0pt;
	mso-level-number-position:left;
	text-indent:-18.0pt;
	mso-ansi-font-size:10.0pt;
	font-family:Wingdings;}
@list l0:level4
	{mso-level-number-format:bullet;
	mso-level-text:\F0A7;
	mso-level-tab-stop:144.0pt;
	mso-level-number-position:left;
	text-indent:-18.0pt;
	mso-ansi-font-size:10.0pt;
	font-family:Wingdings;}
@list l0:level5
	{mso-level-number-format:bullet;
	mso-level-text:\F0A7;
	mso-level-tab-stop:180.0pt;
	mso-level-number-position:left;
	text-indent:-18.0pt;
	mso-ansi-font-size:10.0pt;
	font-family:Wingdings;}
@list l0:level6
	{mso-level-number-format:bullet;
	mso-level-text:\F0A7;
	mso-level-tab-stop:216.0pt;
	mso-level-number-position:left;
	text-indent:-18.0pt;
	mso-ansi-font-size:10.0pt;
	font-family:Wingdings;}
@list l0:level7
	{mso-level-number-format:bullet;
	mso-level-text:\F0A7;
	mso-level-tab-stop:252.0pt;
	mso-level-number-position:left;
	text-indent:-18.0pt;
	mso-ansi-font-size:10.0pt;
	font-family:Wingdings;}
@list l0:level8
	{mso-level-number-format:bullet;
	mso-level-text:\F0A7;
	mso-level-tab-stop:288.0pt;
	mso-level-number-position:left;
	text-indent:-18.0pt;
	mso-ansi-font-size:10.0pt;
	font-family:Wingdings;}
@list l0:level9
	{mso-level-number-format:bullet;
	mso-level-text:\F0A7;
	mso-level-tab-stop:324.0pt;
	mso-level-number-position:left;
	text-indent:-18.0pt;
	mso-ansi-font-size:10.0pt;
	font-family:Wingdings;}
ol
	{margin-bottom:0cm;}
ul
	{margin-bottom:0cm;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body bgcolor=3Dwhite lang=3DDE li=
nk=3Dblue vlink=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><sp=
an style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F49=
7D'>RT for protocol testing nerds: a master thesis and GSoC project about W=
S penetration testing.<o:p></o:p></span></p><p class=3DMsoNormal><span styl=
e=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D'><o:=
p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span style=3D'font-size:11.0=
pt;font-family:"Calibri","sans-serif";color:#1F497D'>There is also an inter=
esting chapter about WS usage and adoption in Android apps ..<o:p></o:p></s=
pan></p><p class=3DMsoNormal><span style=3D'font-size:11.0pt;font-family:"C=
alibri","sans-serif";color:#1F497D'><o:p>&nbsp;</o:p></span></p><p class=3D=
MsoNormal><span style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif=
";color:#1F497D'><o:p>&nbsp;</o:p></span></p><div style=3D'border:none;bord=
er-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt'><div><div style=3D'bord=
er:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm'><p class=
=3DMsoNormal><b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-=
serif";color:windowtext'>Von:</span></b><span style=3D'font-size:10.0pt;fon=
t-family:"Tahoma","sans-serif";color:windowtext'> Robert Koch [mailto:rober=
t.koch@loggia.at] <br><b>Gesendet:</b> Donnerstag, 23. Mai 2013 08:48<br><b=
>An:</b> tobias.oberstein@gmail.com<br><b>Betreff:</b> WebSockets in Penetr=
ation Testing<o:p></o:p></span></p></div></div><p class=3DMsoNormal><o:p>&n=
bsp;</o:p></p><p class=3DMsoNormal>Hello Tobias,<br><br>my master's thesis =
is now available online:<o:p></o:p></p><ul type=3Ddisc><li class=3DMsoNorma=
l style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto;mso-list:l0 l=
evel1 lfo1'>Info-Page: <span class=3Dexldetailsdisplayval><a href=3D"http:/=
/permalink.obvsg.at/AC07815487">http://permalink.obvsg.at/AC07815487</a> </=
span><o:p></o:p></li><li class=3DMsoNormal style=3D'mso-margin-top-alt:auto=
;mso-margin-bottom-alt:auto;mso-list:l0 level1 lfo1'>PDF: <a href=3D"http:/=
/media.obvsg.at/p-AC07815487-2001">http://media.obvsg.at/p-AC07815487-2001<=
/a><o:p></o:p></li></ul><p class=3DMsoNormal>Feel free to share it as you l=
ike. Thanks for helping me out to understand the term &quot;TCP-clean&quot;=
. My master's thesis is based on my Google Summer of Code Project 2012 (<a =
href=3D"http://code.google.com/p/zaproxy/">http://code.google.com/p/zaproxy=
/</a>).<br><br>Best wishes,<br>Robert<o:p></o:p></p></div></div></body></ht=
ml>=

--_000_634914A010D0B943A035D226786325D4422C3D9D7BEXVMBX02012ex_--

From SRS0+fd7b198b0fff3ef3=PI=noemax.com=arman@noemax.com  Thu May 23 02:09:46 2013
Return-Path: <SRS0+fd7b198b0fff3ef3=PI=noemax.com=arman@noemax.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5DE6D21F8FB3 for <hybi@ietfa.amsl.com>; Thu, 23 May 2013 02:09:46 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.448
X-Spam-Level: 
X-Spam-Status: No, score=-2.448 tagged_above=-999 required=5 tests=[AWL=0.150,  BAYES_00=-2.599, HTML_MESSAGE=0.001]
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 7TGQ3UF5KSFP for <hybi@ietfa.amsl.com>; Thu, 23 May 2013 02:09:42 -0700 (PDT)
Received: from mail.noemax.com (mail.noemax.com [74.208.113.37]) by ietfa.amsl.com (Postfix) with ESMTP id 2979B21F9709 for <hybi@ietf.org>; Thu, 23 May 2013 02:09:42 -0700 (PDT)
DKIM-Signature: a=rsa-sha1; t=1369300169; x=1369904969; s=m2048; d=noemax.com; c=relaxed/relaxed; v=1; bh=w8W+5FS5Y44bEFwgLskeee57HVQ=; h=From:Subject:Date:Message-ID:To:Cc:MIME-Version:Content-Type:In-Reply-To:References; b=qMyUwoMtnQ9R0IwFYM5OTDWdnyOtVQvDQBgY8m0q9RM9GHorocNhsRZwU72IxCUGwNN9L2Vb2UX5NjdDNmBxoqzdeqVLOoGtjqNxaqcMzWsYp1yZ3mdhRxnn25FJYAh102WTivYhiIqlzR32jGQDrvnZKMg8PWStTc4n5bspaUhrh6ZFTtP8vAX4KlvoHxjGUxbqHAK8EM+oFhovB0jUaxh4xzA/2Lg6VI6e5YxukE9yZyfD4XJmQTfIb/2+S1IHOpZIb9OtFO8dVBZTP2xjr3sDeNnzswXr/4erq16ZPVT4zPU1ZhljTRtFHniD7nsn1LHnuoYZUR8YNgxlfcAJqw==
Received: from mail.noemax.com by mail.noemax.com (Noemax Mail Server) with ASMTP (SSL) id 201305231209274099; Thu, 23 May 2013 12:09:27 +0300
From: "Arman Djusupov" <arman@noemax.com>
To: "'Joakim Erdfelt'" <joakim@intalio.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com>	<CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com>	<CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com>	<007501ce56f0$67f74080$37e5c180$@noemax.com> <CAG4zZZBpL3=szF6R80SHJeSTYig-bKe8cDPeJS=r2j4Z5oJ_NQ@mail.gmail.com>
In-Reply-To: <CAG4zZZBpL3=szF6R80SHJeSTYig-bKe8cDPeJS=r2j4Z5oJ_NQ@mail.gmail.com>
Date: Thu, 23 May 2013 12:09:30 +0300
Message-ID: <006201ce5795$3eec24d0$bcc46e70$@noemax.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_NextPart_000_0063_01CE57AE.643AE370"
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQGJbPE/yAcYUaa2Hw/8qx+KPIY2GAJfnsrjAhgTLgECNJFvhgKv1fonmVFvGdA=
Content-Language: en-us
Cc: hybi@ietf.org
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 23 May 2013 09:12:58 -0000

This is a multipart message in MIME format.

------=_NextPart_000_0063_01CE57AE.643AE370
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit

Agree on channel slot, header deltas, and simplified extension rules.

 

Concerning the header deltas we can simplify them by applying the
permessage-deflate compression on the entire control channel# 0. It will be
almost like SPDY header compression, only better as it will be applied on
the entire control channel payload. Since we have the permessage-deflate
spec almost ready why not put it to good use? Compressed or not compressed
the control channel could be a simple flag in mux settings.

 

With best regards,

Arman

 

From: Joakim Erdfelt [mailto:joakim@intalio.com] 
Sent: Wednesday, May 22, 2013 6:33 PM
To: Arman Djusupov
Cc: Takeshi Yoshino; hybi@ietf.org
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket

 

+1 for simplification.

 

I'll toss some topics onto the table now:

 * AddChannelRequest - eliminate header deltas

 * AddChannelResponse - eliminate header deltas

 * NewChannelSlot - eliminate the "channel slots" management entirely from
the spec, go back to the earlier, easier to implement, concepts from past
drafts.

 * Simplify the extension rules.  

     Extensions after initial physical connection are applied only to first
logical channel.

     in AddChannelRequest, extensions provided after the mux extension are
used for logical channels.

 

 




--

Joakim Erdfelt <joakim@intalio.com>

webtide.com <http://www.webtide.com/> 

Developer advice, services and support
from the Jetty & CometD experts

eclipse.org/jetty <http://eclipse.org/jetty/>  - cometd.org
<http://cometd.org/> 

 

On Wed, May 22, 2013 at 6:29 AM, Arman Djusupov <arman@noemax.com> wrote:

Sorry, just realized that my message was sent to Takeshi privately by
mistake a few days ago.. (wrong button). It was actually intended for the
list so I am repeating it publicly now:

 

 

Obviously continuing to develop a hybi-mux spec for another year or two
doesn't make sense. Taking into account that new specifications are coming
and that HTTP2.0 will probably provide multiplexing for both WebSocket and
HTTP traffic, at  this stage it would only be worth investing development
time in a "simple and ready to go" hybi-mux specification. Maybe if we
simplify the spec to rock bottom simplicity, even sacrificing some features
so that we can conclude it very fast, then maybe vendors might pick it up.
This way we will give to WebSocket its own mux for now, while the HTTPbis
draft might take much longer to form into the final specification. In any
case, hybi-mux must be reviewed and finalized in a very short time, plus it
should be simpler and easier to implement than both SDPY and HTTPbis based
multiplexing.

 

With best regards,

Arman

 

From: hybi-bounces@ietf.org [mailto:hybi-bounces@ietf.org] On Behalf Of
Takeshi Yoshino
Sent: Sunday, May 19, 2013 7:31 AM
To: hybi@ietf.org
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket

 

I got one comment about this.

 

He/she suggested that we finish the WebSocket dedicated multiplexing quickly
even by simplifying the spec much e.g. cutting some functionality in it that
would take much time to get consensus, and when HTTP/2.0 is matured, switch
to that.

 


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

 


------=_NextPart_000_0063_01CE57AE.643AE370
Content-Type: text/html;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" =
xmlns:o=3D"urn:schemas-microsoft-com:office:office" =
xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" =
xmlns=3D"http://www.w3.org/TR/REC-html40"><head><META =
HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii"><meta name=3DGenerator content=3D"Microsoft Word 14 =
(filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:blue;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:purple;
	text-decoration:underline;}
p.MsoAcetate, li.MsoAcetate, div.MsoAcetate
	{mso-style-priority:99;
	mso-style-link:"Balloon Text Char";
	margin:0in;
	margin-bottom:.0001pt;
	font-size:8.0pt;
	font-family:"Tahoma","sans-serif";}
span.EmailStyle17
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#1F497D;}
span.BalloonTextChar
	{mso-style-name:"Balloon Text Char";
	mso-style-priority:99;
	mso-style-link:"Balloon Text";
	font-family:"Tahoma","sans-serif";}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US link=3Dblue =
vlink=3Dpurple><div class=3DWordSection1><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Agree on channel slot, header deltas, and simplified extension =
rules.<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Concerning the header deltas we can simplify them by applying the =
permessage-deflate compression on the entire control channel# 0. It will =
be almost like SPDY header compression, only better as it will be =
applied on the entire control channel payload. Since we have the =
permessage-deflate spec almost ready why not put it to good use? =
Compressed or not compressed the control channel could be a simple flag =
in mux settings.<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>With best regards,<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Arman<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><b><span =
style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span>=
</b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'> =
Joakim Erdfelt [mailto:joakim@intalio.com] <br><b>Sent:</b> Wednesday, =
May 22, 2013 6:33 PM<br><b>To:</b> Arman Djusupov<br><b>Cc:</b> Takeshi =
Yoshino; hybi@ietf.org<br><b>Subject:</b> Re: [hybi] Call for interest: =
multiplexing dedicated for WebSocket<o:p></o:p></span></p><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p><div><p class=3DMsoNormal>+1 for =
simplification.<o:p></o:p></p><div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p =
class=3DMsoNormal>I'll toss some topics onto the table =
now:<o:p></o:p></p></div><div><p class=3DMsoNormal>&nbsp;* =
AddChannelRequest - eliminate header deltas<o:p></o:p></p></div><div><p =
class=3DMsoNormal>&nbsp;* AddChannelResponse - eliminate header =
deltas<o:p></o:p></p></div><div><p class=3DMsoNormal>&nbsp;* =
NewChannelSlot - eliminate the &quot;channel slots&quot; management =
entirely from the spec, go back to the earlier, easier to implement, =
concepts from past drafts.<o:p></o:p></p></div><div><p =
class=3DMsoNormal>&nbsp;* Simplify the extension rules. =
&nbsp;<o:p></o:p></p></div><div><p class=3DMsoNormal>&nbsp; &nbsp; =
&nbsp;Extensions after initial physical connection are applied only to =
first logical channel.<o:p></o:p></p></div><div><p =
class=3DMsoNormal>&nbsp; &nbsp; &nbsp;in AddChannelRequest, extensions =
provided after the mux extension are used for logical =
channels.<o:p></o:p></p></div><div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div></div><div><p =
class=3DMsoNormal><br clear=3Dall><o:p></o:p></p><div><div><p =
class=3DMsoNormal>--<o:p></o:p></p></div><div><p =
class=3DMsoNormal>Joakim Erdfelt &lt;<a =
href=3D"mailto:joakim@intalio.com" =
target=3D"_blank">joakim@intalio.com</a>&gt;<o:p></o:p></p></div><div><p =
class=3DMsoNormal><a href=3D"http://www.webtide.com/" =
target=3D"_blank">webtide.com</a><o:p></o:p></p></div><div><p =
class=3DMsoNormal>Developer advice, services and support<br>from the =
Jetty &amp; CometD experts<o:p></o:p></p></div><div><p =
class=3DMsoNormal><a href=3D"http://eclipse.org/jetty/" =
target=3D"_blank">eclipse.org/jetty</a>&nbsp;-&nbsp;<a =
href=3D"http://cometd.org/" =
target=3D"_blank">cometd.org</a><o:p></o:p></p></div></div><p =
class=3DMsoNormal =
style=3D'margin-bottom:12.0pt'><o:p>&nbsp;</o:p></p><div><p =
class=3DMsoNormal>On Wed, May 22, 2013 at 6:29 AM, Arman Djusupov &lt;<a =
href=3D"mailto:arman@noemax.com" =
target=3D"_blank">arman@noemax.com</a>&gt; =
wrote:<o:p></o:p></p><div><div><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Sorry, just realized that my message was sent to Takeshi privately by =
mistake a few days ago.. (wrong button). It was actually intended for =
the list so I am repeating it publicly now:</span><o:p></o:p></p><p =
class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Obviously continuing to develop a hybi-mux spec for another year or =
two doesn&#8217;t make sense. Taking into account that new =
specifications are coming and that HTTP2.0 will probably provide =
multiplexing for both WebSocket and HTTP traffic, at&nbsp; this stage it =
would only be worth investing development time in a &#8220;simple and =
ready to go&#8221; hybi-mux specification. Maybe if we simplify the spec =
to rock bottom simplicity, even sacrificing some features so that we can =
conclude it very fast, then maybe vendors might pick it up. This way we =
will give to WebSocket its own mux for now, while the HTTPbis draft =
might take much longer to form into the final specification. In any =
case, hybi-mux must be reviewed and finalized in a very short time, plus =
it should be simpler and easier to implement than both SDPY and HTTPbis =
based multiplexing.</span><o:p></o:p></p><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>With best regards,</span><o:p></o:p></p><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>Arman</span><o:p></o:p></p><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>&nbsp;</span><o:p></o:p></p><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'><b><span =
style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span>=
</b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'> =
<a href=3D"mailto:hybi-bounces@ietf.org" =
target=3D"_blank">hybi-bounces@ietf.org</a> [mailto:<a =
href=3D"mailto:hybi-bounces@ietf.org" =
target=3D"_blank">hybi-bounces@ietf.org</a>] <b>On Behalf Of </b>Takeshi =
Yoshino<br><b>Sent:</b> Sunday, May 19, 2013 7:31 AM<br><b>To:</b> <a =
href=3D"mailto:hybi@ietf.org" =
target=3D"_blank">hybi@ietf.org</a><br><b>Subject:</b> Re: [hybi] Call =
for interest: multiplexing dedicated for =
WebSocket</span><o:p></o:p></p><div><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p><=
/o:p></p><div><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>I got one =
comment about this.<o:p></o:p></p><div><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p><=
/o:p></p></div><div><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>He/she =
suggested that we finish the WebSocket dedicated multiplexing quickly =
even by simplifying the spec much e.g. cutting some functionality in it =
that would take much time to get consensus, and when HTTP/2.0 is =
matured, switch to that.<o:p></o:p></p><div><p class=3DMsoNormal =
style=3D'mso-margin-top-alt:auto;mso-margin-bottom-alt:auto'>&nbsp;<o:p><=
/o:p></p></div></div></div></div></div></div><p class=3DMsoNormal =
style=3D'margin-bottom:12.0pt'><br>______________________________________=
_________<br>hybi mailing list<br><a =
href=3D"mailto:hybi@ietf.org">hybi@ietf.org</a><br><a =
href=3D"https://www.ietf.org/mailman/listinfo/hybi" =
target=3D"_blank">https://www.ietf.org/mailman/listinfo/hybi</a><o:p></o:=
p></p></div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div></div></body></html>
------=_NextPart_000_0063_01CE57AE.643AE370--


From tyoshino@google.com  Mon May 27 00:30:19 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B297C21F91B1 for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 00:30:19 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.977
X-Spam-Level: 
X-Spam-Status: No, score=-1.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 GqA-CoouLp8c for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 00:30:19 -0700 (PDT)
Received: from mail-ea0-x22d.google.com (mail-ea0-x22d.google.com [IPv6:2a00:1450:4013:c01::22d]) by ietfa.amsl.com (Postfix) with ESMTP id 9562D21F9195 for <hybi@ietf.org>; Mon, 27 May 2013 00:30:18 -0700 (PDT)
Received: by mail-ea0-f173.google.com with SMTP id n15so3606392ead.18 for <hybi@ietf.org>; Mon, 27 May 2013 00:30:17 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=zM9Pi9dAqXsZoUdlEO/DDnUJv9jb1ZHpbBLd4NRWFMk=; b=jkuP355+cgczcbzwgVDWuT8AZhI+556qkgIVjBwkapUJ9CFkVrVdtwMMvnDsp0gxpx 6R1yoh1X0NrlUDFWEe2CwPxS8BFfQPWTlNOhLB3upyIi6TYITKMZqnG0LAu/eydxU0II MZBJ08zbeNUNkanFrigtPk911UiiiKc6xHPLnQjGqr0Tz+uVzd21/+UNvUBE1LXExvvy 2gf3/6WwcqxpEa7rGngJQcbG81375OwbxjEBBVuTgdFxBNCDEoOdQE9h5C25GuIh9CPk pWh3Xg7IvVPSuUMsScn9RrClqv245OKjcDCB4IUgf4sWcUq9WVdnMThg+QU4t1PIh5Gj VQew==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=zM9Pi9dAqXsZoUdlEO/DDnUJv9jb1ZHpbBLd4NRWFMk=; b=msu+hBSfXmf70YEbkWdd2pJfqPj7D+jwl+nWx6bFLzm1NXCSu0ZWjV4EdLUk3ZPiFJ j+A1hodtOeochn/jMC7LbnjUVDksKzCF8YQ6GOCkhI0T6K9G9KXxktVurWbvpBNIj1Qg GF5MJDucSd80/CJAxfuyn89GCW+LUPeEZ/uXVbBewUeCL46ZHGElvFrbZHLVbLCHreSO 0RcnMvoJUAIt35R/jo0GqMjLWQX4iHqOojhaLRmhk8lYieBSNEhbSqc2ZoAe/zq9EQJ0 R38zQTeU+09ogbrezfrHxAJghO0IZP2sWvYMCmljvw7x/AvyJM/8a649ivC0AqPw0Oab O92g==
X-Received: by 10.14.110.194 with SMTP id u42mr8983065eeg.117.1369639817690; Mon, 27 May 2013 00:30:17 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.86.67 with HTTP; Mon, 27 May 2013 00:29:57 -0700 (PDT)
In-Reply-To: <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Mon, 27 May 2013 16:29:57 +0900
Message-ID: <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com>
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
Content-Type: multipart/alternative; boundary=089e015384accdfff404ddae1e4a
X-Gm-Message-State: ALoCoQkBmYO9OwQWpFI9J7af+QRnG3mjCR/X0Yg3lyJCfc6FO0Jps1uiu4su53Ab42Wiycg/wYDI4LbQKtBwxG6Obq91fSe+llo2IvuM4jgOQbnewfEL/ofwg+OvZqy78zV9ZwJMehqaWHQXAdeM38QfL4rGqixlNd8yOpA+JDyQR1VmyPWGHyFRDJx0ZJPA6lAicCQRH4pH
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 May 2013 07:30:19 -0000

--089e015384accdfff404ddae1e4a
Content-Type: text/plain; charset=ISO-8859-1

On Wed, May 22, 2013 at 7:01 PM, Tobias Oberstein <
tobias.oberstein@tavendo.de> wrote:

> Sure, multiplexing at the WS level would allow to create generic
> concentrators, but a generic concentrator could not map incoming
> connections to backend boxes based on app data like UserID etc for sharding
> etc.
>

Right


> However, when running _different_ apps on the same device connecting to
> the same origin,


You meant same ws:// address?


> """
>
4.2.1. Frame mapping
>
>    One WebSocket frame is mapped into one proceeding SPDY HEADERS frame
>    and following plural SPDY data frames. The SPDY HEADERS frame must
>    contain WebSocket frame fields and following SPDY data frames contain
>    payload data. These data frames are free to be reframed.
>
>    This plan doesn't care about overheads.
> """
>
> This seems somehow straightforward, but introduces inefficiencies: the
> overhead per WS message is significantly increased and an additional
> framing layer that allows reframing costs CPU cycles.
>

Thanks


> Btw: assuming SPDY runs over TLS exclusively, is there any need for WS
> frame masking anymore?
>

See this post by Eric about BEAST attack. I think it's not easy to conclude
that mask is unncessary for wss.
http://www.educatedguesswork.org/2011/09/security_impact_of_the_rizzodu.html

--089e015384accdfff404ddae1e4a
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Wed, May 22, 2013 at 7:01 PM, Tobias Oberstein <span di=
r=3D"ltr">&lt;<a href=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_bla=
nk">tobias.oberstein@tavendo.de</a>&gt;</span> wrote:<br><div class=3D"gmai=
l_extra">


<div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margi=
n:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204=
);border-left-style:solid;padding-left:1ex">Sure, multiplexing at the WS le=
vel would allow to create generic concentrators, but a generic concentrator=
 could not map incoming connections to backend boxes based on app data like=
 UserID etc for sharding etc.<br>

</blockquote><div><br></div><div style>Right</div><div>=A0</div><blockquote=
 class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left-width:=
1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left=
:1ex">

However, when running _different_ apps on the same device connecting to the=
 same origin,</blockquote><div><br></div><div style>You meant same ws:// ad=
dress?</div><div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin=
:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204)=
;border-left-style:solid;padding-left:1ex">

&quot;&quot;&quot;<br></blockquote><blockquote class=3D"gmail_quote" style=
=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(20=
4,204,204);border-left-style:solid;padding-left:1ex">4.2.1. Frame mapping<b=
r>


<br>
=A0 =A0One WebSocket frame is mapped into one proceeding SPDY HEADERS frame=
<br>
=A0 =A0and following plural SPDY data frames. The SPDY HEADERS frame must<b=
r>
=A0 =A0contain WebSocket frame fields and following SPDY data frames contai=
n<br>
=A0 =A0payload data. These data frames are free to be reframed.<br>
<br>
=A0 =A0This plan doesn&#39;t care about overheads.<br>
&quot;&quot;&quot;<br>
<br>
This seems somehow straightforward, but introduces inefficiencies: the over=
head per WS message is significantly increased and an additional framing la=
yer that allows reframing costs CPU cycles.<br></blockquote><div><br></div>

<div style>Thanks</div><div>=A0</div><blockquote class=3D"gmail_quote" styl=
e=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(2=
04,204,204);border-left-style:solid;padding-left:1ex">
Btw: assuming SPDY runs over TLS exclusively, is there any need for WS fram=
e masking anymore?<br></blockquote><div><br></div><div>See this post by Eri=
c about BEAST attack. I think it&#39;s not easy to conclude that mask is un=
ncessary for wss.</div>

<div><a href=3D"http://www.educatedguesswork.org/2011/09/security_impact_of=
_the_rizzodu.html" target=3D"_blank">http://www.educatedguesswork.org/2011/=
09/security_impact_of_the_rizzodu.html</a><br>
</div><div>=A0</div></div></div></div>

--089e015384accdfff404ddae1e4a--

From tyoshino@google.com  Mon May 27 00:45:35 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1A40B21F94D0 for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 00:45:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.977
X-Spam-Level: 
X-Spam-Status: No, score=-1.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 kmZaX5EPQlSr for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 00:45:34 -0700 (PDT)
Received: from mail-wi0-x232.google.com (mail-wi0-x232.google.com [IPv6:2a00:1450:400c:c05::232]) by ietfa.amsl.com (Postfix) with ESMTP id 58B1A21F9446 for <hybi@ietf.org>; Mon, 27 May 2013 00:45:31 -0700 (PDT)
Received: by mail-wi0-f178.google.com with SMTP id hj6so1312066wib.5 for <hybi@ietf.org>; Mon, 27 May 2013 00:45:30 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=nexdGis9bKAh/o2NWonMt1U0s0ysDBiidX3xRYvi9Pk=; b=Y5UPKP9crDhSXDw6o+JcILSb3G8tEobNLUpKvY+l/mw7avgtuVMFXtOUlqkiO4cyvt PNn6XazJQdRoHD3xwrgh5RRPExdBi9kS+0YWUQL6AllHbID4lMPVcyQZZ0fYFjasz+Mb Ld+l3DVKrQuL9D6wzUZiSBPv0BPYRqS+K9bx21w0gQCxdpomtkk7I6iVmo2ELbQLcaqC 1IGzHygLcWpwSs0sEv4HPuz2X0feBhAfKQOvfqQzSjHLkSyG9Chu86htvaJR66qbxAqZ 7KZoN6MrHdWjVXYtz3GRh5ygitKZQyxEr4j9Pspsw5aLLs80yrdlWMITWe9DRzd3eVao o0tg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=nexdGis9bKAh/o2NWonMt1U0s0ysDBiidX3xRYvi9Pk=; b=mckd7ZXU1T8WSq6aPTXGj7MChSiSrR4+bI8BcyOA7pPPkN5ZMzcehICVCkPZl6PQ81 oTvO7Mszq+qpA2dKoklwOp2i5W7cQaslfYECvgwURoXQKDXoYFXIinQbfsVALQzOyWET xHTH2g46qvZg62E5kDRVlBVUTKtpOCIjsOLM+nBZVcWLn9bH/2cWJ8apkBbXXrE5UlMq beilPAYpFdgjxzSob51CImk4hbDsazyLSFaXqzbXaWLyzWk/dtHvBXpNo7Me8Eme+d49 zoumTtSWlc+xjjbKEIpy8aavlzjpIzFjTPHSqs1v6JnRuhpmWSM6JyPqWaK4K6A/uWj+ lwAQ==
X-Received: by 10.180.187.234 with SMTP id fv10mr7271694wic.29.1369640730353;  Mon, 27 May 2013 00:45:30 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.180.5.136 with HTTP; Mon, 27 May 2013 00:45:09 -0700 (PDT)
In-Reply-To: <CAG4zZZBpL3=szF6R80SHJeSTYig-bKe8cDPeJS=r2j4Z5oJ_NQ@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <CAG4zZZBpL3=szF6R80SHJeSTYig-bKe8cDPeJS=r2j4Z5oJ_NQ@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Mon, 27 May 2013 16:45:09 +0900
Message-ID: <CAH9hSJZS9U2CorrqQFCt10CD=RzQpOM-P-NywXSa3sFw_tdH9A@mail.gmail.com>
To: Joakim Erdfelt <joakim@intalio.com>
Content-Type: multipart/alternative; boundary=001a11c3848e33f64b04ddae5560
X-Gm-Message-State: ALoCoQmLVgXXDhY9JJWbWcR6hQwlQByDHnVikoyZIAJoVd/db/61U2chcCqewLg3VG8STsTo3VQ7nqHm5SsIwFLmH2V+tS6dbOT2Jd0MkTS22lpzcOaMkLrRH/23ZErzX6GhWzfSEn5gcYaUsMa5RC6Uv5mLMVGCGtyCXr4WTvVARyiidH0XiBiaAHpjAQR6Jr1DBzwknZcl
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 May 2013 07:45:35 -0000

--001a11c3848e33f64b04ddae5560
Content-Type: text/plain; charset=ISO-8859-1

On Thu, May 23, 2013 at 12:33 AM, Joakim Erdfelt <joakim@intalio.com> wrote:

>  * Simplify the extension rules.
>      Extensions after initial physical connection are applied only to
> first logical channel.
>      in AddChannelRequest, extensions provided after the mux extension are
> used for logical channels.
>

I'm fine with removing delta encoding though I don't think we need to stop
removing headers that don't make sense for logical channel.

      *  "Sec-WebSocket-Extensions" in an AddChannelRequest is applied to
the multiplexed connection.
      *  A client doesn't include the Upgrade and Sec-WebSocket-Accept
header as they don't make sense for multiplexed connection.

Do you prefer sending whole headers as if it's sent without mux? I agree
that we don't have to worry about any change on Upgrade and
Sec-WebSocket-Accept in the future if we send all as-is, so no strong
objection.

--001a11c3848e33f64b04ddae5560
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Thu, May 23, 2013 at 12:33 AM, Joakim Erdfelt <span dir=
=3D"ltr">&lt;<a href=3D"mailto:joakim@intalio.com" target=3D"_blank">joakim=
@intalio.com</a>&gt;</span> wrote:<br><div class=3D"gmail_extra"><div class=
=3D"gmail_quote">

<blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-=
left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;p=
adding-left:1ex"><div dir=3D"ltr"><div>=A0* Simplify the extension rules. =
=A0</div>


<div>=A0 =A0 =A0Extensions after initial physical connection are applied on=
ly to first logical channel.</div><div>=A0 =A0 =A0in AddChannelRequest, ext=
ensions provided after the mux extension are used for logical channels.</di=
v></div>

</blockquote><div><br></div><div style>I&#39;m fine with removing delta enc=
oding though I don&#39;t think we need to stop removing headers that don&#3=
9;t make sense for logical channel.<br></div><div style><br></div><div styl=
e>

<div>=A0 =A0 =A0 * =A0&quot;Sec-WebSocket-Extensions&quot; in an AddChannel=
Request is applied to the multiplexed connection.</div><div>=A0 =A0 =A0 * =
=A0A client doesn&#39;t include the Upgrade and Sec-WebSocket-Accept header=
 as they don&#39;t make sense for multiplexed connection.<br>

</div><div><br></div><div><div>Do you prefer sending whole headers as if it=
&#39;s sent without mux? I agree that we don&#39;t have to worry about any =
change on Upgrade and Sec-WebSocket-Accept in the future if we send all as-=
is, so no strong objection.</div>

</div><div>=A0</div></div></div></div></div>

--001a11c3848e33f64b04ddae5560--

From tyoshino@google.com  Mon May 27 01:26:08 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CD7C421F8C1A for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 01:26:07 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.977
X-Spam-Level: 
X-Spam-Status: No, score=-1.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 MmNMEfC1QAuw for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 01:26:07 -0700 (PDT)
Received: from mail-wi0-x22e.google.com (mail-wi0-x22e.google.com [IPv6:2a00:1450:400c:c05::22e]) by ietfa.amsl.com (Postfix) with ESMTP id E626E21F9371 for <hybi@ietf.org>; Mon, 27 May 2013 01:25:38 -0700 (PDT)
Received: by mail-wi0-f174.google.com with SMTP id c10so1346616wiw.13 for <hybi@ietf.org>; Mon, 27 May 2013 01:25:38 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=xVFCHKO75zM1Iw/Sb3hQJsmKWDJLqj6fMiIHzKGyhOQ=; b=hRPrOsPToaK2Nr26q+lDvV4rYW+mvbjGIax6epZYsYcjuPLYMrRliZWMlD1ZTPYV/b GXaPpZ6CiL1CAfrn1mAlelEGG7yyQbMAnT9/qlVkHLyygJ2D5naoxuDfDf1IVfuLYmvc si5sKb1iaCFWgeKvBqt2wxEe6F8RMnmc43gQlD0mkFn5/RWope9jq8QXftlKL/O7rdy4 vzS3H6daTTtsJcmh/TKXzh1ZJgwWU795xJAzAZSDo/XgNlhIgAJSHl4A0rYMHHF3s1H0 3BTeWM8ng8Xi0dapF++nMZZJU43Wu5PATHlt2FZGuRMXAi5PLjUuMXfe84uIjA78QcrS m6Rw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=xVFCHKO75zM1Iw/Sb3hQJsmKWDJLqj6fMiIHzKGyhOQ=; b=neZpnU6GoYmYpIuUHIhbD3J9frnLjCoSp3FYNpEtmFrrdVUl2BpPLELq1937qru9Qq sZp3VSDJqQoKmLayqnaClJUS0hc9nZKRqhq/VyFNSXjaeGsZFM9J3zBcp0pZoE0XtqVB KSXxZxczz/86Ci3o4HBzoBVdQ5RfiEO/qgFZ+p6Xji6iZY2PFngrJdlwVgtpZXJjwrHN SyCMKhOAlT7afumzkQVZNqrXv2bEPC12WERuvLEyHNmDwqbIZczhnnhfK/nz52yOWV1Q unVIKdu+rXKj7P8tZhv6UaTEwUvN0zmX+W8otZ+l3NU9xN2SdK2CMREgqT/w75TRkcgW gieg==
X-Received: by 10.180.107.35 with SMTP id gz3mr7582201wib.0.1369643138136; Mon, 27 May 2013 01:25:38 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.180.5.136 with HTTP; Mon, 27 May 2013 01:25:18 -0700 (PDT)
In-Reply-To: <CAFWmRJ2Hbe0x5FeV2T7Gkp3WEsxQHe2=YPBTgvHYLcus3A4rBQ@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <CAFWmRJ2Hbe0x5FeV2T7Gkp3WEsxQHe2=YPBTgvHYLcus3A4rBQ@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Mon, 27 May 2013 17:25:18 +0900
Message-ID: <CAH9hSJYOPvsFPDXLOa29ASd8xavLdvfRK_cVd=Uc=Vaydz1O=w@mail.gmail.com>
To: Simone Bordet <sbordet@intalio.com>
Content-Type: multipart/alternative; boundary=e89a8f3b9b69b7d45204ddaee481
X-Gm-Message-State: ALoCoQnsXjZwUX1l16Yz9h/6/R10O5Rr/mMqd0/iNCdaA2+YXi0MuSxakX6xAIJ+855ntb5g2hXyDjT48nHO49m/LK4Si3xtWZZsYzpgXMIs+OISf6kQVn+1inMdrJqxWL7zdihbvSpxyBmgyazHI0BptAjHdmVDti3oOnGTtFHiJF7Ht4oE+Fr+ui6u64+V3gwePDg0xX1n
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 May 2013 08:26:08 -0000

--e89a8f3b9b69b7d45204ddaee481
Content-Type: text/plain; charset=ISO-8859-1

On Thu, May 23, 2013 at 1:48 AM, Simone Bordet <sbordet@intalio.com> wrote:

> On Wed, May 22, 2013 at 6:26 PM, John A. Tamplin <jat@jaet.org> wrote:
> > I agree -- without flow control, there is very little benefit, and almost
> > nothing that an app couldn't easily do themselves by just adding a simple
> > framing layer inside WS messages (perhaps just the first byte is a
> logical
> > channel number).
>
> In SPDY, the flow control had an initial window of 64k. This proved way
> too small for uploads/downloads, which resulted in a 3x or more slowdown.
> So there needs to be in ws a mechanism to configure this window (which
> SPDY has).
>

NewChannelSlot can do this. I don't think adding SETTINGS is simpler than
NewChannelSlot.

Instead of dropping NewChannelSlot, I'd rather suggest dropping "N mux
control in 1 message" format, "Handshake size" field in AddChannelRequest,
AddChannelResponse, "Reason size" in DropChannel.


> So "simple mux" looks like an oxymoron to me, and makes me wonder what
> useful could be a ws mux solution that does not do dynamic channel
> prioritization.
>
A typical use case Joakim presented to me was to tune video download rate
> in a websocket server: a big priority initially to download faster than the
> client video player plays, to make some buffer, and then reducing the
> priority to a steady state with the video player's speed, or reducing
> priority to almost zero if the user pauses the play.
>

Isn't it in-app priority? Can't script and server address that problem? If
only the client knows current priority, the client can tell the priority to
the peer servlet using a WebSocket message. If the server know priority for
each position, it can just control transmission speed.

Even if we need help by the server, the servlet would call some special
function to tell the server about priority.

Different from HTTP/2.0's case, we don't need to provide a special field in
mux protocol to pass priority, I think.

--e89a8f3b9b69b7d45204ddaee481
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Thu, May 23, 2013 at 1:48 AM, Simone Bordet <span dir=
=3D"ltr">&lt;<a href=3D"mailto:sbordet@intalio.com" target=3D"_blank">sbord=
et@intalio.com</a>&gt;</span> wrote:<br><div class=3D"gmail_extra"><div cla=
ss=3D"gmail_quote">

<blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-=
left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;p=
adding-left:1ex"><div dir=3D"ltr"><div class=3D"im">On Wed, May 22, 2013 at=
 6:26 PM, John A. Tamplin &lt;<a href=3D"mailto:jat@jaet.org" target=3D"_bl=
ank">jat@jaet.org</a>&gt; wrote:<br>

&gt; I agree -- without flow control, there is very little benefit, and alm=
ost<br>
&gt; nothing that an app couldn&#39;t easily do themselves by just adding a=
 simple<br>&gt; framing layer inside WS messages (perhaps just the first by=
te is a logical<br>&gt; channel number).<br><br></div>In SPDY, the flow con=
trol had an initial window of 64k. This proved way too small for uploads/do=
wnloads, which resulted in a 3x or more slowdown.<br>


So there needs to be in ws a mechanism to configure this window (which SPDY=
 has).<br></div></blockquote><div><br></div><div style>NewChannelSlot can d=
o this. I don&#39;t think adding SETTINGS is simpler than NewChannelSlot.</=
div>

<div style><br></div><div style>Instead of dropping NewChannelSlot, I&#39;d=
 rather suggest dropping &quot;N mux control in 1 message&quot; format, &qu=
ot;Handshake size&quot; field in AddChannelRequest, AddChannelResponse, &qu=
ot;Reason size&quot; in DropChannel.</div>

<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px=
 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left=
-style:solid;padding-left:1ex"><div dir=3D"ltr">So &quot;simple mux&quot; l=
ooks like an oxymoron to me, and makes me wonder what useful could be a ws =
mux solution that does not do dynamic channel prioritization.<br>

</div></blockquote><blockquote class=3D"gmail_quote" style=3D"margin:0px 0p=
x 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border=
-left-style:solid;padding-left:1ex"><div dir=3D"ltr">
A typical use case Joakim presented to me was to tune video download rate i=
n a websocket server: a big priority initially to download faster than the =
client video player plays, to make some buffer, and then reducing the prior=
ity to a steady state with the video player&#39;s speed, or reducing priori=
ty to almost zero if the user pauses the play.</div>

</blockquote><div><br></div><div style>Isn&#39;t it in-app priority? Can&#3=
9;t script and server address that problem? If only the client knows curren=
t priority, the client can tell the priority to the peer servlet using a We=
bSocket message. If the server know priority for each position, it can just=
 control transmission speed.</div>

<div style><br></div><div style>Even if we need help by the server, the ser=
vlet would call some special function to tell the server about priority.</d=
iv><div style><br></div><div style>Different from HTTP/2.0&#39;s case, we d=
on&#39;t need to provide a special field in mux protocol to pass priority, =
I think.</div>

<div style>=A0<br></div></div></div></div>

--e89a8f3b9b69b7d45204ddaee481--

From tyoshino@google.com  Mon May 27 01:34:11 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 125E221F8D69 for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 01:34:11 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.977
X-Spam-Level: 
X-Spam-Status: No, score=-1.977 tagged_above=-999 required=5 tests=[AWL=-0.000, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 HX-h7iHOxkVa for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 01:34:10 -0700 (PDT)
Received: from mail-wg0-x22d.google.com (mail-wg0-x22d.google.com [IPv6:2a00:1450:400c:c00::22d]) by ietfa.amsl.com (Postfix) with ESMTP id 6B61321F8C03 for <hybi@ietf.org>; Mon, 27 May 2013 01:34:10 -0700 (PDT)
Received: by mail-wg0-f45.google.com with SMTP id n12so4316144wgh.12 for <hybi@ietf.org>; Mon, 27 May 2013 01:34:09 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=VAi/AVeHfbe8YP9RuC+mi1GyIwUYgbP99XxL8tU8a4I=; b=VNGBv/MT2aX6F7jld1eD6jv/YnwBfsB/wBIQqowKQN3wlykuh+82jbARWbjtstw1kx HDhHs4plhHBGMvaIU2XH0CzAaB4lyWqrrX4P0JpbJnCYy9TvO4+wAAQkkt7LWBOH3Dbq p71GNNQuXs3h2xnvFzh/+jocyMRbLR1L94x2YWGnDuV+dK5Ovt6hylkfs0GSzx/mj0Pv smhO+qQ2YUynpoFLocnJml7CIQvyaU/LF8BRQF01SMfgzsuRfIcQsDdyNfUvUSxIDw/V Yv9t1UfgIV3Q9aJGyJ4zy8PnrxdGiGAP4JUrksYj4eelk76b3dScjvp91LF4OjdqFycb xI5g==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=VAi/AVeHfbe8YP9RuC+mi1GyIwUYgbP99XxL8tU8a4I=; b=T4vw2NgnXuLBwwxumCeG/iaQtUuqEW5CklAKfTbXL643lYDbSOo73IBl+9yOPhCxch 13XocQOVP07P2PooDQdyrJVwmaNoF8oFbLPnRKx79I1r0K7uSznq7yTKrPCU2u9YGG/X zAcyVfFR8aj6Ii5nMKMKXPlgtnSZ67LCoK4CCoEi0klwTLJ829fBOg3kE486WNBCjl9K fKMn7jgjsT/Gu7bm+Qzz4NVbT81Fzz8HIQrq7NC4ICA1MedBlQQgqfa7g/rtv/k5op23 7LgUOEbIMGoOrHajFffQU7s2EsBSL/8HiTCQSE/qGDEnBox8Mg/XO0riI0U75L3d1mCs pbUA==
X-Received: by 10.180.185.244 with SMTP id ff20mr7656145wic.0.1369643649419; Mon, 27 May 2013 01:34:09 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.180.5.136 with HTTP; Mon, 27 May 2013 01:33:49 -0700 (PDT)
In-Reply-To: <CACuKZqEXghX9ONECBGTEB1cD5Pt5cakKih9unjSZAmFnHZ1f2A@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <CACuKZqEXghX9ONECBGTEB1cD5Pt5cakKih9unjSZAmFnHZ1f2A@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Mon, 27 May 2013 17:33:49 +0900
Message-ID: <CAH9hSJbqwem9SHrMrOZvfRTnoDCEmWhdCLGr8gewjAY3BGP9pg@mail.gmail.com>
To: Zhong Yu <zhong.j.yu@gmail.com>
Content-Type: multipart/alternative; boundary=001a11c354ca31617104ddaf0320
X-Gm-Message-State: ALoCoQk6WxeVYJlz6udFCR4jztMb/biWTgiTF0iDlFMcRhTWOP/rH01Nbpy5M2j/+RvYZlBuSK9F+B5Fu+HyojxRkdpm/4c1QK5JhXq67Haqiyw9fvVdbf7O/Wkm+OH3j2Df2lXnK4/Oxj3uaRM69TCoIigCcmAw4KJpq/Ui1qngqMqWghDpNssP653RYxz47O1hOU/S8Hwc
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 May 2013 08:34:11 -0000

--001a11c354ca31617104ddaf0320
Content-Type: text/plain; charset=ISO-8859-1

On Thu, May 23, 2013 at 2:13 AM, Zhong Yu <zhong.j.yu@gmail.com> wrote:

> Contrary to that, most WebSocket API are event based - all incoming
> message data are buffered without any blocking, then the complete message
> is sent to the application callback. In this model, per channel flow
> control doesn't seems to exist.
>

It's possible that one JavaScript takes longer time to process each message
than another.

--001a11c354ca31617104ddaf0320
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Thu, May 23, 2013 at 2:13 AM, Zhong Yu <span dir=3D"ltr=
">&lt;<a href=3D"mailto:zhong.j.yu@gmail.com" target=3D"_blank">zhong.j.yu@=
gmail.com</a>&gt;</span> wrote:<br><div class=3D"gmail_extra"><div class=3D=
"gmail_quote">

<blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-=
left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;p=
adding-left:1ex"><div dir=3D"ltr"><div><div class=3D"h5"><div class=3D"gmai=
l_extra">

<span style=3D"color:rgb(34,34,34)">Contrary to that, most WebSocket API ar=
e event based - all incoming message data are buffered without any blocking=
, then the complete message is sent to the application callback. In this mo=
del, per channel flow control doesn&#39;t seems to exist.</span></div>

</div></div></div></blockquote><div><br></div><div style=3D"text-align:left=
">It&#39;s possible that one JavaScript takes longer time to process each m=
essage than another.</div><div>=A0</div></div></div></div>

--001a11c354ca31617104ddaf0320--

From tyoshino@google.com  Mon May 27 01:35:03 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9F4FD21F8C03 for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 01:35:03 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.977
X-Spam-Level: 
X-Spam-Status: No, score=-1.977 tagged_above=-999 required=5 tests=[AWL=-0.000, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 CH8mY32Regqc for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 01:35:03 -0700 (PDT)
Received: from mail-wi0-x232.google.com (mail-wi0-x232.google.com [IPv6:2a00:1450:400c:c05::232]) by ietfa.amsl.com (Postfix) with ESMTP id BD3E121F89D5 for <hybi@ietf.org>; Mon, 27 May 2013 01:35:02 -0700 (PDT)
Received: by mail-wi0-f178.google.com with SMTP id hj6so1349084wib.11 for <hybi@ietf.org>; Mon, 27 May 2013 01:35:02 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=HTul8iR8c+Dht2j8/IQVhRYOKzKn/EPsRf8sr1ocQLc=; b=cRJJz2WIYuUC4CSyVTCrAldiuXto1n3XCECdA2UJ7n97cNp73DF57iW4ZCx5um2Nd2 kr85moZT0/3OeoIZ+er+/J09vPszQfAWSHFIyhxs7tBNw1iI6v91BTNZ7im0VHN0wZV/ H+/gJaJgHavrZ2uHGVeBH8YriZbK6rXCnFi0tfA8DNpX0Bes9VW+k/y5wvVNl+uc4x+Z YLgpDnyBCRRuy+y08gA9rPbZfgfCGISEHLhMW9hYHq8wxo8Cqq8c3/d4Y093o2aKSHnz 9FVszapYKqgDNDBF2GX3YaTJhDbJDkdmcOwB/uOLLH9MhVdTLIu9aYKXIbT5TviJ3U5G 3A1Q==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=HTul8iR8c+Dht2j8/IQVhRYOKzKn/EPsRf8sr1ocQLc=; b=lLDNyqBltfroAqXPpqqkQfZk1woEUg9IIAjTi3JbbJZw6sVRMzruvWBxugpZWH+Vdm d7aOVCaOIQVvKgFenxJWcd5uE9oPPKmyyJa4g2141G4KpXH0Eq92M0ph+RaalYHjFfCr jA4EwgRQjPSd/2YQgm37MEBLv6wTj3u5w0PbEAf5vxTWe4yUbILu5A6i4MClelvHObxz W1BnvBHT6An1000o0xOlyfSQMPDTNaUD1Ao1PkLnczj9k80xOgm8t8mB03MJ69hLS4Qa UoYdZktPCAPDxrYs6dqYok8JPAlV7f2shUycu4qR34PJgwnUrYjQNvQJ9SfUPTudN2sC XpZQ==
X-Received: by 10.180.184.112 with SMTP id et16mr7471050wic.58.1369643701966;  Mon, 27 May 2013 01:35:01 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.180.5.136 with HTTP; Mon, 27 May 2013 01:34:41 -0700 (PDT)
In-Reply-To: <519D2F3E.3000609@tavendo.de>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <CAFWmRJ2Hbe0x5FeV2T7Gkp3WEsxQHe2=YPBTgvHYLcus3A4rBQ@mail.gmail.com> <519D2F3E.3000609@tavendo.de>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Mon, 27 May 2013 17:34:41 +0900
Message-ID: <CAH9hSJbRc1E1+nbFQ5su+R=0ya6AeOuoK-i5Ud1=4dsV63Vr8w@mail.gmail.com>
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
Content-Type: multipart/alternative; boundary=001a11c2269a532d8504ddaf06eb
X-Gm-Message-State: ALoCoQnxlyVeLEncMf6lRZM8MVMcwmdfxg4zg9U/GElpFZxyoduf6ikwud2zXiFI+OqNoHdDUr8uSe7QP2sficQ//+uhf8/OYYmklzQU5R6QuOO7vVxn1kH98yVxL+20UKctGPojzGpzMQCXqTbOULLcnN36Q+74sIwNrZ8+MXuFeU4cqzaA8Gr8JTz3/Ie3wPd9TMwPLH5Z
Cc: Simone Bordet <sbordet@intalio.com>, "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 May 2013 08:35:03 -0000

--001a11c2269a532d8504ddaf06eb
Content-Type: text/plain; charset=ISO-8859-1

On Thu, May 23, 2013 at 5:49 AM, Tobias Oberstein <
tobias.oberstein@tavendo.de> wrote:

> The current mux RFC reads:
>
> " The sender MUST use a fair mechanism for selecting which logical
> channel's data to send in the next WebSocket message.  Simple
> implementations may choose a round-robin scheduler, while more
> advanced implementations may adjust priority based on the amount
> or frequency of data sent by each logical channel.
> "
>
> This seems to imply that the app has no control over QoS per stream (i.e.
> "chatty" vs "mass data").
>
> IMHO the app should be able to control that (ultimately effecting buffer
> sizes and priorization/scheduling). If the app has no control chances are
> that people will invent something at the app level. IOW: this is something
> where MUX could really deliver added value (better UX), even if you are not
> a big G and concerned about infrastructure resource costs and world scale
> apps.
>

I'll fix the text. It was written considering only environments that runs
scripts without any knowledge about them such as importance, dependency,
etc.

--001a11c2269a532d8504ddaf06eb
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Thu, May 23, 2013 at 5:49 AM, Tobias Oberstein <span di=
r=3D"ltr">&lt;<a href=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_bla=
nk">tobias.oberstein@tavendo.de</a>&gt;</span> wrote:<div class=3D"gmail_ex=
tra"><div class=3D"gmail_quote">

<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div class=3D"im"><span style=3D"color:rgb(3=
4,34,34)">The current mux RFC reads:</span><br></div>
<br>
&quot; The sender MUST use a fair mechanism for selecting which logical<br>
channel&#39;s data to send in the next WebSocket message. =A0Simple<br>
implementations may choose a round-robin scheduler, while more<br>
advanced implementations may adjust priority based on the amount<br>
or frequency of data sent by each logical channel.<br>
&quot;<br>
<br>
This seems to imply that the app has no control over QoS per stream (i.e. &=
quot;chatty&quot; vs &quot;mass data&quot;).<br>
<br>
IMHO the app should be able to control that (ultimately effecting buffer si=
zes and priorization/scheduling). If the app has no control chances are tha=
t people will invent something at the app level. IOW: this is something whe=
re MUX could really deliver added value (better UX), even if you are not a =
big G and concerned about infrastructure resource costs and world scale app=
s.<br>

</blockquote><div><br></div><div style>I&#39;ll fix the text. It was writte=
n considering only environments that runs scripts without any knowledge abo=
ut them such as importance, dependency, etc.</div><div>=A0</div></div></div=
>

</div>

--001a11c2269a532d8504ddaf06eb--

From simone.bordet@gmail.com  Mon May 27 01:47:36 2013
Return-Path: <simone.bordet@gmail.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EB5BF21F8FAF for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 01:47:36 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.977
X-Spam-Level: 
X-Spam-Status: No, score=-1.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622]
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 ZfRLAtdEJwBG for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 01:47:33 -0700 (PDT)
Received: from mail-pd0-f170.google.com (mail-pd0-f170.google.com [209.85.192.170]) by ietfa.amsl.com (Postfix) with ESMTP id 0E76721F8F5D for <hybi@ietf.org>; Mon, 27 May 2013 01:47:33 -0700 (PDT)
Received: by mail-pd0-f170.google.com with SMTP id x10so6252330pdj.29 for <hybi@ietf.org>; Mon, 27 May 2013 01:47:32 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=UcSZDdIeggty+MtSEaU3yo9Oz9uq9DlAbRmAf4HcX0Y=; b=PoyrtdaJgAOcsUHyWrXzrICZ+vZz7Fx9fhoCuC9zPyhvJ7fOw97TesjjwjlLKRyp2+ 8GeVjJrmMjQK5FDi4n/JZvsFJ7gA3DzLQPaSBQ/D0lZf74yjL05EqrVp9r7sTh+CT0SA ptEM86ZoBFTp+I+1uo9YxVHKuCG91vol3kbkztJCJOJ3zfGxezy5vHrgwsdBES0OESZ9 Ac5h7UON9M5s+M0v8qvC+XuW7U/zAtEz2ZMME33eaRvMcXJ8Czof9XTDGL2AUJjcJn6f 63QBJUlOpIFj8C0kzbk/YRXMPXKIRlCKf/MpbOGNFwMlqTKyDUAZESPKpElrri6Uw8o+ Vkpw==
MIME-Version: 1.0
X-Received: by 10.67.22.66 with SMTP id hq2mr28366905pad.131.1369644452645; Mon, 27 May 2013 01:47:32 -0700 (PDT)
Sender: simone.bordet@gmail.com
Received: by 10.68.147.8 with HTTP; Mon, 27 May 2013 01:47:32 -0700 (PDT)
In-Reply-To: <CAH9hSJYOPvsFPDXLOa29ASd8xavLdvfRK_cVd=Uc=Vaydz1O=w@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <CAFWmRJ2Hbe0x5FeV2T7Gkp3WEsxQHe2=YPBTgvHYLcus3A4rBQ@mail.gmail.com> <CAH9hSJYOPvsFPDXLOa29ASd8xavLdvfRK_cVd=Uc=Vaydz1O=w@mail.gmail.com>
Date: Mon, 27 May 2013 10:47:32 +0200
X-Google-Sender-Auth: K9uolKnaCaly2SzQcm5rtI_LGCM
Message-ID: <CAFWmRJ2M0Gtoz80+6v+=0Ldm9+xE2brqD2shVcBPuNz+QGiKHg@mail.gmail.com>
From: Simone Bordet <sbordet@intalio.com>
To: Takeshi Yoshino <tyoshino@google.com>
Content-Type: text/plain; charset=UTF-8
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 May 2013 08:47:37 -0000

Hi,

On Mon, May 27, 2013 at 10:25 AM, Takeshi Yoshino <tyoshino@google.com> wrote:
> Isn't it in-app priority? Can't script and server address that problem? If
> only the client knows current priority, the client can tell the priority to
> the peer servlet using a WebSocket message. If the server know priority for
> each position, it can just control transmission speed.
>
> Even if we need help by the server, the servlet would call some special
> function to tell the server about priority.
>
> Different from HTTP/2.0's case, we don't need to provide a special field in
> mux protocol to pass priority, I think.

Sure, but then you force every application to reinvent this mechanism
on top of WebSocket mux.
At that point, one could argue that application messages could have
their own "channel_id" in their messages, and applications can do mux
themselves, if they need it.

--
Simone Bordet
----
http://cometd.org
http://webtide.com
http://intalio.com
Developer advice, training, services and support
from the Jetty & CometD experts.
Intalio, the modern way to build business applications.

From tyoshino@google.com  Mon May 27 01:52:20 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4384021F8FE5 for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 01:52:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.477
X-Spam-Level: 
X-Spam-Status: No, score=-2.477 tagged_above=-999 required=5 tests=[AWL=0.500,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, 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 fzXAcz3m1ggF for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 01:52:15 -0700 (PDT)
Received: from mail-ee0-f46.google.com (mail-ee0-f46.google.com [74.125.83.46]) by ietfa.amsl.com (Postfix) with ESMTP id B82E621F8F2C for <hybi@ietf.org>; Mon, 27 May 2013 01:52:14 -0700 (PDT)
Received: by mail-ee0-f46.google.com with SMTP id e49so3705432eek.5 for <hybi@ietf.org>; Mon, 27 May 2013 01:52:13 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=yQkBuN7/7UafyQhB2OyDeKNsC7eMm21c0Gna/R/q36s=; b=QBPEYE8JYs4L/MFdc8WRcYvvvee8MuvNKMgYxk7YZrWUFgzc5z1O9/j+Lso9/rFVFE he1g5HxRlQcuk7x+cIrBoy/9WHMsrOSNArNWhKezqBnLQOH6Ai/yi0G74XemjWgYSu72 k3YmZJUZY1kpwIsomAHghafpb7RytA19P75cnwUl0kkWYqbmfmSeEhOS9FrXuKi08AGN TD+EXMRGe956wvmEKUFtQo4/ACLaLokfULkrNEdXiy+5Hp9uXJiZdjtP5Ngr4zZgUSdm dLARcG7UgmM3kVxA1YjbvnJYgF/APecSpVmGCUnUY/arlG6Krcjp7u7a/BAIAdkV4aT1 IZhQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=yQkBuN7/7UafyQhB2OyDeKNsC7eMm21c0Gna/R/q36s=; b=WAi5uuUuqNR8EInjN/kIxsOMAHslEmuUspWWBm/Xk0XhaJCHhQvnPhUtEi6KWSKzV6 AxuzUkWf0XkTpsUVq8zmjUOTp74My6OaX7386FgvsRoEkwy1ouRlRSKmBWGAVBW92yc3 rfKoEWPfpe7/l9yJz+sZ8IH7zj6KdNbmwZRRBcpkDHtMW5AT2sBpfNlowMmKixGn0m/O X6dXK3/sKp4j9XiIicDUw6opwJOp43H16VF9WJ8RudtYaAICBRxHfdAq8mclPsUCOKqr EVnDDtC8meGHYbfL5UQaLt7C2NIrrTex/Q/kdYDe75C5Z5DyOPvrdCPGfTvOc8MZgMt/ rsyw==
X-Received: by 10.14.194.73 with SMTP id l49mr9051133een.99.1369644733850; Mon, 27 May 2013 01:52:13 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.86.67 with HTTP; Mon, 27 May 2013 01:51:53 -0700 (PDT)
In-Reply-To: <519D2E2D.4080809@tavendo.de>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <519D02E7.6040009@tavendo.de> <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com> <519D2E2D.4080809@tavendo.de>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Mon, 27 May 2013 17:51:53 +0900
Message-ID: <CAH9hSJZw7sLhQH6i2O9-rRKc88Bmh0EaJEcpxSB5kUKprP6YyQ@mail.gmail.com>
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
Content-Type: multipart/alternative; boundary=047d7b343e3ad4c10404ddaf437c
X-Gm-Message-State: ALoCoQkSp5O61gc+Y6v4jYitG1S9Kepyj9d1QDZP4c+FQdKas3STl2usXBj9/dOK7g3E9nx9FaPP0RK9fsJMiZI1gFgmbM8ZFebxIltEWx4eKjRGiLFVe4LqS3aWk87Iol+S8pmpdV+bD6VKAExktVGmlZbw1WXt1ma7Tm6PygQ8rUgNFO2OR9QfHhvKEYU2wenoyEi3vyGT
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 May 2013 08:52:20 -0000

--047d7b343e3ad4c10404ddaf437c
Content-Type: text/plain; charset=ISO-8859-1

On Thu, May 23, 2013 at 5:44 AM, Tobias Oberstein <
tobias.oberstein@tavendo.de> wrote:

> Having those buffers in user space (in app) allows one to tune the buffer
> size _per stream_. Thus, we can have large buffers for streams that carry
> mass data, and small buffers for chatty streams. Thats a real advantage.
>

Exactly.


> How about having a shared Web worker in the background that handles all
> messaging for the app over a single WS?
>

One restriction of Web Worker is that we cannot share a worker between
webapps with different origins while WebSocket can connect to a host
different from its origin.

For load balancer vendors, it's better that multiplexing is standardized
than each service provider develops its own.

Basically multiplexing should be done as close to network as possible. If
it's done in application layer, for example when compression is used, such
a load balancer need to decode the compression to see the multiplexing
information. That doesn't make them happy.

--

As Google, our motivation of mux is to make sure that total meaningful
traffic per server cost (memory, port, machine, etc.) small enough compared
to HTTP/Comet approach.

--047d7b343e3ad4c10404ddaf437c
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Thu, May 23, 2013 at 5:44 AM, Tobias Oberstein <span di=
r=3D"ltr">&lt;<a href=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_bla=
nk">tobias.oberstein@tavendo.de</a>&gt;</span> wrote:<br><div class=3D"gmai=
l_extra">


<div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margi=
n:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204=
);border-left-style:solid;padding-left:1ex">Having those buffers in user sp=
ace (in app) allows one to tune the buffer size _per stream_. Thus, we can =
have large buffers for streams that carry mass data, and small buffers for =
chatty streams. Thats a real advantage.<br>


</blockquote><div><br></div><div>Exactly.</div><div>=A0</div><blockquote cl=
ass=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px=
;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1e=
x">


<div><span style=3D"color:rgb(34,34,34)">How about having a shared Web work=
er in the background that handles all messaging for the app over a single W=
S?</span><br></div></blockquote><div><br></div><div>One restriction of Web =
Worker is that we cannot share a worker between webapps with different orig=
ins while WebSocket can connect to a host different from its origin.<br>


</div><div><div><br></div><div>For load balancer vendors, it&#39;s better t=
hat multiplexing is standardized than each service provider develops its ow=
n.</div><div><br></div><div>Basically multiplexing should be done as close =
to network as possible.=A0If it&#39;s done in application layer, for exampl=
e when compression is used, such a load balancer need to decode the compres=
sion to see the multiplexing information. That doesn&#39;t make them happy.=
<br>


</div><div><br></div><div>--</div><div><br></div><div><div>As Google, our m=
otivation of mux is to make sure that total meaningful traffic per server c=
ost (memory, port, machine, etc.) small enough compared to HTTP/Comet appro=
ach.</div>


</div></div><div>=A0</div></div></div></div>

--047d7b343e3ad4c10404ddaf437c--

From tyoshino@google.com  Mon May 27 02:03:41 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3579A21F90EF for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 02:03:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.977
X-Spam-Level: 
X-Spam-Status: No, score=-1.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 mlDvWGVezlHS for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 02:03:40 -0700 (PDT)
Received: from mail-ea0-x231.google.com (mail-ea0-x231.google.com [IPv6:2a00:1450:4013:c01::231]) by ietfa.amsl.com (Postfix) with ESMTP id 7FC8421F90AC for <hybi@ietf.org>; Mon, 27 May 2013 02:03:40 -0700 (PDT)
Received: by mail-ea0-f177.google.com with SMTP id q14so3751251eaj.8 for <hybi@ietf.org>; Mon, 27 May 2013 02:03:39 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=b1KCcCPvswmrKB/dCaQzqUsCthKx0p8ZeWm08tl8w1o=; b=cnRfNKjJQIKryOrfKeMbBe6toPzRP3nxS2bZpuMwMw0NWi3hgJOSzGS2V7/YdycJox 4jxAHVYy0fz9PoWMPLaiwahcfF+vfqLjcK8OdMQhlnLhFQJ6Ukph7ckg7M06SRoiM4XQ ErevSQrUOEuroHX/DcwDsvQtj05fMOoZ2G0a78oX8IFBD0r1bXsZ5zNscIopRC8K9TsU 0NRuWybAmVqeix6z0mijJIOQzHTeLZToen5rr70pWRA2V1J/Ckj5UucA1/BqDOIyco3Z NJtw9C/c/MZHU+5MqwIcRD/6DO1K8gVhFt9qU+v79LNJDmrXnDhxnwJcx6CVo+HUByzc JmpQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=b1KCcCPvswmrKB/dCaQzqUsCthKx0p8ZeWm08tl8w1o=; b=GhXOc345RVQVf8Fi6dnd21yQ0Kc5TGtQ6Cip5dgSUuJ4IEDINrDN+kfEdQz7aAH2MP Alz/p4lG8opa3zSv9RQb7JowQFzmMNt0O23yxfW4yG2NBSfpButTCgRE4F6rSUSaIe+t Fb8v/luRCPdCnE23MzGNyiyUaOtmQOvbaEKP/SrEZc+YvLs4S13j0J36VkWZuj88G/mS jvpQZYhOhhEqBTh9CeSjQU9ZApADli1VTgVOYlKkhkb/1NpuKKDmYP/yI0H7/Kbbw+Sg 0X7rFrPofYEtf6UKxkxIPyFbWdBEo967gCNXuslSimBJivnxtVsWW34LsNGSglRjkDsB yBug==
X-Received: by 10.14.178.134 with SMTP id f6mr9225704eem.129.1369645419565; Mon, 27 May 2013 02:03:39 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.86.67 with HTTP; Mon, 27 May 2013 02:03:19 -0700 (PDT)
In-Reply-To: <CAFWmRJ2M0Gtoz80+6v+=0Ldm9+xE2brqD2shVcBPuNz+QGiKHg@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <CAFWmRJ2Hbe0x5FeV2T7Gkp3WEsxQHe2=YPBTgvHYLcus3A4rBQ@mail.gmail.com> <CAH9hSJYOPvsFPDXLOa29ASd8xavLdvfRK_cVd=Uc=Vaydz1O=w@mail.gmail.com> <CAFWmRJ2M0Gtoz80+6v+=0Ldm9+xE2brqD2shVcBPuNz+QGiKHg@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Mon, 27 May 2013 18:03:19 +0900
Message-ID: <CAH9hSJb6FJK3qYENDXMKV8_s+Vv8b6SRnqjEaZnWUUBu6EuOkg@mail.gmail.com>
To: Simone Bordet <sbordet@intalio.com>
Content-Type: multipart/alternative; boundary=047d7b6242c0b3e09404ddaf6cad
X-Gm-Message-State: ALoCoQkb6cPwb/otAQmRHfjFUUuExzzKG7EUfc7jnWPcGtl/Du67OqCFpqQKSFMc/B3NrbQyzMa+sY+Zlnj/Omb9BtI8n7h761L7KtkhXYuF1ydMX8OhiGcbPLW4N4oVJsgeOc13Kj47zP0xrRm0n0X7Nv1sl8BLbNyPJqiemAEJFonBmc5qvpAz08eZ6FQFBcX04cDi9VEC
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 May 2013 09:03:41 -0000

--047d7b6242c0b3e09404ddaf6cad
Content-Type: text/plain; charset=ISO-8859-1

On Mon, May 27, 2013 at 5:47 PM, Simone Bordet <sbordet@intalio.com> wrote:

> > Different from HTTP/2.0's case, we don't need to provide a special field
> in
> > mux protocol to pass priority, I think.
>
> Sure, but then you force every application to reinvent this mechanism
> on top of WebSocket mux.
> At that point, one could argue that application messages could have
> their own "channel_id" in their messages, and applications can do mux
> themselves, if they need it.


Oh, ... good point.

If there gonna be inter-app prioritization (then load balancer would want
to look at ToS), frequent priority change (performance matters. better than
JS level processing), and wide adoption of priority (standardized method is
preferred), right.

--047d7b6242c0b3e09404ddaf6cad
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Mon, May 27, 2013 at 5:47 PM, Simone Bordet <span dir=
=3D"ltr">&lt;<a href=3D"mailto:sbordet@intalio.com" target=3D"_blank">sbord=
et@intalio.com</a>&gt;</span> wrote:<br><div class=3D"gmail_extra"><div cla=
ss=3D"gmail_quote">

<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div class=3D"im">&gt; Different from HTTP/2=
.0&#39;s case, we don&#39;t need to provide a special field in<br>
&gt; mux protocol to pass priority, I think.<br>
<br>
</div>Sure, but then you force every application to reinvent this mechanism=
<br>
on top of WebSocket mux.<br>
At that point, one could argue that application messages could have<br>
their own &quot;channel_id&quot; in their messages, and applications can do=
 mux<br>
themselves, if they need it.</blockquote><div><br></div><div style>Oh, ... =
good point.</div><div><br></div><div style>If there gonna be inter-app prio=
ritization (then load balancer would want to look at ToS), frequent priorit=
y change (performance matters. better than JS level processing), and wide a=
doption of priority (standardized method is preferred), right.</div>

<div>=A0</div></div></div></div>

--047d7b6242c0b3e09404ddaf6cad--

From tobias.oberstein@tavendo.de  Mon May 27 04:59:22 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A12A221F91C4 for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 04:59:22 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.998
X-Spam-Level: 
X-Spam-Status: No, score=-1.998 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, J_CHICKENPOX_24=0.6, WEIRD_PORT=0.001]
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 DKdb4mec9EPC for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 04:59:18 -0700 (PDT)
Received: from EXHUB020-3.exch020.serverdata.net (exhub020-3.exch020.serverdata.net [206.225.164.30]) by ietfa.amsl.com (Postfix) with ESMTP id B57C721F911B for <hybi@ietf.org>; Mon, 27 May 2013 04:59:14 -0700 (PDT)
Received: from EXVMBX020-12.exch020.serverdata.net ([169.254.3.33]) by EXHUB020-3.exch020.serverdata.net ([206.225.164.30]) with mapi; Mon, 27 May 2013 04:59:13 -0700
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
To: Takeshi Yoshino <tyoshino@google.com>
Date: Mon, 27 May 2013 04:59:09 -0700
Thread-Topic: [hybi] Call for interest: multiplexing dedicated for WebSocket
Thread-Index: Ac5arAyvD1Tcy7RjSrylZZe22aqswgAIvDrA
Message-ID: <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net> <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com>
In-Reply-To: <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: de-DE, en-US
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 May 2013 11:59:22 -0000

>>However, when running _different_ apps on the same device connecting to t=
he same origin,

>You meant same ws:// address?
=A0
Yes, sorry not origin.

I had a look into the MUX RFC again .. couldn't find anser to the following=
 (may have missed sth):

What WS addresses would be eligible to be multiplexed over a single WS conn=
ection?

a) ws://somehost.com:999/
b) ws://somehost.com:999/foo
c) ws://somehost.com:999/bar

d) wss://somehost.com:999/
e) wss://somehost.com:999/foo
f) wss://somehost.com:999/bar

All of a) - f) are to the same target IP:port and hence could share the sam=
e TCP.

However, d) - f) use wss, and hence have a TLS handshake right after TCP es=
tablishment.

So d) - f) cannot be multiplexed over the same physical WS as a) - c)?

Or can an implementation just "silently" transport a)-c) also over wss, and=
 hence multiplex all of a) - f) over 1 physical WS?

Lastly, a)-c) are to the same target IP:port and also WS schema (ws, not ws=
s) - and hence can be multiplexed over 1 physical WS even though they are t=
o different URL paths?

IOW: in the context of ws-mux, what is "same target"?

The combination of IP, port and WS schema (but not necessarily same URL pat=
h)?

>>Btw: assuming SPDY runs over TLS exclusively, is there any need for WS fr=
ame masking anymore?

>See this post by Eric about BEAST attack. I think it's not easy to conclud=
e that mask is unncessary for wss.
>http://www.educatedguesswork.org/2011/09/security_impact_of_the_rizzodu.ht=
ml
=A0
Ok, I see. Non-trivial.

From tobias.oberstein@tavendo.de  Mon May 27 05:52:03 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 700D921F9452 for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 05:52:03 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -0.305
X-Spam-Level: 
X-Spam-Status: No, score=-0.305 tagged_above=-999 required=5 tests=[AWL=-1.694, BAYES_00=-2.599, FRT_STOCK2=3.988]
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 FcKjHWIcpyXg for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 05:51:59 -0700 (PDT)
Received: from EXHUB020-2.exch020.serverdata.net (exhub020-2.exch020.serverdata.net [206.225.164.29]) by ietfa.amsl.com (Postfix) with ESMTP id A86AC21F93BB for <hybi@ietf.org>; Mon, 27 May 2013 05:51:58 -0700 (PDT)
Received: from EXVMBX020-12.exch020.serverdata.net ([169.254.3.33]) by EXHUB020-2.exch020.serverdata.net ([206.225.164.29]) with mapi; Mon, 27 May 2013 05:51:53 -0700
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
To: Takeshi Yoshino <tyoshino@google.com>
Date: Mon, 27 May 2013 05:51:51 -0700
Thread-Topic: [hybi] Call for interest: multiplexing dedicated for WebSocket
Thread-Index: Ac5at3zlbfyyn8pnSr+JNCrU7IV1VQAG4csw
Message-ID: <634914A010D0B943A035D226786325D4422C3DA775@EXVMBX020-12.exch020.serverdata.net>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <519D02E7.6040009@tavendo.de> <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com> <519D2E2D.4080809@tavendo.de> <CAH9hSJZw7sLhQH6i2O9-rRKc88Bmh0EaJEcpxSB5kUKprP6YyQ@mail.gmail.com>
In-Reply-To: <CAH9hSJZw7sLhQH6i2O9-rRKc88Bmh0EaJEcpxSB5kUKprP6YyQ@mail.gmail.com>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: de-DE, en-US
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 27 May 2013 12:52:03 -0000

>>Having those buffers in user space (in app) allows one to tune the buffer=
 size _per stream_. Thus, we can have large buffers for streams that carry =
mass data, and small buffers for chatty streams. Thats a real >>advantage.

>Exactly.

In the meantime, I have read more about modern TCP/IP stacks.

E.g. beginning with FreeBSD 7, the TCP/IP stack auto-tunes the TCP send/rec=
eive buffer sizes _per socket_. This is controlled via:

net.inet.tcp.sendbuf_auto=3D1         # TCP send buffer size autotuning (de=
fault: on)
net.inet.tcp.sendbuf_inc=3D16384      # TCP send buffer autotuning step siz=
e
net.inet.tcp.recvbuf_auto=3D1         # TCP receive buffer size autotuning =
(default: on)
net.inet.tcp.recvbuf_inc=3D524288     # TCP receive buffer autotuning step =
size
net.inet.tcp.sendbuf_max=3D16777216   # TCP send buffer maximum size (tune =
up for Fat-Long-Pipes)
net.inet.tcp.recvbuf_max=3D16777216   # TCP receive buffer maximum size (tu=
ne up for Fat-Long-Pipes)

"""
TCP Autotuning

Beginning with Linux 2.6, Mac OSX 10.5, Windows Vista, and FreeBSD 7.0, bot=
h sender and receiver autotuning became available, eliminating the need to =
set the TCP send and receive buffers by hand for each path. However the max=
imum buffer sizes are still too small for many high-speed network path, and=
 must be increased as described on the pages for each operating system.
"""
http://fasterdata.es.net/host-tuning/background/

Apart from autotuning, there is also the option to control send/receive buf=
fer sizes _per socket_ using setsockopt() with SO_SNDBUF / SO_RCVBUF.

=3D=3D=3D

Given above, how would the following 2 approaches compare?

a)
- Client 1 <=3D> Server 1 connected with 2 TCP connections.
- TCP autotuning on both client and server OR manually set buffer sizes: sm=
all for TCP-1, and large for TCP-2
- First TCP _only_ carries "chatty" traffic, and second TCP carries "mass d=
ata" traffic.

b)
- Client 1 <=3D> Server 1 connected with 1 TCP connection
- TCP autotuning OR large buffers set system wide
- multiplexing done over the single TCP that carries both "chatty" and "mas=
s data" traffic and
- the multiplexing is further optimizing buffer size (app level) and does p=
riorization of traffic (chatty =3D high-prio, mass =3D low-prio):

"Compare": I am mostly interested in the end-user experience:

i) Is the chatty traffic still low-latency, even in the presence of concurr=
ent mass-data traffic?
ii) Is the throughput of mass-data traffic concurrently to the chatty traff=
ic still (nearly) as high as when only doing mass data transfer?
=A0
>>How about having a shared Web worker in the background that handles all m=
essaging for the app over a single WS?
>One restriction of Web Worker is that we cannot share a worker between web=
apps with different origins while WebSocket can connect to a host different=
 from its origin.

Ah, right. Same origin policy. Thus, with WS-MUX, there is _more_ potential=
ly TCP sharing than whats possible with shared WebWorkers.

However, for a given single app (from a single origin) running in multiple =
browser tabs, the shared WebWorker seems not so bad. Works today. No need f=
or WS-MUX or HTTP 2.0 MUX + WS over HTTP 2.0.

>For load balancer vendors, it's better that multiplexing is standardized t=
han each service provider develops its own.

I don't have experience running G scale infrastructure. However, I'd like t=
o understand why LBs need to have a look into WS at all .. why not Layer-4 =
LB? Doing balancing based on hash of source IP?

>Basically multiplexing should be done as close to network as possible.=A0I=
f it's done in application layer, for example when compression is used, suc=
h a load balancer need to decode the compression to see the >multiplexing i=
nformation. That doesn't make them happy.

This seems to assume that=20

- either the LB is done above layer 4 and/or
- the logical WS connections contained in the physical WS would need to be =
balanced to _different_ backend nodes.

>As Google, our motivation of mux is to make sure that total meaningful tra=
ffic per server cost (memory, port, machine, etc.) small enough compared to=
 HTTP/Comet approach.
=A0
I see. For me, I am mostly interested in the use case: 1 app doing both cha=
tty and mass-data coms.

From tyoshino@google.com  Mon May 27 23:23:02 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3BAAD21F8EA4 for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 23:23:02 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.677
X-Spam-Level: 
X-Spam-Status: No, score=-1.677 tagged_above=-999 required=5 tests=[AWL=-0.300, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, J_CHICKENPOX_24=0.6, NO_RELAYS=-0.001, WEIRD_PORT=0.001]
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 iZTWU8Hq9m0R for <hybi@ietfa.amsl.com>; Mon, 27 May 2013 23:23:01 -0700 (PDT)
Received: from mail-ea0-x233.google.com (mail-ea0-x233.google.com [IPv6:2a00:1450:4013:c01::233]) by ietfa.amsl.com (Postfix) with ESMTP id CFD9B21F8481 for <hybi@ietf.org>; Mon, 27 May 2013 23:23:00 -0700 (PDT)
Received: by mail-ea0-f179.google.com with SMTP id z16so4337245ead.10 for <hybi@ietf.org>; Mon, 27 May 2013 23:22:59 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=alB3WFcjFlnHHPGgyLUd+OYJhC233sCjMO9vaLiW5pk=; b=iBcElCQy6zJZd+09PH0JSGwa7wrzEN9E8aDfydLGydA9OH5eRdxhiRykNh/jvfsCAA 2jFhDDa0bIqHElZSu+A1GnSIp/g+9SGbXnPTRPrazGUd5iuw+MaUrmv7Ev7r9l4FT9Ju +Hw8m5T8Oo/s7w9pZgLptFclrXH6V2Ht6s4oUpjSxAcTtK/cQuNn2VRauwugYErSh85r sRVfz34/H4huMWrUZp1P9WQeq+HJgm79NuBbFitk+SMb6Xk10CT3aBlgmXeTN2K928EC m7ZbqXT4qDZTc+08pXYQp2OHBZ+XfE9o5XJhhK3+TylwCqUlvHOw0CKW6dBymOgpH4DH AtOw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=alB3WFcjFlnHHPGgyLUd+OYJhC233sCjMO9vaLiW5pk=; b=d++hTGlrlzSGhRUUJgnkaJFihDeyngHZ9dafsdGU1HCmHAt27MlWGAMEUGbZzn4+2h +FW3UKl58RJYHpLN5ky55Ehj/1uLrJ5+6mE6IcbgA35od8NJAX4mF91Zv+xiNNE5WkMk Dd1xIz+0b/6BuXFBFxAPIJUs4VWPVM9DYMSj2AJqWO57CrBdx2DOE4lGh7c7p9o5kSDD /LqJ/TSWtMwuNMsrO8MJx8b8q7rlgGl8HH42WoUlfr0iK6v8bBF01Ox0dQAQ5LNzFKmn Zh2yCTIgamkNF7kMHVXOOmTsfZmF+ZtR0U69lOPNWzc04vIRoXeUy6qXieWJiOQ8veap HxMg==
X-Received: by 10.15.32.142 with SMTP id a14mr12583678eev.152.1369722179489; Mon, 27 May 2013 23:22:59 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.86.67 with HTTP; Mon, 27 May 2013 23:22:39 -0700 (PDT)
In-Reply-To: <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net> <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Tue, 28 May 2013 15:22:39 +0900
Message-ID: <CAH9hSJaAzpmkm2fvALma5YPhWs-E+5u1vOxdLZggJyJgEYGRUQ@mail.gmail.com>
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
Content-Type: multipart/alternative; boundary=089e01635210f349d304ddc14b7e
X-Gm-Message-State: ALoCoQmKn366fCGkH6GALQomyc70K9KE82ye7sK0sdqWYFpoOVJpqS78NR+zgFZk+goJebZU3m9edlKV/wtQzFo+MTky03s4dABg1LXNFqkjeaCbrc615V1PzznctH5Q+52vOWS04/CraXTHw74IvNkCDGJwqgzWQR6Y8eCdSwQUhPMmkDMMs8tlOL+tzPoNa4mnVCdsfI31
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 28 May 2013 06:23:02 -0000

--089e01635210f349d304ddc14b7e
Content-Type: text/plain; charset=ISO-8859-1

On Mon, May 27, 2013 at 8:59 PM, Tobias Oberstein <
tobias.oberstein@tavendo.de> wrote:

> >>However, when running _different_ apps on the same device connecting to
> the same origin,
>
> >You meant same ws:// address?
>
> Yes, sorry not origin.
>
> I had a look into the MUX RFC again .. couldn't find anser to the
> following (may have missed sth):
>

It's not yet specified.


> What WS addresses would be eligible to be multiplexed over a single WS
> connection?
>
> a) ws://somehost.com:999/
> b) ws://somehost.com:999/foo
> c) ws://somehost.com:999/bar
>
> d) wss://somehost.com:999/
> e) wss://somehost.com:999/foo
> f) wss://somehost.com:999/bar
>
> All of a) - f) are to the same target IP:port and hence could share the
> same TCP.
>

I think it's fine.


> However, d) - f) use wss, and hence have a TLS handshake right after TCP
> establishment.
>

Right unless TLS parameters differ (such as client cert) between paths.

See also SPDY's CREDENTIAL frame.
https://github.com/grmocg/SPDY-Specification/blob/gh-pages/draft-mbelshe-spdy-00.txt


> So d) - f) cannot be multiplexed over the same physical WS as a) - c)?
>

It's fine and up to the client.


> Or can an implementation just "silently" transport a)-c) also over wss,
> and hence multiplex all of a) - f) over 1 physical WS?
>
> Lastly, a)-c) are to the same target IP:port and also WS schema (ws, not
> wss) - and hence can be multiplexed over 1 physical WS even though they are
> to different URL paths?
>
> IOW: in the context of ws-mux, what is "same target"?


Basically ip:port. We don't need to isolate them based on hostname, I
believe, unless it's required by some security/privacy policy.

--089e01635210f349d304ddc14b7e
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Mon, May 27, 2013 at 8:59 PM, Tobias Oberstein <span di=
r=3D"ltr">&lt;<a href=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_bla=
nk">tobias.oberstein@tavendo.de</a>&gt;</span> wrote:<br><div class=3D"gmai=
l_extra">


<div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margi=
n:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204=
);border-left-style:solid;padding-left:1ex"><div>&gt;&gt;However, when runn=
ing _different_ apps on the same device connecting to the same origin,<br>



<br>
&gt;You meant same ws:// address?<br>
=A0<br>
</div>Yes, sorry not origin.<br>
<br>
I had a look into the MUX RFC again .. couldn&#39;t find anser to the follo=
wing (may have missed sth):<br></blockquote><div><br></div><div>It&#39;s no=
t yet specified.</div><div>=A0</div><blockquote class=3D"gmail_quote" style=
=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(20=
4,204,204);border-left-style:solid;padding-left:1ex">



What WS addresses would be eligible to be multiplexed over a single WS conn=
ection?<br>
<br>
a) ws://<a href=3D"http://somehost.com:999/" target=3D"_blank">somehost.com=
:999/</a><br>
b) ws://<a href=3D"http://somehost.com:999/foo" target=3D"_blank">somehost.=
com:999/foo</a><br>
c) ws://<a href=3D"http://somehost.com:999/bar" target=3D"_blank">somehost.=
com:999/bar</a><br>
<br>
d) wss://<a href=3D"http://somehost.com:999/" target=3D"_blank">somehost.co=
m:999/</a><br>
e) wss://<a href=3D"http://somehost.com:999/foo" target=3D"_blank">somehost=
.com:999/foo</a><br>
f) wss://<a href=3D"http://somehost.com:999/bar" target=3D"_blank">somehost=
.com:999/bar</a><br>
<br>
All of a) - f) are to the same target IP:port and hence could share the sam=
e TCP.<br></blockquote><div><br></div><div>I think it&#39;s fine.</div><div=
>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8=
ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-sty=
le:solid;padding-left:1ex">


However, d) - f) use wss, and hence have a TLS handshake right after TCP es=
tablishment.<br></blockquote><div><br></div><div>Right unless TLS parameter=
s differ (such as client cert) between paths.</div><div><br></div><div styl=
e>

See also SPDY&#39;s CREDENTIAL frame.=A0<a href=3D"https://github.com/grmoc=
g/SPDY-Specification/blob/gh-pages/draft-mbelshe-spdy-00.txt">https://githu=
b.com/grmocg/SPDY-Specification/blob/gh-pages/draft-mbelshe-spdy-00.txt</a>=
</div>

<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px=
 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left=
-style:solid;padding-left:1ex">
So d) - f) cannot be multiplexed over the same physical WS as a) - c)?<br><=
/blockquote><div><br></div><div>It&#39;s fine and up to the client.</div><d=
iv>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0=
.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-s=
tyle:solid;padding-left:1ex">


Or can an implementation just &quot;silently&quot; transport a)-c) also ove=
r wss, and hence multiplex all of a) - f) over 1 physical WS?<br>
<br>
Lastly, a)-c) are to the same target IP:port and also WS schema (ws, not ws=
s) - and hence can be multiplexed over 1 physical WS even though they are t=
o different URL paths?<br>
<br>
IOW: in the context of ws-mux, what is &quot;same target&quot;?</blockquote=
><div><br></div><div>Basically ip:port. We don&#39;t need to isolate them b=
ased on hostname, I believe, unless it&#39;s required by some security/priv=
acy policy.</div>

<div>=A0</div></div></div></div>

--089e01635210f349d304ddc14b7e--

From tyoshino@google.com  Tue May 28 00:19:41 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E09E721F92FB for <hybi@ietfa.amsl.com>; Tue, 28 May 2013 00:19:40 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.827
X-Spam-Level: 
X-Spam-Status: No, score=-1.827 tagged_above=-999 required=5 tests=[AWL=0.150,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 ccSWApQd4EQR for <hybi@ietfa.amsl.com>; Tue, 28 May 2013 00:19:40 -0700 (PDT)
Received: from mail-ea0-x231.google.com (mail-ea0-x231.google.com [IPv6:2a00:1450:4013:c01::231]) by ietfa.amsl.com (Postfix) with ESMTP id 1A73A21F8F06 for <hybi@ietf.org>; Tue, 28 May 2013 00:19:39 -0700 (PDT)
Received: by mail-ea0-f177.google.com with SMTP id q14so4349332eaj.22 for <hybi@ietf.org>; Tue, 28 May 2013 00:19:39 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :content-type; bh=foz/C8T23ymS1aDq4qQp+TsUuDFMDGT70aNAxf0p+bM=; b=lxku8Rvvi8v7ObzKmVXyXQrjfJ/uUNYF82Qhub0590RoKtnTtdeXZo9o1F4KH27RB0 xRMQraEfCK8NDNTo3mfons7KwDMUu2IuCF2ZhfqeDju65ao7Vvs+wlZl8wmBq23MJbdF 10yR48E1smCYKLAV/1RaDOfAcURI6kehiyLAbecceEmJYRseuURkUsxSaqwXza2vmJ69 37x/b1L8NCY1mLK2lTR1xmIdHGYefugGOancQUQM6Z0tvX3uqlrNcMCLr6Q+WADtNnAS m4RR/+A9KhguxU5nCjSGH9hK6ePCm1luS0n/+r75uXYyFfXMeug5H6wK+JV91AWIf50h fLdw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :content-type:x-gm-message-state; bh=foz/C8T23ymS1aDq4qQp+TsUuDFMDGT70aNAxf0p+bM=; b=fuDd7lnHXUlGVekfmWZd5S5cQIMQm7mLVoX4UG/eZzcGKNF+Hu8JCfRHVDF9/mH80u sDnhvyGX389cGp4/RCuS0Wrq3+YFPyAxSwrggKSBlNrSHj6rZ4M4oKTzHfSu+yzE00xD vQ4JzGajlj+hodAG3ZKcP7oEH1mXf3PLlcVVass+I7GzPios+v7yAuyqXblSg89o17Pq tVHNi3VoWgH8tM0ykUNtDSWqzCOHaYkns6sv2urBN4gUrHC61L/mf4kR4lsNHTNjvXsh 1qa9bpUdnVPQGaFPoiJ177y5vtDWE0ecEA6S6LaA/HppAjrxy/+QHViQYnPbwjnR4Ao/ sYxA==
X-Received: by 10.14.216.201 with SMTP id g49mr56789222eep.42.1369725578926; Tue, 28 May 2013 00:19:38 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.86.67 with HTTP; Tue, 28 May 2013 00:19:18 -0700 (PDT)
In-Reply-To: <CAH9hSJYDH47Ya1FNp80xjeD=pwYJAqcx=XDr=SnBbNDD+f-r4Q@mail.gmail.com>
References: <20130425140626.10027.9016.idtracker@ietfa.amsl.com> <CAH9hSJYDH47Ya1FNp80xjeD=pwYJAqcx=XDr=SnBbNDD+f-r4Q@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Tue, 28 May 2013 16:19:18 +0900
Message-ID: <CAH9hSJYa8QA9VkOwS6GEr0+8iJcnQcpMy3X_fKwGQOuJRr=sEw@mail.gmail.com>
To: "hybi@ietf.org" <hybi@ietf.org>
Content-Type: multipart/alternative; boundary=047d7b603cf092a4ce04ddc2160d
X-Gm-Message-State: ALoCoQlouBbuMjwteL3OWvhkKp8XlZL5/RY66/C0Sp2CXxGBXKEM6lHZongRjBoUNybV0LUy8o7CbrnR+SgQGnYWeP3pdrARvQXmZvRROqmn/jtmEifdnC0TXCeAA1VnqWvZvRwCR34BMEL5fKhr8cu1zbMlV+GMHKRxkGTk6RccFCpZ5fEi+fROyOWDgeW8cTIRt1Ec6QEJ
Subject: Re: [hybi] I-D Action: draft-ietf-hybi-permessage-compression-09.txt
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 28 May 2013 07:19:41 -0000

--047d7b603cf092a4ce04ddc2160d
Content-Type: text/plain; charset=ISO-8859-1

Hi all,

Please post your opinions and experience (implemented, faced any
difficulties, found issues, etc.).

--047d7b603cf092a4ce04ddc2160d
Content-Type: text/html; charset=ISO-8859-1

<div dir="ltr">Hi all,<div><br></div><div style>Please post your opinions and experience (implemented, faced any difficulties, found issues, etc.).</div><div><br></div></div>

--047d7b603cf092a4ce04ddc2160d--

From scott@adligo.com  Tue May 28 12:44:09 2013
Return-Path: <scott@adligo.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 668C221F9600 for <hybi@ietfa.amsl.com>; Tue, 28 May 2013 12:44:09 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.376
X-Spam-Level: 
X-Spam-Status: No, score=-1.376 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, J_CHICKENPOX_24=0.6, NO_RELAYS=-0.001, WEIRD_PORT=0.001]
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 9KqBM1S7rIfD for <hybi@ietfa.amsl.com>; Tue, 28 May 2013 12:44:08 -0700 (PDT)
Received: from mail-ie0-x229.google.com (mail-ie0-x229.google.com [IPv6:2607:f8b0:4001:c03::229]) by ietfa.amsl.com (Postfix) with ESMTP id 101CD21F92A5 for <hybi@ietf.org>; Tue, 28 May 2013 12:44:07 -0700 (PDT)
Received: by mail-ie0-f169.google.com with SMTP id u16so22985215iet.0 for <hybi@ietf.org>; Tue, 28 May 2013 12:44:07 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type:x-gm-message-state; bh=cDm3aMJGC9Eif2Tt+pEqnAINFMMFLr8NIwiCGAIBdxU=; b=Zm7WbZx9xUlcfQevZGaOf5wnpsjWBe5v6xGv2p39w916BvUdfy0uazBDRsY8Jq/xBN oWH1AZWVqQHlSWiFVDu32Cs0tS245aF9fSD8QnTgt/cP9SywP9EuOQ9rdpVuPt7rLtf7 9U4gO76iOpYEBR2mR8MC/WAS1kJ9SXq3Pc2UX8IJC0z12bhI09IZI1xuN1Y/QlwqBGVY bDLXz2XDSckn2YhxpLOP/nD3ToGja3fIXGuiCaRPbwYI43Mhw8CIDUWNMO8WBfGK+LQw WzY4Mdb8/c5ssdBbuQVToXiSS5KiDASuwY+Qlu32I5oD4Lb3LXgueV+poJT9Z5I45H0v sUAw==
MIME-Version: 1.0
X-Received: by 10.50.114.161 with SMTP id jh1mr7814226igb.112.1369770247350; Tue, 28 May 2013 12:44:07 -0700 (PDT)
Received: by 10.64.148.71 with HTTP; Tue, 28 May 2013 12:44:07 -0700 (PDT)
In-Reply-To: <mailman.3.1369767603.10801.hybi@ietf.org>
References: <mailman.3.1369767603.10801.hybi@ietf.org>
Date: Tue, 28 May 2013 14:44:07 -0500
Message-ID: <CANEdHmgDD4OCQf009FijtEjU=LzhLoNLZHvXsBxgROmzAyR4+w@mail.gmail.com>
From: Scott Morgan <scott@adligo.com>
To: hybi@ietf.org
Content-Type: multipart/alternative; boundary=089e0122ac44047acb04ddcc7de4
X-Gm-Message-State: ALoCoQmPF4F5iVrzFOD1+dT8JhKSAYp3yX+I5Z4Td9njiEZfgLORb2yyX30+V+4RNV4850bKLEw9
Subject: Re: [hybi] hybi Digest, Vol 51, Issue 14
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 28 May 2013 19:44:09 -0000

--089e0122ac44047acb04ddcc7de4
Content-Type: text/plain; charset=ISO-8859-1

Hi,

    Also what will happen if multiple to the multiplexed ws if one of two
browser tabs calls close on the ws connection?
   Does the other tab force keep the ws connection open?
   Does the other tab have to auto recover and reopen the connection?

Sorry if I missed this somewhere in the drafts.
Just my 2cents.

   It's important for a attempt to block browser ip address spoofing on the
first ws connection in a the following use case;

1 Browser connects to a wss server
2 User does some authentication (ie user id, password)
3 wss server does authentication using user id, password and ip address
(for either some privileges or as a additional authentication parameter for
a success fail style authentication)
4 after successful auth message at the browser an the wss connection is
closed
 and a different wss is opened on the same port with a different path in
the url
5 The wss server only allows the different wss connection to open if there
is a http session from the original wss connection from the same ip address
of the first connection.

Not that this is a fool proof solution to ip address spoofing, but it would
at least provide some level of ip address spoofing support (as the spoofer
would need to know that it had to spoof twice).  If the connection stayed
open, the original spoof would remain in effect and this sort of attempt to
block ip address spoofing would be rendered completely in-effective.

Cheers,
Scott



On Tue, May 28, 2013 at 2:00 PM, <hybi-request@ietf.org> wrote:

> If you have received this digest without all the individual message
> attachments you will need to update your digest options in your list
> subscription.  To do so, go to
>
> https://www.ietf.org/mailman/listinfo/hybi
>
> Click the 'Unsubscribe or edit options' button, log in, and set "Get
> MIME or Plain Text Digests?" to MIME.  You can set this option
> globally for all the list digests you receive at this point.
>
>
>
> Send hybi mailing list submissions to
>         hybi@ietf.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         https://www.ietf.org/mailman/listinfo/hybi
> or, via email, send a message with subject or body 'help' to
>         hybi-request@ietf.org
>
> You can reach the person managing the list at
>         hybi-owner@ietf.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of hybi digest..."
>
>
> Today's Topics:
>
>    1. Re: Call for interest: multiplexing dedicated for WebSocket
>       (Takeshi Yoshino)
>    2. Re: I-D Action: draft-ietf-hybi-permessage-compression-09.txt
>       (Takeshi Yoshino)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Tue, 28 May 2013 15:22:39 +0900
> From: Takeshi Yoshino <tyoshino@google.com>
> To: Tobias Oberstein <tobias.oberstein@tavendo.de>
> Cc: "hybi@ietf.org" <hybi@ietf.org>
> Subject: Re: [hybi] Call for interest: multiplexing dedicated for
>         WebSocket
> Message-ID:
>         <
> CAH9hSJaAzpmkm2fvALma5YPhWs-E+5u1vOxdLZggJyJgEYGRUQ@mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> On Mon, May 27, 2013 at 8:59 PM, Tobias Oberstein <
> tobias.oberstein@tavendo.de> wrote:
>
> > >>However, when running _different_ apps on the same device connecting to
> > the same origin,
> >
> > >You meant same ws:// address?
> >
> > Yes, sorry not origin.
> >
> > I had a look into the MUX RFC again .. couldn't find anser to the
> > following (may have missed sth):
> >
>
> It's not yet specified.
>
>
> > What WS addresses would be eligible to be multiplexed over a single WS
> > connection?
> >
> > a) ws://somehost.com:999/
> > b) ws://somehost.com:999/foo
> > c) ws://somehost.com:999/bar
> >
> > d) wss://somehost.com:999/
> > e) wss://somehost.com:999/foo
> > f) wss://somehost.com:999/bar
> >
> > All of a) - f) are to the same target IP:port and hence could share the
> > same TCP.
> >
>
> I think it's fine.
>
>
> > However, d) - f) use wss, and hence have a TLS handshake right after TCP
> > establishment.
> >
>
> Right unless TLS parameters differ (such as client cert) between paths.
>
> See also SPDY's CREDENTIAL frame.
>
> https://github.com/grmocg/SPDY-Specification/blob/gh-pages/draft-mbelshe-spdy-00.txt
>
>
> > So d) - f) cannot be multiplexed over the same physical WS as a) - c)?
> >
>
> It's fine and up to the client.
>
>
> > Or can an implementation just "silently" transport a)-c) also over wss,
> > and hence multiplex all of a) - f) over 1 physical WS?
> >
> > Lastly, a)-c) are to the same target IP:port and also WS schema (ws, not
> > wss) - and hence can be multiplexed over 1 physical WS even though they
> are
> > to different URL paths?
> >
> > IOW: in the context of ws-mux, what is "same target"?
>
>
> Basically ip:port. We don't need to isolate them based on hostname, I
> believe, unless it's required by some security/privacy policy.
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://www.ietf.org/mail-archive/web/hybi/attachments/20130528/0180f4a2/attachment.html
> >
>
> ------------------------------
>
> Message: 2
> Date: Tue, 28 May 2013 16:19:18 +0900
> From: Takeshi Yoshino <tyoshino@google.com>
> To: "hybi@ietf.org" <hybi@ietf.org>
> Subject: Re: [hybi] I-D Action:
>         draft-ietf-hybi-permessage-compression-09.txt
> Message-ID:
>         <CAH9hSJYa8QA9VkOwS6GEr0+8iJcnQcpMy3X_fKwGQOuJRr=
> sEw@mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hi all,
>
> Please post your opinions and experience (implemented, faced any
> difficulties, found issues, etc.).
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://www.ietf.org/mail-archive/web/hybi/attachments/20130528/74b6e651/attachment.htm
> >
>
> ------------------------------
>
> _______________________________________________
> hybi mailing list
> hybi@ietf.org
> https://www.ietf.org/mailman/listinfo/hybi
>
>
> End of hybi Digest, Vol 51, Issue 14
> ************************************
>

--089e0122ac44047acb04ddcc7de4
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hi,<div><br></div><div style>=A0 =A0 Also what will happen=
 if multiple to the=A0<span style=3D"font-family:arial,sans-serif;font-size=
:13px">multiplexed ws if one of two browser tabs calls close on the ws conn=
ection?</span></div>
<div style><span style=3D"font-family:arial,sans-serif;font-size:13px">=A0 =
=A0Does the other tab force keep the ws connection open?</span></div><div s=
tyle><span style=3D"font-family:arial,sans-serif;font-size:13px">=A0 =A0Doe=
s the other tab have to auto recover and reopen the connection?</span></div=
>
<div style><span style=3D"font-family:arial,sans-serif;font-size:13px"><br>=
</span></div><div style><span style=3D"font-family:arial,sans-serif;font-si=
ze:13px">Sorry if I missed this somewhere in the drafts.</span></div><div s=
tyle>
<span style=3D"font-family:arial,sans-serif;font-size:13px">Just my 2cents.=
</span></div><div style><span style=3D"font-family:arial,sans-serif;font-si=
ze:13px"><br></span></div><div style><span style=3D"font-family:arial,sans-=
serif;font-size:13px">=A0 =A0It&#39;s important for a attempt to block brow=
ser ip address spoofing on the first ws connection in a the following use c=
ase;</span></div>
<div style><span style=3D"font-family:arial,sans-serif;font-size:13px"><br>=
</span></div><div style><span style=3D"font-family:arial,sans-serif;font-si=
ze:13px">1 Browser connects to a wss server</span></div><div style><span st=
yle=3D"font-family:arial,sans-serif;font-size:13px">2 User does some authen=
tication (ie user id, password)</span></div>
<div style><span style=3D"font-family:arial,sans-serif;font-size:13px">3 ws=
s server does authentication using user id, password and ip address</span><=
/div><div style><span style=3D"font-family:arial,sans-serif;font-size:13px"=
>(for either some privileges or as a additional authentication parameter fo=
r a success fail style authentication)</span></div>
<div style><span style=3D"font-family:arial,sans-serif;font-size:13px">4 af=
ter successful auth message at the browser an the=A0</span><font face=3D"ar=
ial, sans-serif">wss connection is closed</font></div><div style><font face=
=3D"arial, sans-serif">=A0and a different wss is opened on the same port wi=
th a different path in the url</font></div>
<div style><font face=3D"arial, sans-serif">5 The wss server only allows th=
e different wss connection to open if there is a http session from the orig=
inal wss connection from the same ip address of the first connection.</font=
></div>
<div style><font face=3D"arial, sans-serif"><br></font></div><div style><fo=
nt face=3D"arial, sans-serif">Not that this is a fool proof solution to ip =
address spoofing, but it would at least provide some level of ip address sp=
oofing support (as the spoofer would need to know that it had to spoof twic=
e). =A0If the connection stayed open, the original spoof would remain in ef=
fect and this sort of attempt to block ip address spoofing would be rendere=
d completely in-effective.</font></div>
<div style><br></div><div style><span style=3D"font-family:arial,sans-serif=
;font-size:13px">Cheers,</span></div><div style><span style=3D"font-family:=
arial,sans-serif;font-size:13px">Scott</span></div><div style><span style=
=3D"font-family:arial,sans-serif;font-size:13px"><br>
</span></div></div><div class=3D"gmail_extra"><br><br><div class=3D"gmail_q=
uote">On Tue, May 28, 2013 at 2:00 PM,  <span dir=3D"ltr">&lt;<a href=3D"ma=
ilto:hybi-request@ietf.org" target=3D"_blank">hybi-request@ietf.org</a>&gt;=
</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">If you have received this digest without all=
 the individual message<br>
attachments you will need to update your digest options in your list<br>
subscription. =A0To do so, go to<br>
<br>
<a href=3D"https://www.ietf.org/mailman/listinfo/hybi" target=3D"_blank">ht=
tps://www.ietf.org/mailman/listinfo/hybi</a><br>
<br>
Click the &#39;Unsubscribe or edit options&#39; button, log in, and set &qu=
ot;Get<br>
MIME or Plain Text Digests?&quot; to MIME. =A0You can set this option<br>
globally for all the list digests you receive at this point.<br>
<br>
<br>
<br>
Send hybi mailing list submissions to<br>
=A0 =A0 =A0 =A0 <a href=3D"mailto:hybi@ietf.org">hybi@ietf.org</a><br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
=A0 =A0 =A0 =A0 <a href=3D"https://www.ietf.org/mailman/listinfo/hybi" targ=
et=3D"_blank">https://www.ietf.org/mailman/listinfo/hybi</a><br>
or, via email, send a message with subject or body &#39;help&#39; to<br>
=A0 =A0 =A0 =A0 <a href=3D"mailto:hybi-request@ietf.org">hybi-request@ietf.=
org</a><br>
<br>
You can reach the person managing the list at<br>
=A0 =A0 =A0 =A0 <a href=3D"mailto:hybi-owner@ietf.org">hybi-owner@ietf.org<=
/a><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than &quot;Re: Contents of hybi digest...&quot;<br>
<br>
<br>
Today&#39;s Topics:<br>
<br>
=A0 =A01. Re: Call for interest: multiplexing dedicated for WebSocket<br>
=A0 =A0 =A0 (Takeshi Yoshino)<br>
=A0 =A02. Re: I-D Action: draft-ietf-hybi-permessage-compression-09.txt<br>
=A0 =A0 =A0 (Takeshi Yoshino)<br>
<br>
<br>
----------------------------------------------------------------------<br>
<br>
Message: 1<br>
Date: Tue, 28 May 2013 15:22:39 +0900<br>
From: Takeshi Yoshino &lt;<a href=3D"mailto:tyoshino@google.com">tyoshino@g=
oogle.com</a>&gt;<br>
To: Tobias Oberstein &lt;<a href=3D"mailto:tobias.oberstein@tavendo.de">tob=
ias.oberstein@tavendo.de</a>&gt;<br>
Cc: &quot;<a href=3D"mailto:hybi@ietf.org">hybi@ietf.org</a>&quot; &lt;<a h=
ref=3D"mailto:hybi@ietf.org">hybi@ietf.org</a>&gt;<br>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for<br>
=A0 =A0 =A0 =A0 WebSocket<br>
Message-ID:<br>
=A0 =A0 =A0 =A0 &lt;<a href=3D"mailto:CAH9hSJaAzpmkm2fvALma5YPhWs-E%2B5u1vO=
xdLZggJyJgEYGRUQ@mail.gmail.com">CAH9hSJaAzpmkm2fvALma5YPhWs-E+5u1vOxdLZggJ=
yJgEYGRUQ@mail.gmail.com</a>&gt;<br>
Content-Type: text/plain; charset=3D&quot;iso-8859-1&quot;<br>
<br>
On Mon, May 27, 2013 at 8:59 PM, Tobias Oberstein &lt;<br>
<a href=3D"mailto:tobias.oberstein@tavendo.de">tobias.oberstein@tavendo.de<=
/a>&gt; wrote:<br>
<br>
&gt; &gt;&gt;However, when running _different_ apps on the same device conn=
ecting to<br>
&gt; the same origin,<br>
&gt;<br>
&gt; &gt;You meant same ws:// address?<br>
&gt;<br>
&gt; Yes, sorry not origin.<br>
&gt;<br>
&gt; I had a look into the MUX RFC again .. couldn&#39;t find anser to the<=
br>
&gt; following (may have missed sth):<br>
&gt;<br>
<br>
It&#39;s not yet specified.<br>
<br>
<br>
&gt; What WS addresses would be eligible to be multiplexed over a single WS=
<br>
&gt; connection?<br>
&gt;<br>
&gt; a) ws://<a href=3D"http://somehost.com:999/" target=3D"_blank">somehos=
t.com:999/</a><br>
&gt; b) ws://<a href=3D"http://somehost.com:999/foo" target=3D"_blank">some=
host.com:999/foo</a><br>
&gt; c) ws://<a href=3D"http://somehost.com:999/bar" target=3D"_blank">some=
host.com:999/bar</a><br>
&gt;<br>
&gt; d) wss://<a href=3D"http://somehost.com:999/" target=3D"_blank">someho=
st.com:999/</a><br>
&gt; e) wss://<a href=3D"http://somehost.com:999/foo" target=3D"_blank">som=
ehost.com:999/foo</a><br>
&gt; f) wss://<a href=3D"http://somehost.com:999/bar" target=3D"_blank">som=
ehost.com:999/bar</a><br>
&gt;<br>
&gt; All of a) - f) are to the same target IP:port and hence could share th=
e<br>
&gt; same TCP.<br>
&gt;<br>
<br>
I think it&#39;s fine.<br>
<br>
<br>
&gt; However, d) - f) use wss, and hence have a TLS handshake right after T=
CP<br>
&gt; establishment.<br>
&gt;<br>
<br>
Right unless TLS parameters differ (such as client cert) between paths.<br>
<br>
See also SPDY&#39;s CREDENTIAL frame.<br>
<a href=3D"https://github.com/grmocg/SPDY-Specification/blob/gh-pages/draft=
-mbelshe-spdy-00.txt" target=3D"_blank">https://github.com/grmocg/SPDY-Spec=
ification/blob/gh-pages/draft-mbelshe-spdy-00.txt</a><br>
<br>
<br>
&gt; So d) - f) cannot be multiplexed over the same physical WS as a) - c)?=
<br>
&gt;<br>
<br>
It&#39;s fine and up to the client.<br>
<br>
<br>
&gt; Or can an implementation just &quot;silently&quot; transport a)-c) als=
o over wss,<br>
&gt; and hence multiplex all of a) - f) over 1 physical WS?<br>
&gt;<br>
&gt; Lastly, a)-c) are to the same target IP:port and also WS schema (ws, n=
ot<br>
&gt; wss) - and hence can be multiplexed over 1 physical WS even though the=
y are<br>
&gt; to different URL paths?<br>
&gt;<br>
&gt; IOW: in the context of ws-mux, what is &quot;same target&quot;?<br>
<br>
<br>
Basically ip:port. We don&#39;t need to isolate them based on hostname, I<b=
r>
believe, unless it&#39;s required by some security/privacy policy.<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: &lt;<a href=3D"http://www.ietf.org/mail-archive/web/hybi/attachments/2=
0130528/0180f4a2/attachment.html" target=3D"_blank">http://www.ietf.org/mai=
l-archive/web/hybi/attachments/20130528/0180f4a2/attachment.html</a>&gt;<br=
>

<br>
------------------------------<br>
<br>
Message: 2<br>
Date: Tue, 28 May 2013 16:19:18 +0900<br>
From: Takeshi Yoshino &lt;<a href=3D"mailto:tyoshino@google.com">tyoshino@g=
oogle.com</a>&gt;<br>
To: &quot;<a href=3D"mailto:hybi@ietf.org">hybi@ietf.org</a>&quot; &lt;<a h=
ref=3D"mailto:hybi@ietf.org">hybi@ietf.org</a>&gt;<br>
Subject: Re: [hybi] I-D Action:<br>
=A0 =A0 =A0 =A0 draft-ietf-hybi-permessage-compression-09.txt<br>
Message-ID:<br>
=A0 =A0 =A0 =A0 &lt;CAH9hSJYa8QA9VkOwS6GEr0+8iJcnQcpMy3X_fKwGQOuJRr=3D<a hr=
ef=3D"mailto:sEw@mail.gmail.com">sEw@mail.gmail.com</a>&gt;<br>
Content-Type: text/plain; charset=3D&quot;iso-8859-1&quot;<br>
<br>
Hi all,<br>
<br>
Please post your opinions and experience (implemented, faced any<br>
difficulties, found issues, etc.).<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: &lt;<a href=3D"http://www.ietf.org/mail-archive/web/hybi/attachments/2=
0130528/74b6e651/attachment.htm" target=3D"_blank">http://www.ietf.org/mail=
-archive/web/hybi/attachments/20130528/74b6e651/attachment.htm</a>&gt;<br>

<br>
------------------------------<br>
<br>
_______________________________________________<br>
hybi mailing list<br>
<a href=3D"mailto:hybi@ietf.org">hybi@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/hybi" target=3D"_blank">ht=
tps://www.ietf.org/mailman/listinfo/hybi</a><br>
<br>
<br>
End of hybi Digest, Vol 51, Issue 14<br>
************************************<br>
</blockquote></div><br></div>

--089e0122ac44047acb04ddcc7de4--

From tyoshino@google.com  Tue May 28 20:37:06 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CFCEA21F8749 for <hybi@ietfa.amsl.com>; Tue, 28 May 2013 20:37:06 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.877
X-Spam-Level: 
X-Spam-Status: No, score=-1.877 tagged_above=-999 required=5 tests=[AWL=0.100,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 iiSJ0zhc3wrJ for <hybi@ietfa.amsl.com>; Tue, 28 May 2013 20:37:06 -0700 (PDT)
Received: from mail-wg0-x234.google.com (mail-wg0-x234.google.com [IPv6:2a00:1450:400c:c00::234]) by ietfa.amsl.com (Postfix) with ESMTP id C45CD21F8681 for <hybi@ietf.org>; Tue, 28 May 2013 20:37:05 -0700 (PDT)
Received: by mail-wg0-f52.google.com with SMTP id z11so6082422wgg.19 for <hybi@ietf.org>; Tue, 28 May 2013 20:37:04 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=caDwCC+I5aLN9jPvJV+a6+qUYnBrj4HuPnry4Zgyh7s=; b=lE0KP4p1K8Dm1Gfik1kRyHl+ahMbPEcWofSn5c4HjAOPUTYf81E89epcUC+ZbBCEqo pxP0PQ5umuJ24n0cCdedAbMy6YhmSRiLu2vVepBowS6CC/4OUKFRiVRj5krkIGXfXO7z B9q2aQgduZM5EkewNEkE16CTVzIVkiOct9scWpJuWOZFo0MNlev0bu7DCRwwrceNih9b MpsgLjYJl2MJ4Hi4CJauUP+oS/qlUnHytLTLJ1jDB3OuCxZ0FNm/nPAJokS7Ar64zVzI n/1FsYA244N92faeQSV4s8PNGBED1hLpOZgmIVJzduebeWEpWo+uUuK19ISjctoLMMaZ r4fg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=caDwCC+I5aLN9jPvJV+a6+qUYnBrj4HuPnry4Zgyh7s=; b=G/3eJ5o/C6NyJZ4GJ2BsOfJfmRdMuPBbzaIMXqpLVeVM0+gtjPwPPzzS1kUhxODKDl 41dcx2CYyPVg8cuL6xMgRb9mt2euvnSkOEE6HCLyZA91B1EXogl8zxr0/vb2BtCPO2v5 l57mNMt8XaSb2l8fOlc3V1pWgTSHrporMjwr70WsykCcWop9LalneTo2zWClsZpz6iLk fYTRi8XQZckzpIH5BfJ+52dzClfRxTrTQsmYHi2hYq9anpptWK5YoHrofhBX0t/4dOFo Ch04zGRhhoNQ/WlJnS1ju2a4d+tPcB70koRffas5/YqEmXxs+Md68jWcXMz1GrLYOAEj GN+w==
X-Received: by 10.180.11.176 with SMTP id r16mr500546wib.58.1369798624746; Tue, 28 May 2013 20:37:04 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.180.5.136 with HTTP; Tue, 28 May 2013 20:36:44 -0700 (PDT)
In-Reply-To: <CANEdHmgDD4OCQf009FijtEjU=LzhLoNLZHvXsBxgROmzAyR4+w@mail.gmail.com>
References: <mailman.3.1369767603.10801.hybi@ietf.org> <CANEdHmgDD4OCQf009FijtEjU=LzhLoNLZHvXsBxgROmzAyR4+w@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Wed, 29 May 2013 12:36:44 +0900
Message-ID: <CAH9hSJYfhnmCOF7a9uovsYLObqdXwop35JB6r3PscFwfeO=R=g@mail.gmail.com>
To: Scott Morgan <scott@adligo.com>
Content-Type: multipart/alternative; boundary=001a11c222027138dc04ddd3181a
X-Gm-Message-State: ALoCoQkp0Y9Q/MDtsdP3HBtv0tmzBXklF/B3aWDhuV3LbYDhyvrc1ybCkk+AAJKbPSmiPFW+XZVsmcWd640yBLCtMFt+uLtzVLfy7sLFm1uWRMU1B83J9IWVwbRp0y7HuUH7cfWmyfgdau8c79L7JXolTNEoI/tUnkkUKLJWiwj3TXWIhzw712ID1rxJwHZFnt7JYSWx4AsQ
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] hybi Digest, Vol 51, Issue 14
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 29 May 2013 03:37:06 -0000

--001a11c222027138dc04ddd3181a
Content-Type: text/plain; charset=ISO-8859-1

On Wed, May 29, 2013 at 4:44 AM, Scott Morgan <scott@adligo.com> wrote:

>     Also what will happen if multiple to the multiplexed ws if one of two
> browser tabs calls close on the ws connection?
>    Does the other tab force keep the ws connection open?
>    Does the other tab have to auto recover and reopen the connection?
>

close() call on a WebSocket instance that is multiplexed in network stack
should result in close of the logical channel used by the WebSocket
instance. Not close on the physical WebSocket connection. Once the last
logical connection is closed, the network stack may start closing the
physical connection. It can be also delayed.
http://tools.ietf.org/html/draft-ietf-hybi-websocket-multiplexing-10#section-15
is
the section discussing that.


> Sorry if I missed this somewhere in the drafts.
> Just my 2cents.
>
>    It's important for a attempt to block browser ip address spoofing on
> the first ws connection in a the following use case;
>
>
Could you elaborate?


> 1 Browser connects to a wss server
> 2 User does some authentication (ie user id, password)
> 3 wss server does authentication using user id, password and ip address
> (for either some privileges or as a additional authentication parameter
> for a success fail style authentication)
>

That is that only access from some IP address X will have some high
privilege?


>  4 after successful auth message at the browser an the wss connection is
> closed
>  and a different wss is opened on the same port with a different path in
> the url
> 5 The wss server only allows the different wss connection to open if there
> is a http session from the original wss connection from the same ip address
> of the first connection.
>

Does the server get a session parameter over the new connection and also
check the source IP address? Or the server here just checks if there was
successful authentication on some prior connection with the same IP address
as the new one?


>
> Not that this is a fool proof solution to ip address spoofing, but it
> would at least provide some level of ip address spoofing support (as the
> spoofer would need to know that it had to spoof twice).  If the connection
> stayed open, the original spoof would remain in effect and this sort of
> attempt to block ip address spoofing would be rendered completely
> in-effective.
>

--001a11c222027138dc04ddd3181a
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Wed, May 29, 2013 at 4:44 AM, Scott Morgan <span dir=3D=
"ltr">&lt;<a href=3D"mailto:scott@adligo.com" target=3D"_blank">scott@adlig=
o.com</a>&gt;</span> wrote:<br><div class=3D"gmail_extra"><div class=3D"gma=
il_quote">

<blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-=
left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;p=
adding-left:1ex"><div dir=3D"ltr"><div>=A0 =A0 Also what will happen if mul=
tiple to the=A0<span style=3D"font-family:arial,sans-serif;font-size:13px">=
multiplexed ws if one of two browser tabs calls close on the ws connection?=
</span></div>


<div><span style=3D"font-family:arial,sans-serif;font-size:13px">=A0 =A0Doe=
s the other tab force keep the ws connection open?</span></div><div><span s=
tyle=3D"font-family:arial,sans-serif;font-size:13px">=A0 =A0Does the other =
tab have to auto recover and reopen the connection?</span></div>

</div></blockquote><div><br></div><div style>close() call on a WebSocket in=
stance that is multiplexed in network stack should result in close of the l=
ogical channel used by the WebSocket instance. Not close on the physical We=
bSocket connection. Once the last logical connection is closed, the network=
 stack may start closing the physical connection. It can be also delayed. =
=A0<a href=3D"http://tools.ietf.org/html/draft-ietf-hybi-websocket-multiple=
xing-10#section-15">http://tools.ietf.org/html/draft-ietf-hybi-websocket-mu=
ltiplexing-10#section-15</a>=A0is the section discussing that.</div>

<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px=
 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left=
-style:solid;padding-left:1ex"><div dir=3D"ltr"><div><span style=3D"font-fa=
mily:arial,sans-serif;font-size:13px"></span></div>

<div><span style=3D"font-family:arial,sans-serif;font-size:13px">Sorry if I=
 missed this somewhere in the drafts.</span></div><div>
<span style=3D"font-family:arial,sans-serif;font-size:13px">Just my 2cents.=
</span></div><div><span style=3D"font-family:arial,sans-serif;font-size:13p=
x"><br></span></div><div><span style=3D"font-family:arial,sans-serif;font-s=
ize:13px">=A0 =A0It&#39;s important for a attempt to block browser ip addre=
ss spoofing on the first ws connection in a the following use case;</span><=
/div>


<div><span style=3D"font-family:arial,sans-serif;font-size:13px"><br></span=
></div></div></blockquote><div><br></div><div><div>Could you elaborate?</di=
v></div><div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px=
 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);bor=
der-left-style:solid;padding-left:1ex">

<div dir=3D"ltr"><div><span style=3D"font-family:arial,sans-serif;font-size=
:13px"></span></div><div><span style=3D"font-family:arial,sans-serif;font-s=
ize:13px">1 Browser connects to a wss server</span></div><div><span style=
=3D"font-family:arial,sans-serif;font-size:13px">2 User does some authentic=
ation (ie user id, password)</span></div>


<div><span style=3D"font-family:arial,sans-serif;font-size:13px">3 wss serv=
er does authentication using user id, password and ip address</span></div><=
div><span style=3D"font-family:arial,sans-serif;font-size:13px">(for either=
 some privileges or as a additional authentication parameter for a success =
fail style authentication)</span></div>

</div></blockquote><div><br></div><div style>That is that only access from =
some IP address X will have some high privilege?</div><div>=A0</div><blockq=
uote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left-wi=
dth:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-=
left:1ex">

<div dir=3D"ltr">
<div><span style=3D"font-family:arial,sans-serif;font-size:13px">4 after su=
ccessful auth message at the browser an the=A0</span><font face=3D"arial, s=
ans-serif">wss connection is closed</font></div><div><font face=3D"arial, s=
ans-serif">=A0and a different wss is opened on the same port with a differe=
nt path in the url</font></div>


<div><font face=3D"arial, sans-serif">5 The wss server only allows the diff=
erent wss connection to open if there is a http session from the original w=
ss connection from the same ip address of the first connection.</font></div=
>

</div></blockquote><div><br></div><div style>Does the server get a session =
parameter over the new connection and also check the source IP address? Or =
the server here just checks if there was successful authentication on some =
prior connection with the same IP address as the new one?</div>

<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px=
 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left=
-style:solid;padding-left:1ex"><div dir=3D"ltr">
<div><font face=3D"arial, sans-serif"><br></font></div><div><font face=3D"a=
rial, sans-serif">Not that this is a fool proof solution to ip address spoo=
fing, but it would at least provide some level of ip address spoofing suppo=
rt (as the spoofer would need to know that it had to spoof twice). =A0If th=
e connection stayed open, the original spoof would remain in effect and thi=
s sort of attempt to block ip address spoofing would be rendered completely=
 in-effective.</font></div>

</div></blockquote><div>=A0<br></div></div></div></div>

--001a11c222027138dc04ddd3181a--

From tyoshino@google.com  Tue May 28 22:43:33 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D16EE21F8EB2 for <hybi@ietfa.amsl.com>; Tue, 28 May 2013 22:43:33 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.392
X-Spam-Level: 
X-Spam-Status: No, score=0.392 tagged_above=-999 required=5 tests=[AWL=-2.219,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, FRT_STOCK2=3.988, HTML_MESSAGE=0.001, J_CHICKENPOX_52=0.6, NO_RELAYS=-0.001]
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 oqr8jiACoVtl for <hybi@ietfa.amsl.com>; Tue, 28 May 2013 22:43:32 -0700 (PDT)
Received: from mail-we0-x229.google.com (mail-we0-x229.google.com [IPv6:2a00:1450:400c:c03::229]) by ietfa.amsl.com (Postfix) with ESMTP id 245B721F8EBB for <hybi@ietf.org>; Tue, 28 May 2013 22:43:31 -0700 (PDT)
Received: by mail-we0-f169.google.com with SMTP id q55so6104405wes.14 for <hybi@ietf.org>; Tue, 28 May 2013 22:43:31 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=qAiCr5lG5RjxJ8dr732LA7r8kS8Fmo8nKWhRl5FQwT4=; b=UqM0zgF1HHaCaKuru0B60OW9cZP+D4S11KcEKn0p2abvGPUP+9Sob0il4HNqW3Kocw CId6xcNO9SWavW6oXzcfLQIhwTshjgjJwSKcj9rNeH1wyiYstziTb+TVGkHwM3QaX0N+ Pdmy24tiLaRX1qlDdCiLsvW6iyCpqoTAm7FYIGkyZ722hhysHeLctApqfTn+536ERvWW QNRD0kL5p5WVRKGM1DZFyXtmbh37P0+QOR0y4tqCU9pvURCuCCskbZqXtQdEWfsqtuH1 YTiVOVaNzi5h+DE+TXwYruosXBxv3/0syNs9RlV9yuup0EapF99xwnn9mxZHY4ftoZJk PJgg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=qAiCr5lG5RjxJ8dr732LA7r8kS8Fmo8nKWhRl5FQwT4=; b=prNSNOOs0IEnKHLMYlNLrVWNIEowDXPT7P6/fxpXCPcqkIy3Vptc/8f4P1AardOUAT wRDpncjvlNIofe1czu0KFEcNGRxYM95vbOIH3bjcZM2h+CMpY0NKNQYOoeWCyq+EWpTX bOX/hMtbJJFsKImAM8VuMb10RskdvcerjGRObq3NZS+mskH0F8xA3+dvsJW27d6Cm4qC 5Dza1XLGSiU0zy5bohtQMgmsVbMPzZvOieXlnqalRdJWdh0lfM1/BnFYVlvj92uwWEjJ /JdJo6olNFiGNbcGkXfw8SOoZrG6148NPzdbF/943tz/g3/lNB083j1U3fAVPaqgBxBf dY1Q==
X-Received: by 10.180.11.176 with SMTP id r16mr778560wib.58.1369806211164; Tue, 28 May 2013 22:43:31 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.180.5.136 with HTTP; Tue, 28 May 2013 22:43:11 -0700 (PDT)
In-Reply-To: <634914A010D0B943A035D226786325D4422C3DA775@EXVMBX020-12.exch020.serverdata.net>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <519D02E7.6040009@tavendo.de> <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com> <519D2E2D.4080809@tavendo.de> <CAH9hSJZw7sLhQH6i2O9-rRKc88Bmh0EaJEcpxSB5kUKprP6YyQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C3DA775@EXVMBX020-12.exch020.serverdata.net>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Wed, 29 May 2013 14:43:11 +0900
Message-ID: <CAH9hSJYms5rPSvDDNeMYoZ19UmRLJUd7YM7d98ONUkB4YZnM=Q@mail.gmail.com>
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
Content-Type: multipart/alternative; boundary=001a11c22202a0bdec04ddd4dc86
X-Gm-Message-State: ALoCoQmNzCBUZEUT0NEB1qruE1uwsLnHrBWsS7e7sps2LIz7uW7bEzxIgZ7zVLRGN19Zsp6qNvb4g129m1uxfmkQ2ltKTlFAT+0Gd+GgvxbrPj3YDfMRKIKUKuTJ3tArDgHbM/lndMhE9j25CmfAFX1sQoJ19hZJ4XN1wjn8QkNqfL5RTp1oExgpACP1mKUiYiKLfdg4EMuv
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 29 May 2013 05:43:33 -0000

--001a11c22202a0bdec04ddd4dc86
Content-Type: text/plain; charset=ISO-8859-1

On Mon, May 27, 2013 at 9:51 PM, Tobias Oberstein <
tobias.oberstein@tavendo.de> wrote:

> >>Having those buffers in user space (in app) allows one to tune the
> buffer size _per stream_. Thus, we can have large buffers for streams that
> carry mass data, and small buffers for chatty streams. Thats a real
> >>advantage.
>
> >Exactly.
>
> In the meantime, I have read more about modern TCP/IP stacks.
>
> E.g. beginning with FreeBSD 7, the TCP/IP stack auto-tunes the TCP
> send/receive buffer sizes _per socket_. This is controlled via:
>
> net.inet.tcp.sendbuf_auto=1         # TCP send buffer size autotuning
> (default: on)
> net.inet.tcp.sendbuf_inc=16384      # TCP send buffer autotuning step size
> net.inet.tcp.recvbuf_auto=1         # TCP receive buffer size autotuning
> (default: on)
> net.inet.tcp.recvbuf_inc=524288     # TCP receive buffer autotuning step
> size
> net.inet.tcp.sendbuf_max=16777216   # TCP send buffer maximum size (tune
> up for Fat-Long-Pipes)
> net.inet.tcp.recvbuf_max=16777216   # TCP receive buffer maximum size
> (tune up for Fat-Long-Pipes)
>
> """
> TCP Autotuning
>
> Beginning with Linux 2.6, Mac OSX 10.5, Windows Vista, and FreeBSD 7.0,
> both sender and receiver autotuning became available, eliminating the need
> to set the TCP send and receive buffers by hand for each path. However the
> maximum buffer sizes are still too small for many high-speed network path,
> and must be increased as described on the pages for each operating system.
> """
> http://fasterdata.es.net/host-tuning/background/
>
> Apart from autotuning, there is also the option to control send/receive
> buffer sizes _per socket_ using setsockopt() with SO_SNDBUF / SO_RCVBUF.
>
> ===
>
> Given above, how would the following 2 approaches compare?
>
> a)
> - Client 1 <=> Server 1 connected with 2 TCP connections.
> - TCP autotuning on both client and server OR manually set buffer sizes:
> small for TCP-1, and large for TCP-2
> - First TCP _only_ carries "chatty" traffic, and second TCP carries "mass
> data" traffic.
>
> b)
> - Client 1 <=> Server 1 connected with 1 TCP connection
> - TCP autotuning OR large buffers set system wide
> - multiplexing done over the single TCP that carries both "chatty" and
> "mass data" traffic and
> - the multiplexing is further optimizing buffer size (app level) and does
> priorization of traffic (chatty = high-prio, mass = low-prio):
>
> "Compare": I am mostly interested in the end-user experience:
>
> i) Is the chatty traffic still low-latency, even in the presence of
> concurrent mass-data traffic?
>

a) it depends on process scheduler and network queue design of the server
and client, I think. I don't know much about per-socket buffer autotuning,
but it probably just improves memory usage and throughput but doesn't have
much effect on latency of chatty traffic?

b) the endpoints need to cut mass data into smaller chunks to reserve slots
for inserting chatty traffic. e.g. if acceptable delay for chatty traffic
is x ms, set the maximum send unit of mass traffic to (current through put)
* x.

To allow one to control the peer's message scheduling, both per-socket
buffer size and ws-mux flow control are insufficient, I think. Needs to
exchange ToS info.


> ii) Is the throughput of mass-data traffic concurrently to the chatty
> traffic still (nearly) as high as when only doing mass data transfer?
>

I think minimum delay requirement for the chatty traffic determines
fragmentation on mass-data traffic regardless it's done in OS's network
stack or ws-mux code.


> >>How about having a shared Web worker in the background that handles all
> messaging for the app over a single WS?
> >One restriction of Web Worker is that we cannot share a worker between
> webapps with different origins while WebSocket can connect to a host
> different from its origin.
>
> Ah, right. Same origin policy. Thus, with WS-MUX, there is _more_
> potentially TCP sharing than whats possible with shared WebWorkers.
>
> However, for a given single app (from a single origin) running in multiple
> browser tabs, the shared WebWorker seems not so bad. Works today. No need
> for WS-MUX or HTTP 2.0 MUX + WS over HTTP 2.0.
>

Yes. Thanks for reminding us about shared worker approach. We haven't
evaluated it well.


> >For load balancer vendors, it's better that multiplexing is standardized
> than each service provider develops its own.
>
> I don't have experience running G scale infrastructure. However, I'd like
> to understand why LBs need to have a look into WS at all .. why not Layer-4
> LB? Doing balancing based on hash of source IP?
>

Traffic for various services would be multiplexed into one TCP. We wanna
demultiplex it and forward to designated backends. This means that there're
again lots of TCP connections between LB-backends, but we can make any
optimization there as we want.


> >Basically multiplexing should be done as close to network as possible. If
> it's done in application layer, for example when compression is used, such
> a load balancer need to decode the compression to see the >multiplexing
> information. That doesn't make them happy.
>
> This seems to assume that
>
> - either the LB is done above layer 4 and/or
> - the logical WS connections contained in the physical WS would need to be
> balanced to _different_ backend nodes.
>

Yes. Sorry that we haven't told the story in our mind in detail.


> >As Google, our motivation of mux is to make sure that total meaningful
> traffic per server cost (memory, port, machine, etc.) small enough compared
> to HTTP/Comet approach.
>
> I see. For me, I am mostly interested in the use case: 1 app doing both
> chatty and mass-data coms.
>

OK

--001a11c22202a0bdec04ddd4dc86
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Mon, May 27, 2013 at 9:51 PM, Tobias Oberstein <span di=
r=3D"ltr">&lt;<a href=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_bla=
nk">tobias.oberstein@tavendo.de</a>&gt;</span> wrote:<br><div class=3D"gmai=
l_extra">




<div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margi=
n:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204=
);border-left-style:solid;padding-left:1ex"><div>&gt;&gt;Having those buffe=
rs in user space (in app) allows one to tune the buffer size _per stream_. =
Thus, we can have large buffers for streams that carry mass data, and small=
 buffers for chatty streams. Thats a real &gt;&gt;advantage.<br>





<br>
&gt;Exactly.<br>
<br>
</div>In the meantime, I have read more about modern TCP/IP stacks.<br>
<br>
E.g. beginning with FreeBSD 7, the TCP/IP stack auto-tunes the TCP send/rec=
eive buffer sizes _per socket_. This is controlled via:<br>
<br>
net.inet.tcp.sendbuf_auto=3D1 =A0 =A0 =A0 =A0 # TCP send buffer size autotu=
ning (default: on)<br>
net.inet.tcp.sendbuf_inc=3D16384 =A0 =A0 =A0# TCP send buffer autotuning st=
ep size<br>
net.inet.tcp.recvbuf_auto=3D1 =A0 =A0 =A0 =A0 # TCP receive buffer size aut=
otuning (default: on)<br>
net.inet.tcp.recvbuf_inc=3D524288 =A0 =A0 # TCP receive buffer autotuning s=
tep size<br>
net.inet.tcp.sendbuf_max=3D16777216 =A0 # TCP send buffer maximum size (tun=
e up for Fat-Long-Pipes)<br>
net.inet.tcp.recvbuf_max=3D16777216 =A0 # TCP receive buffer maximum size (=
tune up for Fat-Long-Pipes)<br>
<br>
&quot;&quot;&quot;<br>
TCP Autotuning<br>
<br>
Beginning with Linux 2.6, Mac OSX 10.5, Windows Vista, and FreeBSD 7.0, bot=
h sender and receiver autotuning became available, eliminating the need to =
set the TCP send and receive buffers by hand for each path. However the max=
imum buffer sizes are still too small for many high-speed network path, and=
 must be increased as described on the pages for each operating system.<br>





&quot;&quot;&quot;<br>
<a href=3D"http://fasterdata.es.net/host-tuning/background/" target=3D"_bla=
nk">http://fasterdata.es.net/host-tuning/background/</a><br>
<br>
Apart from autotuning, there is also the option to control send/receive buf=
fer sizes _per socket_ using setsockopt() with SO_SNDBUF / SO_RCVBUF.<br>
<br>
=3D=3D=3D<br>
<br>
Given above, how would the following 2 approaches compare?<br>
<br>
a)<br>
- Client 1 &lt;=3D&gt; Server 1 connected with 2 TCP connections.<br>
- TCP autotuning on both client and server OR manually set buffer sizes: sm=
all for TCP-1, and large for TCP-2<br>
- First TCP _only_ carries &quot;chatty&quot; traffic, and second TCP carri=
es &quot;mass data&quot; traffic.<br>
<br>
b)<br>
- Client 1 &lt;=3D&gt; Server 1 connected with 1 TCP connection<br>
- TCP autotuning OR large buffers set system wide<br>
- multiplexing done over the single TCP that carries both &quot;chatty&quot=
; and &quot;mass data&quot; traffic and<br>
- the multiplexing is further optimizing buffer size (app level) and does p=
riorization of traffic (chatty =3D high-prio, mass =3D low-prio):<br>
<br>
&quot;Compare&quot;: I am mostly interested in the end-user experience:<br>
<br>
i) Is the chatty traffic still low-latency, even in the presence of concurr=
ent mass-data traffic?<br></blockquote><div><br></div><div>a) it depends on=
 process scheduler and network queue design of the server and client, I thi=
nk. I don&#39;t know much about per-socket buffer autotuning, but it probab=
ly just improves memory usage and throughput but doesn&#39;t have much effe=
ct on latency of chatty traffic?</div>



<div><br></div><div>b) the endpoints need to cut mass data into smaller chu=
nks to reserve slots for inserting chatty traffic. e.g. if acceptable delay=
 for chatty traffic is x ms, set the maximum send unit of mass traffic to (=
current through put) * x.</div>


<div><br></div><div>To allow one to control the peer&#39;s message scheduli=
ng, both per-socket buffer size and ws-mux flow control are insufficient, I=
 think. Needs to exchange ToS info.</div>
<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px=
 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left=
-style:solid;padding-left:1ex">
ii) Is the throughput of mass-data traffic concurrently to the chatty traff=
ic still (nearly) as high as when only doing mass data transfer?<br></block=
quote><div><br></div><div>I think minimum delay requirement for the chatty =
traffic determines fragmentation on mass-data traffic regardless it&#39;s d=
one in OS&#39;s network stack or ws-mux code.</div>


<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px=
 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left=
-style:solid;padding-left:1ex">

<div>&gt;&gt;How about having a shared Web worker in the background that ha=
ndles all messaging for the app over a single WS?<br>
&gt;One restriction of Web Worker is that we cannot share a worker between =
webapps with different origins while WebSocket can connect to a host differ=
ent from its origin.<br>
<br>
</div>Ah, right. Same origin policy. Thus, with WS-MUX, there is _more_ pot=
entially TCP sharing than whats possible with shared WebWorkers.<br>
<br>
However, for a given single app (from a single origin) running in multiple =
browser tabs, the shared WebWorker seems not so bad. Works today. No need f=
or WS-MUX or HTTP 2.0 MUX + WS over HTTP 2.0.<br></blockquote><div><br>




</div><div>Yes. Thanks for reminding us about shared worker approach. We ha=
ven&#39;t evaluated it well.</div><div>=A0</div><blockquote class=3D"gmail_=
quote" style=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-=
color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">





<div>&gt;For load balancer vendors, it&#39;s better that multiplexing is st=
andardized than each service provider develops its own.<br>
<br>
</div>I don&#39;t have experience running G scale infrastructure. However, =
I&#39;d like to understand why LBs need to have a look into WS at all .. wh=
y not Layer-4 LB? Doing balancing based on hash of source IP?<br></blockquo=
te>




<div><br></div><div>Traffic for various services would be multiplexed into =
one TCP. We wanna demultiplex it and forward to designated backends. This m=
eans that there&#39;re again lots of TCP connections between LB-backends, b=
ut we can make any optimization there as we want.</div>




<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px=
 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left=
-style:solid;padding-left:1ex">
<div>&gt;Basically multiplexing should be done as close to network as possi=
ble.=A0If it&#39;s done in application layer, for example when compression =
is used, such a load balancer need to decode the compression to see the &gt=
;multiplexing information. That doesn&#39;t make them happy.<br>





<br>
</div>This seems to assume that<br>
<br>
- either the LB is done above layer 4 and/or<br>
- the logical WS connections contained in the physical WS would need to be =
balanced to _different_ backend nodes.<br></blockquote><div><br></div><div>=
Yes. Sorry that we haven&#39;t told the story in our mind in detail.</div>




<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px=
 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left=
-style:solid;padding-left:1ex">
<div>&gt;As Google, our motivation of mux is to make sure that total meanin=
gful traffic per server cost (memory, port, machine, etc.) small enough com=
pared to HTTP/Comet approach.<br>
=A0<br>
</div>I see. For me, I am mostly interested in the use case: 1 app doing bo=
th chatty and mass-data coms.<br></blockquote><div><br></div><div>OK</div><=
/div><br></div></div>

--001a11c22202a0bdec04ddd4dc86--

From ricea@google.com  Wed May 29 21:12:03 2013
Return-Path: <ricea@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 92BE121F8FB3 for <hybi@ietfa.amsl.com>; Wed, 29 May 2013 21:12:03 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.376
X-Spam-Level: 
X-Spam-Status: No, score=-1.376 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, J_CHICKENPOX_24=0.6, NO_RELAYS=-0.001, WEIRD_PORT=0.001]
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 rJAPrDErXjDA for <hybi@ietfa.amsl.com>; Wed, 29 May 2013 21:12:02 -0700 (PDT)
Received: from mail-bk0-x22d.google.com (mail-bk0-x22d.google.com [IPv6:2a00:1450:4008:c01::22d]) by ietfa.amsl.com (Postfix) with ESMTP id 2BA7511E80A2 for <hybi@ietf.org>; Wed, 29 May 2013 21:12:01 -0700 (PDT)
Received: by mail-bk0-f45.google.com with SMTP id je9so4636557bkc.32 for <hybi@ietf.org>; Wed, 29 May 2013 21:12:01 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=2Gqa62vkwEscMwNSQQXCZ/3uD7BK4B5o6PS37hvSpOU=; b=adw/ePRZdOg0J3yHmBNbUO9fJ3SxOIZGgdEDLHoJGKRgrmFp5GjZav/FL9aWF1r252 PIcTSCWAwEHFrhWQTyKgUAuKwdzhy3/9robfeyAr7QKjWRobf9e4V1RoDcfGv+htozRZ hw5I6TQC7cp+cWt8vpwrqSA/7SD+bGV3bwuYnwO/c+K3h0RuhRshU7atjEhpwxHyJmdf rJbWzp93m67umnWn81oooTdWjS7o4n58IhckCK/JRu9fN0NoY5Wu+U2AajZgW43QJjJa biUkhMldiUbh665C85p/tSkBw5szoA1z1Iygxvn0lPJHAZwkZyG8zSIfyav9qcIPXfI9 1fag==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:x-gm-message-state; bh=2Gqa62vkwEscMwNSQQXCZ/3uD7BK4B5o6PS37hvSpOU=; b=apJccTgnUaXCrbcQ75WQx37KNrLln/GGoddtG993V51Y/1iDoCIG7mCHXIHCyh2g+5 aQdRjDC0HpB3uayc/1V37SkrT6R71F2kyTiUjcpYiXEV2WVVkI8RDaVIOatYkcAUAJXr BTmmA0qQlpPKQpO0LpptwfBQnfsWdDWiEKHIMTtwR+vyxEyRi63LvbPdqKsHaSyh7eKI k+R5Mfr5z0Px0/ocU6EOVHuYNJ2dxWg6zX3P6wTW9QQ9UDZ1A+WlfqgV9jE2uhPMAsdg FxB3qmGtylrhnLwmcHgy2HZg/aSpQ+w3MPpHVkYHoSgKP1x6BRWC580sm4Ar620bHDJd GkfA==
MIME-Version: 1.0
X-Received: by 10.205.115.73 with SMTP id fd9mr1147932bkc.58.1369887121039; Wed, 29 May 2013 21:12:01 -0700 (PDT)
Received: by 10.204.26.140 with HTTP; Wed, 29 May 2013 21:12:00 -0700 (PDT)
In-Reply-To: <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net> <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net>
Date: Thu, 30 May 2013 13:12:00 +0900
Message-ID: <CAHixhFrTk79A07BjQCgvep_+bmA4rGG1ZvqmoS6gsQYNPyPoZA@mail.gmail.com>
From: Adam Rice <ricea@google.com>
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
Content-Type: multipart/alternative; boundary=14dae9c0971c3bbed704dde7b3cf
X-Gm-Message-State: ALoCoQlmMsRNDf9ZL5bv09aXIym1Q5JPHsBJrt/0+c/3QT5KRUsDiQibz2e0fCtG+c38o1SJkDWczxmuzBFDB+je4EVzIF59/SlSwrBbpGGN3ynTcTdOB7Q2rJNKlkSWc/K7BNml5GI/FuI0Gr5qbr3FRfMt0qSljQFcx2YeCO4VrdAlHXt0f4UKCNIaYB11xn4gGNFs5G3i
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 04:12:03 -0000

--14dae9c0971c3bbed704dde7b3cf
Content-Type: text/plain; charset=ISO-8859-1

On 27 May 2013 20:59, Tobias Oberstein <tobias.oberstein@tavendo.de> wrote:

> I had a look into the MUX RFC again .. couldn't find anser to the
> following (may have missed sth):
>
> What WS addresses would be eligible to be multiplexed over a single WS
> connection?
>
> a) ws://somehost.com:999/
> b) ws://somehost.com:999/foo
> c) ws://somehost.com:999/bar
>
> d) wss://somehost.com:999/
> e) wss://somehost.com:999/foo
> f) wss://somehost.com:999/bar
>
> All of a) - f) are to the same target IP:port and hence could share the
> same TCP.
>
> However, d) - f) use wss, and hence have a TLS handshake right after TCP
> establishment.
>

While you certainly can serve both ws: and wss: on a single port, and I can
see cases where you might want to, I would not in general consider it to be
a good idea.


> So d) - f) cannot be multiplexed over the same physical WS as a) - c)?
>
> Or can an implementation just "silently" transport a)-c) also over wss,
> and hence multiplex all of a) - f) over 1 physical WS?
>

The handshake does not currently include the schema, so there would be no
way to communicate to the server that a)-c) were supposed to be ws:, not
wss:.

Even if this was amended, both client and server would have to be careful
that no ambient authority leaked from the wss: channels to the ws:
channels. For example: the client would have to be careful not to send
"secure" cookies with the ws: handshakes, and the server would have to be
careful not to apply any authority contained in a client TLS certificate to
the ws: logical channels.

For this reason, I think it would be easiest not to attempt to multiplex
ws: and wss: onto a single TCP/IP connection.


> Lastly, a)-c) are to the same target IP:port and also WS schema (ws, not
> wss) - and hence can be multiplexed over 1 physical WS even though they are
> to different URL paths?
>

Yes. These semantics are inherited from HTTP/1.1.

Adam Rice

--14dae9c0971c3bbed704dde7b3cf
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On 27 May 2013 20:59, Tobias Oberstein <span dir=3D"ltr">&=
lt;<a href=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_blank" class=
=3D"cremed">tobias.oberstein@tavendo.de</a>&gt;</span> wrote:<br><div class=
=3D"gmail_extra">
<div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margi=
n:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I had a look into the MUX RFC again .. couldn&#39;t find anser to the follo=
wing (may have missed sth):<br>
<br>
What WS addresses would be eligible to be multiplexed over a single WS conn=
ection?<br>
<br>
a) ws://<a href=3D"http://somehost.com:999/" target=3D"_blank" class=3D"cre=
med">somehost.com:999/</a><br>
b) ws://<a href=3D"http://somehost.com:999/foo" target=3D"_blank" class=3D"=
cremed">somehost.com:999/foo</a><br>
c) ws://<a href=3D"http://somehost.com:999/bar" target=3D"_blank" class=3D"=
cremed">somehost.com:999/bar</a><br>
<br>
d) wss://<a href=3D"http://somehost.com:999/" target=3D"_blank" class=3D"cr=
emed">somehost.com:999/</a><br>
e) wss://<a href=3D"http://somehost.com:999/foo" target=3D"_blank" class=3D=
"cremed">somehost.com:999/foo</a><br>
f) wss://<a href=3D"http://somehost.com:999/bar" target=3D"_blank" class=3D=
"cremed">somehost.com:999/bar</a><br>
<br>
All of a) - f) are to the same target IP:port and hence could share the sam=
e TCP.<br>
<br>
However, d) - f) use wss, and hence have a TLS handshake right after TCP es=
tablishment.<br></blockquote><div><br></div><div>While you certainly can se=
rve both ws: and wss: on a single port, and I can see cases where you might=
 want to, I would not in general consider it to be a good idea.</div>
<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex">
So d) - f) cannot be multiplexed over the same physical WS as a) - c)?<br>
<br>
Or can an implementation just &quot;silently&quot; transport a)-c) also ove=
r wss, and hence multiplex all of a) - f) over 1 physical WS?<br></blockquo=
te><div><br></div><div>The handshake does not currently include the schema,=
 so there would be no way to communicate to the server that a)-c) were supp=
osed to be ws:, not wss:.</div>
<div><br></div><div>Even if this was amended, both client and server would =
have to be careful that no ambient authority leaked from the wss: channels =
to the ws: channels. For example: the client would have to be careful not t=
o send &quot;secure&quot; cookies with the ws: handshakes, and the server w=
ould have to be careful not to apply any authority contained in a client TL=
S certificate to the ws: logical channels.</div>
<div><br></div><div>For this reason, I think it would be easiest not to att=
empt to multiplex ws: and wss: onto a single TCP/IP connection.</div><div>=
=A0<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;b=
order-left:1px #ccc solid;padding-left:1ex">

Lastly, a)-c) are to the same target IP:port and also WS schema (ws, not ws=
s) - and hence can be multiplexed over 1 physical WS even though they are t=
o different URL paths?<br></blockquote><div><br></div><div>Yes. These seman=
tics are inherited from HTTP/1.1.</div>
<div><br></div><div>Adam Rice</div></div></div></div>

--14dae9c0971c3bbed704dde7b3cf--

From tyoshino@google.com  Thu May 30 00:17:00 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BF1CE21F9590 for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 00:17:00 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.458
X-Spam-Level: 
X-Spam-Status: No, score=-1.458 tagged_above=-999 required=5 tests=[AWL=0.519,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 dWwFRncA6IEC for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 00:17:00 -0700 (PDT)
Received: from mail-ea0-x22b.google.com (mail-ea0-x22b.google.com [IPv6:2a00:1450:4013:c01::22b]) by ietfa.amsl.com (Postfix) with ESMTP id 193A121F88EA for <hybi@ietf.org>; Thu, 30 May 2013 00:16:59 -0700 (PDT)
Received: by mail-ea0-f171.google.com with SMTP id b15so6243842eae.30 for <hybi@ietf.org>; Thu, 30 May 2013 00:16:59 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:from:date:message-id:subject:to:content-type; bh=uJ0Xt2+SJghwj94wtSocI/+zT+Td3zg4L93u/+ymB4Y=; b=HmsKBeF0Qrd8z8GMJCI01icqSNBnkyovRAprwIZxhbAZ9DFn/cuO43FiTtoTssnuEA b+UPtZsqMK8Y/EIC2G3KNy4wBREMenyEJXvhJCjIg1gfVSi0HG2Vq+aiETpW4RDQBNv8 jSVsmGU2lNG1c1bA20eipBy7xkKx4/UPRR0UR0m7hb2ew83cABCWrxS8Bt02j2W1X4r4 onBtAWDqb1YhudPTjbBInOsYGs4HAPiOw/IMOVhNc4m2fYZGLRqMloev5ZgVBQUjLn+r MR8FBa9XGmYhC+j+GsKZr3nboNie1CsXbgWCFKLJ05QDVzRbqDQDGpH7sNThA+WrTQtZ 1Yvg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:from:date:message-id:subject:to:content-type :x-gm-message-state; bh=uJ0Xt2+SJghwj94wtSocI/+zT+Td3zg4L93u/+ymB4Y=; b=hSzumZndthQBfQwiflGWFnw0x4gvXeu3auO9eKsbdYi4i3uA0xKyhgBjPNpNzirqeH hz8mOHTEAbgd1w+p9FYYtOW2AqgcjG+QzO8fKIZo/e+mmUhb8B9cEi/SGRFAXHDm+kt6 ofxngH7vJAs57k97xOhlt8fOprRj38ZOKLokZDZ+pPTa9Mwqs/7Nah9wIdCqG1gvwDLF ztJ8p0kM5vrFLc5tswr/O3cktDbl1z7ftR8TbsGUTAwv2qyAytZRVtREoVYuGcmuD5pc S5cqX0pjnzQ92hXDBWzk149/GoHeMWcRoP8mBarQ+tFItjL5lbZ3Y21XgyKwvs5qHy5N 0Bow==
X-Received: by 10.14.194.73 with SMTP id l49mr8550611een.99.1369898218837; Thu, 30 May 2013 00:16:58 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.223.86.67 with HTTP; Thu, 30 May 2013 00:16:38 -0700 (PDT)
From: Takeshi Yoshino <tyoshino@google.com>
Date: Thu, 30 May 2013 16:16:38 +0900
Message-ID: <CAH9hSJaUnFRve-JMUm1pzdxreHsd5R13gUa0s9_S2u1t6Y87rQ@mail.gmail.com>
To: "hybi@ietf.org" <hybi@ietf.org>
Content-Type: multipart/alternative; boundary=047d7b343e3ab690bc04ddea482f
X-Gm-Message-State: ALoCoQmFL6qmXss5788d5iteBmcqIWz6N43DovS2Sd8UumauyeF9G7t2sW8AOv5rNfb/grtiiMWtGEwf6qdkEtIWNxBqZ8Bz7x3ARFdTHKZSUDJL2/f9RVffpdjG2hwNWiEqDtqTyCFbJhFcW5o7xPqv1WSPTZciJjSkD1UBpgaVtthkij0IQMn9cwhL4UnHlxNqfCUvPVw9
Subject: [hybi] Simplify WebSocket multiplexing
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 07:17:00 -0000

--047d7b343e3ab690bc04ddea482f
Content-Type: text/plain; charset=ISO-8859-1

(Branching from the thread "Call for interest: multiplexing dedicated for
WebSocket")

It seemed that we had rough consensus on accelerating WebSocket dedicated
multiplexing by simplifying it if we keep working on it.

I'd like to hear your opinion on this in this separate thread. Below listed
some points we could drop/simplify including both ones suggested in the
thread and some I came up with now.

1) Header delta encoding in AddChannelRequest
2) Header delta encoding in AddChannelResponse
3) Channel slot
4) Only one mux control in one message
5) Use the length encoding for encoding mux channel ID
6) Drop flow control

----

My opinion is to do 1, 2 and 4.

--047d7b343e3ab690bc04ddea482f
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">(Branching from the thread &quot;Call for interest: multip=
lexing dedicated for WebSocket&quot;)<div><br></div><div style>It seemed th=
at we had rough consensus on accelerating WebSocket dedicated multiplexing =
by simplifying it if we keep working on it.</div>

<div><br></div><div style>I&#39;d like to hear your opinion on this in this=
 separate thread. Below listed some points we could drop/simplify including=
 both ones suggested in the thread and some I came up with now.</div><div>

<br></div><div><div>1) Header delta encoding in AddChannelRequest</div><div=
>2) Header delta encoding in AddChannelResponse</div><div>3) Channel slot</=
div><div style>4) Only one mux control in one message</div></div><div style=
>

5) Use the length encoding for encoding mux channel ID</div><div style>6) D=
rop flow control</div><div style><br></div><div style>----<br></div><div st=
yle><br></div><div style>My opinion is to do 1, 2 and 4.</div></div>

--047d7b343e3ab690bc04ddea482f--

From tyoshino@google.com  Thu May 30 00:28:37 2013
Return-Path: <tyoshino@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 113EE21F9602 for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 00:28:37 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.545
X-Spam-Level: 
X-Spam-Status: No, score=-1.545 tagged_above=-999 required=5 tests=[AWL=0.432,  BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 nKa-IOYdijkH for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 00:28:36 -0700 (PDT)
Received: from mail-we0-x22d.google.com (mail-we0-x22d.google.com [IPv6:2a00:1450:400c:c03::22d]) by ietfa.amsl.com (Postfix) with ESMTP id 33F6D21F962E for <hybi@ietf.org>; Thu, 30 May 2013 00:28:34 -0700 (PDT)
Received: by mail-we0-f173.google.com with SMTP id p57so6932678wes.4 for <hybi@ietf.org>; Thu, 30 May 2013 00:28:34 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=vQXdUWNkT/MdeJhlJ84Po9vG9I/392Kaquigg7wdEzU=; b=esgvbNqQPzMbFIrbp3aOobnGXwkByOPbZ19pjpnRsuKkcyKVAc+Hu8cZiTQL7D2ppY vwsTyf8lAwvLIab5MoqG1PFJTKAcKHwnwyFT78PmVXo4Pf4bX7h39pAB9yekiBaa+OEy 1P/Tj1yxuVGfPvdk5LfS2GU8k1Rfd9Faggtn403IWKB9rH6Uq3BFJcZIqXhzsmPb5huV xFSs79QjCZOLmKKUs0oLcN7ep3jFVq8ETJLYzJsNNnKdUtSOzJyU9nQleA/Nhfw08qfx BdBwGiBgTfg3z6iR+dkRkdFkUsu5SoQiISXTG+07s9ifNO8OeKaP9/69LcR2x0InoYEQ vyPQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:x-gm-message-state; bh=vQXdUWNkT/MdeJhlJ84Po9vG9I/392Kaquigg7wdEzU=; b=Ex1WFpaKE3fIqppq1fxNAI+ZMfMIP7rTb5R/Cezqd5RYElk7WtbRypYC18tcx3Ui01 cQPFZgiSmdxkvXfC1o5rE2e7DnkCEaeMYuvPMQWj2e7lA1Vjk17rWHD+urCNe6a8V3FC aIF8yQI5yLtxQ+nVGTn7wepsp7Fa+yOG88tCF75oKZUQ1ZnoVyqLf6Q4YjgoofGJeePE cskeuf4EJTTaf3IbpPvUsAOSUemYmR/HThgm1Adu43rUzAlZQ7URLbVMnYnXC1YOX2CB VprS+Xfw7/Uae3ymCZZx1egZwifRSoWTqGAWYUt3XjyEWR1QhSyluVPpxc55FH2fCj3V C5Ew==
X-Received: by 10.194.87.71 with SMTP id v7mr3263331wjz.33.1369898914306; Thu, 30 May 2013 00:28:34 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.180.5.136 with HTTP; Thu, 30 May 2013 00:28:14 -0700 (PDT)
In-Reply-To: <CAHixhFrTk79A07BjQCgvep_+bmA4rGG1ZvqmoS6gsQYNPyPoZA@mail.gmail.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net> <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net> <CAHixhFrTk79A07BjQCgvep_+bmA4rGG1ZvqmoS6gsQYNPyPoZA@mail.gmail.com>
From: Takeshi Yoshino <tyoshino@google.com>
Date: Thu, 30 May 2013 16:28:14 +0900
Message-ID: <CAH9hSJYFa+bqN=e7x87W+Xvq-st70nbzUXniQaPme2fzspCjWA@mail.gmail.com>
To: Adam Rice <ricea@google.com>
Content-Type: multipart/alternative; boundary=047d7bf10ab02a702704ddea72ed
X-Gm-Message-State: ALoCoQmkUuPs89OKr7khGWNLEtsBvdxXUs1mVFYWODuuseU23d9UNjB/D6cUbgdvhjug3X/62DZEGd3SQmx2QP9vcLAEuUNT98v4IU/AoascIXSN2cpXMYgI2dRbTVTjd4y4IHicAYGauBVc5FEUCJK7ztpYXEHVRp6VFNuRcJD0BlQmwW2m4yuW39yooWIJL1q0CbPGPaJC
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 07:28:37 -0000

--047d7bf10ab02a702704ddea72ed
Content-Type: text/plain; charset=ISO-8859-1

On Thu, May 30, 2013 at 1:12 PM, Adam Rice <ricea@google.com> wrote:

> So d) - f) cannot be multiplexed over the same physical WS as a) - c)?
>>
>> Or can an implementation just "silently" transport a)-c) also over wss,
>> and hence multiplex all of a) - f) over 1 physical WS?
>>
>
> The handshake does not currently include the schema, so there would be no
> way to communicate to the server that a)-c) were supposed to be ws:, not
> wss:.
>

Right. I missed that point.


>  Even if this was amended, both client and server would have to be
> careful that no ambient authority leaked from the wss: channels to the ws:
> channels. For example: the client would have to be careful not to send
> "secure" cookies with the ws: handshakes, and the server would have to be
> careful not to apply any authority contained in a client TLS certificate to
> the ws: logical channels.
>
> For this reason, I think it would be easiest not to attempt to multiplex
> ws: and wss: onto a single TCP/IP connection.
>

Compression contexts, etc. also need to be isolated carefully to protect it
from attack like CRIME.

--047d7bf10ab02a702704ddea72ed
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Thu, May 30, 2013 at 1:12 PM, Adam Rice <span dir=3D"lt=
r">&lt;<a href=3D"mailto:ricea@google.com" target=3D"_blank">ricea@google.c=
om</a>&gt;</span> wrote:<br><div class=3D"gmail_extra"><div class=3D"gmail_=
quote"><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-=
left:1px #ccc solid;padding-left:1ex">

<div dir=3D"ltr"><div class=3D"gmail_extra"><div class=3D"gmail_quote"><div=
 class=3D"im"><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex">So d) - f) cannot be multiplex=
ed over the same physical WS as a) - c)?<br>


<br>
Or can an implementation just &quot;silently&quot; transport a)-c) also ove=
r wss, and hence multiplex all of a) - f) over 1 physical WS?<br></blockquo=
te><div><br></div></div><div>The handshake does not currently include the s=
chema, so there would be no way to communicate to the server that a)-c) wer=
e supposed to be ws:, not wss:.</div>

</div></div></div></blockquote><div><br></div><div style>Right. I missed th=
at point.</div><div style>=A0<br></div><blockquote class=3D"gmail_quote" st=
yle=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div =
dir=3D"ltr">

<div class=3D"gmail_extra"><div class=3D"gmail_quote">
<div>Even if this was amended, both client and server would have to be care=
ful that no ambient authority leaked from the wss: channels to the ws: chan=
nels. For example: the client would have to be careful not to send &quot;se=
cure&quot; cookies with the ws: handshakes, and the server would have to be=
 careful not to apply any authority contained in a client TLS certificate t=
o the ws: logical channels.<br>

</div>
<div><br></div><div>For this reason, I think it would be easiest not to att=
empt to multiplex ws: and wss: onto a single TCP/IP connection.</div></div>=
</div></div></blockquote><div><br></div><div style>Compression contexts, et=
c. also need to be isolated carefully to protect it from attack like CRIME.=
</div>

<div>=A0</div></div></div></div>

--047d7bf10ab02a702704ddea72ed--

From sustrik@250bpm.com  Thu May 30 01:13:10 2013
Return-Path: <sustrik@250bpm.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8AAF921F97D8 for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 01:13:10 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 1.72
X-Spam-Level: *
X-Spam-Status: No, score=1.72 tagged_above=-999 required=5 tests=[BAYES_40=-0.185, HELO_EQ_SK=1.35, HOST_EQ_SK=0.555]
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 rzSF4vczDSGW for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 01:13:05 -0700 (PDT)
Received: from mail.moloch.sk (chrocht.moloch.sk [62.176.169.44]) by ietfa.amsl.com (Postfix) with ESMTP id 5F3A121F973F for <hybi@ietf.org>; Thu, 30 May 2013 01:13:05 -0700 (PDT)
Received: by mail.moloch.sk (Postfix, from userid 107) id 4EC19180008F; Thu, 30 May 2013 10:13:03 +0200 (CEST)
To: <hybi@ietf.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Date: Thu, 30 May 2013 10:13:03 +0200
From: Martin Sustrik <sustrik@250bpm.com>
In-Reply-To: <CAH9hSJaUnFRve-JMUm1pzdxreHsd5R13gUa0s9_S2u1t6Y87rQ@mail.gmail.com>
References: <CAH9hSJaUnFRve-JMUm1pzdxreHsd5R13gUa0s9_S2u1t6Y87rQ@mail.gmail.com>
Message-ID: <2edca26ed71bbd097ccc16fcfa32bcb7@webmail.moloch.sk>
X-Sender: sustrik@250bpm.com
User-Agent: Roundcube Webmail/0.8.1
Subject: Re: [hybi] Simplify WebSocket multiplexing
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 08:13:10 -0000

On 2013-05-30 09:16, Takeshi Yoshino wrote:

> 6) Drop flow control

Without flow control, pushback on one channel will block all other 
channels, which defeats the whole point of multiplexing. Not a good idea 
IMO.

Martin

From tobias.oberstein@tavendo.de  Thu May 30 01:30:37 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A6ACC21F96B5 for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 01:30:37 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.598
X-Spam-Level: 
X-Spam-Status: No, score=-2.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OCl4zMapJNWE for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 01:30:32 -0700 (PDT)
Received: from EXHUB020-3.exch020.serverdata.net (exhub020-3.exch020.serverdata.net [206.225.164.30]) by ietfa.amsl.com (Postfix) with ESMTP id E91A121F8A03 for <hybi@ietf.org>; Thu, 30 May 2013 01:30:31 -0700 (PDT)
Received: from [109.193.209.160] (109.193.209.160) by smtpx20.serverdata.net (206.225.164.35) with Microsoft SMTP Server (TLS) id 8.3.264.0; Thu, 30 May 2013 01:30:30 -0700
Message-ID: <51A70DF9.5010601@tavendo.de>
Date: Thu, 30 May 2013 10:29:45 +0200
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: Takeshi Yoshino <tyoshino@google.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net> <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net> <CAHixhFrTk79A07BjQCgvep_+bmA4rGG1ZvqmoS6gsQYNPyPoZA@mail.gmail.com> <CAH9hSJYFa+bqN=e7x87W+Xvq-st70nbzUXniQaPme2fzspCjWA@mail.gmail.com>
In-Reply-To: <CAH9hSJYFa+bqN=e7x87W+Xvq-st70nbzUXniQaPme2fzspCjWA@mail.gmail.com>
Content-Type: multipart/alternative; boundary="------------080708080102070705040802"
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 08:30:37 -0000

--------------080708080102070705040802
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
Content-Transfer-Encoding: 7bit

Am 30.05.2013 09:28, schrieb Takeshi Yoshino:
> On Thu, May 30, 2013 at 1:12 PM, Adam Rice <ricea@google.com 
> <mailto:ricea@google.com>> wrote:
>
>         So d) - f) cannot be multiplexed over the same physical WS as
>         a) - c)?
>
>         Or can an implementation just "silently" transport a)-c) also
>         over wss, and hence multiplex all of a) - f) over 1 physical WS?
>
>
>     The handshake does not currently include the schema, so there
>     would be no way to communicate to the server that a)-c) were
>     supposed to be ws:, not wss:.
>
>
> Right. I missed that point.
>
>     Even if this was amended, both client and server would have to be
>     careful that no ambient authority leaked from the wss: channels to
>     the ws: channels. For example: the client would have to be careful
>     not to send "secure" cookies with the ws: handshakes, and the
>     server would have to be careful not to apply any authority
>     contained in a client TLS certificate to the ws: logical channels.
>
>     For this reason, I think it would be easiest not to attempt to
>     multiplex ws: and wss: onto a single TCP/IP connection.
>
>
> Compression contexts, etc. also need to be isolated carefully to 
> protect it from attack like CRIME.

I also agree with Adam. To summarize, given the possible "issues" of

- client cert based auth
- "secure" cookies / ambient authority
- compression contexts / CRIME

(plus complexities like: what if an app opens ws://domain.com, and then 
_later_ opens wss://domain.com? Upgrade to TLS not possible anymore .. 
cannot TLS handshake in the middle of an already established non-TLS TCP 
where some traffic already traveled .. the WS handshake at least.)

I think it would be good to include explicit text into the MUX draft, e.g.

"""
An implementation MUST NOT multiplex multiple logical channels to 
WebSocket addresses with different schemes (ws: versus wss:) on a single 
physical WebSocket connection.

For example, two WebSocket connections to ws://domain.com and 
wss://domain.com must not be shared on a single physical WebSocket 
connection to domain.com.
"""

I also think it would be helpful to include explicit text describing 
what WebSocket addresses are eligible for sharing, e.g.

"""
An implementation MAY multiplex multiple logical channels to different 
WebSocket addresses if and only if the WebSocket addresses share the 
same WebSocket scheme, the hostname resolves to the same IP address and 
the same port.

For example, connections to all of the following WebSocket addresses may 
be multiplexed over a single physical WebSocket connection (assum 
domain.com, domain2.com and sub1.domain.com all resolve to the same IP 
address):

ws://domain.com
ws://domain.com:80
ws://domain.com/foo
ws://domain.com:80/foo
ws://domain.com/bar
ws://domain2.com
ws://sub1.domain.com
ws://sub1.domain.com:80
ws://sub1.domain.com:80/bar

Note: When a WS address does not specify an explicit port, the port is 
the default port for the given WS scheme (ws: = 80 and wss: = 443).
"""




/Tobias


--------------080708080102070705040802
Content-Type: text/html; charset="ISO-8859-1"
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <div class="moz-cite-prefix">Am 30.05.2013 09:28, schrieb Takeshi
      Yoshino:<br>
    </div>
    <blockquote
cite="mid:CAH9hSJYFa+bqN=e7x87W+Xvq-st70nbzUXniQaPme2fzspCjWA@mail.gmail.com"
      type="cite">
      <div dir="ltr">On Thu, May 30, 2013 at 1:12 PM, Adam Rice <span
          dir="ltr">&lt;<a moz-do-not-send="true"
            href="mailto:ricea@google.com" target="_blank">ricea@google.com</a>&gt;</span>
        wrote:<br>
        <div class="gmail_extra">
          <div class="gmail_quote">
            <blockquote class="gmail_quote" style="margin:0 0 0
              .8ex;border-left:1px #ccc solid;padding-left:1ex">
              <div dir="ltr">
                <div class="gmail_extra">
                  <div class="gmail_quote">
                    <div class="im">
                      <blockquote class="gmail_quote" style="margin:0 0
                        0 .8ex;border-left:1px #ccc
                        solid;padding-left:1ex">So d) - f) cannot be
                        multiplexed over the same physical WS as a) -
                        c)?<br>
                        <br>
                        Or can an implementation just "silently"
                        transport a)-c) also over wss, and hence
                        multiplex all of a) - f) over 1 physical WS?<br>
                      </blockquote>
                      <div><br>
                      </div>
                    </div>
                    <div>The handshake does not currently include the
                      schema, so there would be no way to communicate to
                      the server that a)-c) were supposed to be ws:, not
                      wss:.</div>
                  </div>
                </div>
              </div>
            </blockquote>
            <div><br>
            </div>
            <div style="">Right. I missed that point.</div>
            <div style="">&nbsp;<br>
            </div>
            <blockquote class="gmail_quote" style="margin:0 0 0
              .8ex;border-left:1px #ccc solid;padding-left:1ex">
              <div dir="ltr">
                <div class="gmail_extra">
                  <div class="gmail_quote">
                    <div>Even if this was amended, both client and
                      server would have to be careful that no ambient
                      authority leaked from the wss: channels to the ws:
                      channels. For example: the client would have to be
                      careful not to send "secure" cookies with the ws:
                      handshakes, and the server would have to be
                      careful not to apply any authority contained in a
                      client TLS certificate to the ws: logical
                      channels.<br>
                    </div>
                    <div><br>
                    </div>
                    <div>For this reason, I think it would be easiest
                      not to attempt to multiplex ws: and wss: onto a
                      single TCP/IP connection.</div>
                  </div>
                </div>
              </div>
            </blockquote>
            <div><br>
            </div>
            <div style="">Compression contexts, etc. also need to be
              isolated carefully to protect it from attack like CRIME.</div>
            <div>&nbsp;</div>
          </div>
        </div>
      </div>
    </blockquote>
    <br>
    I also agree with Adam. To summarize, given the possible "issues" of<br>
    <br>
    - client cert based auth<br>
    - "secure" cookies / ambient authority<br>
    - compression contexts / CRIME<br>
    <br>
    (plus complexities like: what if an app opens <a class="moz-txt-link-freetext" href="ws://domain.com">ws://domain.com</a>, and
    then _later_ opens <a class="moz-txt-link-freetext" href="wss://domain.com">wss://domain.com</a>? Upgrade to TLS not possible
    anymore .. cannot TLS handshake in the middle of an already
    established non-TLS TCP where some traffic already traveled .. the
    WS handshake at least.)<br>
    <br>
    I think it would be good to include explicit text into the MUX
    draft, e.g.<br>
    <br>
    """<br>
    An implementation MUST NOT multiplex multiple logical channels to
    WebSocket addresses with different schemes (ws: versus wss:) on a
    single physical WebSocket connection.<br>
    <br>
    For example, two WebSocket connections to <a class="moz-txt-link-freetext" href="ws://domain.com">ws://domain.com</a> and
    <a class="moz-txt-link-freetext" href="wss://domain.com">wss://domain.com</a> must not be shared on a single physical WebSocket
    connection to domain.com.<br>
    """<br>
    <br>
    I also think it would be helpful to include explicit text describing
    what WebSocket addresses are eligible for sharing, e.g.<br>
    <br>
    """<br>
    An implementation MAY multiplex multiple logical channels to
    different WebSocket addresses if and only if the WebSocket addresses
    share the same WebSocket scheme, the hostname resolves to the same
    IP address and the same port.<br>
    <br>
    For example, connections to all of the following WebSocket addresses
    may be multiplexed over a single physical WebSocket connection
    (assum domain.com, domain2.com and sub1.domain.com all resolve to
    the same IP address):<br>
    <br>
    <a class="moz-txt-link-freetext" href="ws://domain.com">ws://domain.com</a><br>
    <a class="moz-txt-link-freetext" href="ws://domain.com:80">ws://domain.com:80</a><br>
    <a class="moz-txt-link-freetext" href="ws://domain.com/foo">ws://domain.com/foo</a><br>
    <a class="moz-txt-link-freetext" href="ws://domain.com:80/foo">ws://domain.com:80/foo</a><br>
    <a class="moz-txt-link-freetext" href="ws://domain.com/bar">ws://domain.com/bar</a><br>
    <a class="moz-txt-link-freetext" href="ws://domain2.com">ws://domain2.com</a><br>
    <a class="moz-txt-link-freetext" href="ws://sub1.domain.com">ws://sub1.domain.com</a><br>
    <a class="moz-txt-link-freetext" href="ws://sub1.domain.com:80">ws://sub1.domain.com:80</a><br>
    <a class="moz-txt-link-freetext" href="ws://sub1.domain.com:80/bar">ws://sub1.domain.com:80/bar</a><br>
    <br>
    Note: When a WS address does not specify an explicit port, the port
    is the default port for the given WS scheme (ws: = 80 and wss: =
    443).<br>
    """<br>
    <br>
    <br>
    <br>
    <br>
    /Tobias<br>
    <br>
  </body>
</html>

--------------080708080102070705040802--

From tobias.oberstein@tavendo.de  Thu May 30 03:11:41 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EA98F21F9473 for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 03:11:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.452
X-Spam-Level: 
X-Spam-Status: No, score=-1.452 tagged_above=-999 required=5 tests=[AWL=1.147,  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 H4ckGaElQxSu for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 03:11:36 -0700 (PDT)
Received: from EXHUB020-1.exch020.serverdata.net (exhub020-1.exch020.serverdata.net [206.225.164.28]) by ietfa.amsl.com (Postfix) with ESMTP id AED7621F91B8 for <hybi@ietf.org>; Thu, 30 May 2013 03:11:36 -0700 (PDT)
Received: from EXVMBX020-12.exch020.serverdata.net ([169.254.3.90]) by EXHUB020-1.exch020.serverdata.net ([206.225.164.28]) with mapi; Thu, 30 May 2013 03:11:35 -0700
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
To: Takeshi Yoshino <tyoshino@google.com>, "hybi@ietf.org" <hybi@ietf.org>
Date: Thu, 30 May 2013 03:11:38 -0700
Thread-Topic: [hybi] Simplify WebSocket multiplexing
Thread-Index: Ac5dBbQ3WPmIwuLHQzqzW2cnu7PaEQACnkhQ
Message-ID: <634914A010D0B943A035D226786325D4422DA06B10@EXVMBX020-12.exch020.serverdata.net>
References: <CAH9hSJaUnFRve-JMUm1pzdxreHsd5R13gUa0s9_S2u1t6Y87rQ@mail.gmail.com>
In-Reply-To: <CAH9hSJaUnFRve-JMUm1pzdxreHsd5R13gUa0s9_S2u1t6Y87rQ@mail.gmail.com>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: de-DE, en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Subject: Re: [hybi] Simplify WebSocket multiplexing
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 10:11:42 -0000

>I'd like to hear your opinion on this in this separate thread. Below liste=
d some points we could drop/simplify including >both ones suggested in the =
thread and some I came up with now.
>
>1) Header delta encoding in AddChannelRequest
>2) Header delta encoding in AddChannelResponse
>3) Channel slot
>4) Only one mux control in one message
>5) Use the length encoding for encoding mux channel ID
>6) Drop flow control
>
>My opinion is to do 1, 2 and 4.

+1 for dropping: 1, 2, 4

rgd. 5: an implementation needs code for such encoding already for base WS.=
 And it's trivial anyway. So IMHO no gain in dropping.

rgd. 3: no opinion right now .. I need to look further into the draft.

rgd. 6: without flow-control, the whole MUX effort is IMHO not worth the ef=
forts anymore. John already said that, and I agree.=20

=3D=3D

However, my current thinking is this:

WS-MUX might be designed with the following features:

a) no flow-control and no priorization (QoS)
b) flow-control, but no priorization (QoS)
c) flow-control _and_ priorization (QoS)

=3D>

a) Not worth the effort.


b) I am still struggling with understanding the exact situations under whic=
h WS-MUX will improve things over whats possible today with stuff like shar=
ed Web workers, per-socket TCP autotuning and layer-4 load-balancers.

- An app served from a single origin can have all it's WS connections (from=
 multiple tabs etc) shared over a single TCP using a shared Web worker appr=
oach.

- An app where all WS traffic for a given frontend can be processed at a si=
ngle backend can efficiently scale-out using a layer-4 LB that distributes =
traffic based on consistent hash of source IP.

- An app that wants max. compatibility with network infrastructure will use=
 WSS and hence intermediaries (other than MITM proxies) won't be able to lo=
ok into the traffic anyway.

As Takeshi pointed out, there are 2 situations where above is insufficient:

i) a single app served from different origins ("widgets") =3D> cannot use s=
hared Web workers

ii) different apps from same vendor where the connections from the differen=
t apps need to go to different backend nodes ("services") =3D> needs an app=
/service/WSMUX-aware layer-7 LB
=20
Given i) and ii) "only" (are there more?), we (for our use cases) probably =
would have troubles justifying the efforts of doing WS-MUX ..


c)
Is this out of scope rgd WS-MUX?

Priorization/ToS: this is where we see use cases for us.

/Tobias

From tobias.oberstein@tavendo.de  Thu May 30 04:29:21 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B874521F9421 for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 04:28:59 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.298
X-Spam-Level: 
X-Spam-Status: No, score=-2.298 tagged_above=-999 required=5 tests=[AWL=-0.299, BAYES_00=-2.599, J_CHICKENPOX_52=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 WyyFNSkgzDbr for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 04:28:42 -0700 (PDT)
Received: from EXHUB020-3.exch020.serverdata.net (exhub020-3.exch020.serverdata.net [206.225.164.30]) by ietfa.amsl.com (Postfix) with ESMTP id 42DCB21F9410 for <hybi@ietf.org>; Thu, 30 May 2013 04:28:39 -0700 (PDT)
Received: from [109.193.209.160] (109.193.209.160) by smtpx20.serverdata.net (206.225.164.35) with Microsoft SMTP Server (TLS) id 8.3.264.0; Thu, 30 May 2013 04:28:36 -0700
Message-ID: <51A737E6.2090707@tavendo.de>
Date: Thu, 30 May 2013 13:28:38 +0200
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: Takeshi Yoshino <tyoshino@google.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <CAH9hSJZUG1f+3Uk=t2=A5i4O9=wPvAisspM=pgmGEH9emTL9-Q@mail.gmail.com> <CAH9hSJZai_UuxW4O6mZcEJT2DJoURtLo16XNci1qkYVWv4HVdg@mail.gmail.com> <007501ce56f0$67f74080$37e5c180$@noemax.com> <519CD6A1.7080708@ericsson.com> <519CE075.4000106@tavendo.de> <CAM5k6X9WmO80hiQZ6_GqK66PAd3of=2ZRi9=VrWj52apA1+=5g@mail.gmail.com> <519D02E7.6040009@tavendo.de> <CAM5k6X8Z-JxrgTy3NAc-wC7zK_AfsWAhqzNKaEY_yZzmz=pZcQ@mail.gmail.com> <519D2E2D.4080809@tavendo.de> <CAH9hSJZw7sLhQH6i2O9-rRKc88Bmh0EaJEcpxSB5kUKprP6YyQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C3DA775@EXVMBX020-12.exch020.serverdata.net> <CAH9hSJYms5rPSvDDNeMYoZ19UmRLJUd7YM7d98ONUkB4YZnM=Q@mail.gmail.com>
In-Reply-To: <CAH9hSJYms5rPSvDDNeMYoZ19UmRLJUd7YM7d98ONUkB4YZnM=Q@mail.gmail.com>
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 11:29:21 -0000

>     Given above, how would the following 2 approaches compare?
>
>     a)
>     - Client 1 <=> Server 1 connected with 2 TCP connections.
>     - TCP autotuning on both client and server OR manually set buffer
>     sizes: small for TCP-1, and large for TCP-2
>     - First TCP _only_ carries "chatty" traffic, and second TCP carries
>     "mass data" traffic.
>
>     b)
>     - Client 1 <=> Server 1 connected with 1 TCP connection
>     - TCP autotuning OR large buffers set system wide
>     - multiplexing done over the single TCP that carries both "chatty"
>     and "mass data" traffic and
>     - the multiplexing is further optimizing buffer size (app level) and
>     does priorization of traffic (chatty = high-prio, mass = low-prio):
>
>     "Compare": I am mostly interested in the end-user experience:
>
>     i) Is the chatty traffic still low-latency, even in the presence of
>     concurrent mass-data traffic?
>
>
> a) it depends on process scheduler and network queue design of the
> server and client, I think. I don't know much about per-socket buffer
> autotuning, but it probably just improves memory usage and throughput
> but doesn't have much effect on latency of chatty traffic?

Improving memory, but not necessarily latency: yep, plausible.

>
> b) the endpoints need to cut mass data into smaller chunks to reserve
> slots for inserting chatty traffic. e.g. if acceptable delay for chatty
> traffic is x ms, set the maximum send unit of mass traffic to (current
> through put) * x.
>
> To allow one to control the peer's message scheduling, both per-socket
> buffer size and ws-mux flow control are insufficient, I think. Needs to
> exchange ToS info.

I now can see that flow control and priorization are somewhat orthogonal 
issues.

Priorization and ToS is not goal of WS-MUX at all?

>
>     ii) Is the throughput of mass-data traffic concurrently to the
>     chatty traffic still (nearly) as high as when only doing mass data
>     transfer?
>
>
> I think minimum delay requirement for the chatty traffic determines
> fragmentation on mass-data traffic regardless it's done in OS's network
> stack or ws-mux code.

Makes sense. The lower the latency requirement, the more fragmentation, 
and the less efficiency on mass data.

>
>      >>How about having a shared Web worker in the background that
>     handles all messaging for the app over a single WS?
>      >One restriction of Web Worker is that we cannot share a worker
>     between webapps with different origins while WebSocket can connect
>     to a host different from its origin.
>
>     Ah, right. Same origin policy. Thus, with WS-MUX, there is _more_
>     potentially TCP sharing than whats possible with shared WebWorkers.
>
>     However, for a given single app (from a single origin) running in
>     multiple browser tabs, the shared WebWorker seems not so bad. Works
>     today. No need for WS-MUX or HTTP 2.0 MUX + WS over HTTP 2.0.
>
>
> Yes. Thanks for reminding us about shared worker approach. We haven't
> evaluated it well.
>
>      >For load balancer vendors, it's better that multiplexing is
>     standardized than each service provider develops its own.
>
>     I don't have experience running G scale infrastructure. However, I'd
>     like to understand why LBs need to have a look into WS at all .. why
>     not Layer-4 LB? Doing balancing based on hash of source IP?
>
>
> Traffic for various services would be multiplexed into one TCP. We wanna
> demultiplex it and forward to designated backends. This means that
> there're again lots of TCP connections between LB-backends, but we can
> make any optimization there as we want.

I see. So the load-balancer needs to be a layer-7 one, and not only be 
WS-MUX enabled, but also app/service aware? Something like this:

E.g. user has 2 tabs open which (logically) connect to

wss://google.com/gmail
wss://google.com/gplus

Both would be transported over 1 physical WS (via WS-MUX).

The LB would demux traffic for logical channel

wss://google.com/gmail

to backend nodes of cluster "GMAIL", while it would demux traffic for

wss://google.com/gplus

to backend nodes of cluster "GPLUS"?

>
>      >Basically multiplexing should be done as close to network as
>     possible. If it's done in application layer, for example when
>     compression is used, such a load balancer need to decode the
>     compression to see the >multiplexing information. That doesn't make
>     them happy.
>
>     This seems to assume that
>
>     - either the LB is done above layer 4 and/or
>     - the logical WS connections contained in the physical WS would need
>     to be balanced to _different_ backend nodes.
>
>
> Yes. Sorry that we haven't told the story in our mind in detail.

I think it might be helpful to add text to the MUX draft that describe 
the use cases WS-MUX addresses and how it improves thing over 
alternatives. Motivation is always good and helps by providing context 
and examples.

/Tobias

From SRS0+3db7b28a961e272e=PP=noemax.com=arman@noemax.com  Thu May 30 07:38:46 2013
Return-Path: <SRS0+3db7b28a961e272e=PP=noemax.com=arman@noemax.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A2D6421F87D2 for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 07:38:22 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.598
X-Spam-Level: 
X-Spam-Status: No, score=-2.598 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Jft1yescz-2W for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 07:38:10 -0700 (PDT)
Received: from mail.noemax.com (mail.noemax.com [74.208.113.37]) by ietfa.amsl.com (Postfix) with ESMTP id 77B3F21F8E6E for <hybi@ietf.org>; Thu, 30 May 2013 07:38:09 -0700 (PDT)
DKIM-Signature: a=rsa-sha1; t=1369924684; x=1370529484; s=m2048; d=noemax.com; c=relaxed/relaxed; v=1; bh=EK9S2URlM2rDbMVgglzwFdqVN6Y=; h=From:Subject:Date:Message-ID:To:MIME-Version:Content-Type:In-Reply-To:References; b=ngo+K/c5vmQKV5btoj6nIZYlzKVubOWAuo/19U/IN/OmheCc6qfV6T3kyRb2GUMqopscYE2y1xBAztb7e1L1YhSrVUyWmU328qlb7EIPs6ewV2RSe+MqGDHu0J8eD1gd9DPChkKIf56eTf65l/9MnknEEGqupneAv+iT8ZaIYemcCoiGJVfGvOo0NuClivJf1XONlNThXXVtcn2enNeX3bf1rtg132LOhQasMioBXQn2aTUknm8D0Aj3o/mVyOvSJUUWJ01JLRSleaOKLyCpMK4iiig9iWNOHpk7KcasvHvQLDdtgF5dTnQyUHCn4UeGoYBfl5Z2wytdHoQOAayPyQ==
Received: from mail.noemax.com by mail.noemax.com (Noemax Mail Server) with ASMTP (SSL) id 201305301738030291; Thu, 30 May 2013 17:38:03 +0300
From: "Arman Djusupov" <arman@noemax.com>
To: "'Takeshi Yoshino'" <tyoshino@google.com>, <hybi@ietf.org>
References: <CAH9hSJaUnFRve-JMUm1pzdxreHsd5R13gUa0s9_S2u1t6Y87rQ@mail.gmail.com>
In-Reply-To: <CAH9hSJaUnFRve-JMUm1pzdxreHsd5R13gUa0s9_S2u1t6Y87rQ@mail.gmail.com>
Date: Thu, 30 May 2013 17:38:02 +0300
Message-ID: <005001ce5d43$4bcb74e0$e3625ea0$@noemax.com>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="----=_NextPart_000_0051_01CE5D5C.7118ACE0"
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQJIpJn/mIepNG7p5bRkeMd7fgmhTpgpPOwQ
Content-Language: en-us
Subject: Re: [hybi] Simplify WebSocket multiplexing
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 14:41:46 -0000

This is a multipart message in MIME format.

------=_NextPart_000_0051_01CE5D5C.7118ACE0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit

Agree on dropping 1,2, 4.

 

Flow control (6) cannot be removed, mux would not work without it.

 

I would also simplify the initial handshake by removing the quota from the
handshake. Let the server be required to send the first set of channel slots
right after replying to the HTTP handshake, this will not cause any delays
and there will 

be a unified initial quota mechanism for all channels including first
logical channel. So the client sends a handshake requesting the mux
extension, as reply it gets an HTTP response confirming the handshake and
the set of new channel slots in the first frame.

 

Another part of spec to simplify is the DropChannel close codes. There are
too many codes that cover protocol violations. Since we are not designing a
unit-test framework for bad implementations I think most of those are not
needed. Any protocol violation should result in dropping the physical
connection, there is simply no reason to continue communicating with a pier
that does not adhere to the protocol and obviously is not compatible. Polite
behavior in such cases can only be abused by a DoS attacker, where a
malicious client keeps violating the protocol to force the remote endpoint
to send "error reports", wasting its time and I/O threads.

 

With best regards,

Arman

 

 

From: hybi-bounces@ietf.org [mailto:hybi-bounces@ietf.org] On Behalf Of
Takeshi Yoshino
Sent: Thursday, May 30, 2013 10:17 AM
To: hybi@ietf.org
Subject: [hybi] Simplify WebSocket multiplexing

 

(Branching from the thread "Call for interest: multiplexing dedicated for
WebSocket")

 

It seemed that we had rough consensus on accelerating WebSocket dedicated
multiplexing by simplifying it if we keep working on it.

 

I'd like to hear your opinion on this in this separate thread. Below listed
some points we could drop/simplify including both ones suggested in the
thread and some I came up with now.

 

1) Header delta encoding in AddChannelRequest

2) Header delta encoding in AddChannelResponse

3) Channel slot

4) Only one mux control in one message

5) Use the length encoding for encoding mux channel ID

6) Drop flow control

 

----

 

My opinion is to do 1, 2 and 4.


------=_NextPart_000_0051_01CE5D5C.7118ACE0
Content-Type: text/html;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" =
xmlns:o=3D"urn:schemas-microsoft-com:office:office" =
xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" =
xmlns=3D"http://www.w3.org/TR/REC-html40"><head><META =
HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii"><meta name=3DGenerator content=3D"Microsoft Word 14 =
(filtered medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
	{font-family:Tahoma;
	panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0in;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:#0563C1;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:#954F72;
	text-decoration:underline;}
span.EmailStyle17
	{mso-style-type:personal-reply;
	font-family:"Calibri","sans-serif";
	color:#44546A;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri","sans-serif";
	mso-ligatures:standard;}
@page WordSection1
	{size:8.5in 11.0in;
	margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DEN-US =
link=3D"#0563C1" vlink=3D"#954F72"><div class=3DWordSection1><p =
class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>Agree on </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>dropping </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>1,2, 4.<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>Flow control </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>(6) </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>cannot be removed, mux would not work without =
it.<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>I would also simplify</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'> the</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'> initial handshake by removing </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>the </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>quota from the handshake. Let </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>the </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>server be required to send </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>the </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>first set of channel slots right after reply</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>ing</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'> </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>to</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'> the HTTP handshake, this will not cause any delays and there will =
<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>be </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>a </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>unified initial quota mechanism</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'> for all channels including first logical channel</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>. So </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>the </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>client sends </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>a </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>handshake requesting </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>the </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>mux extension</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>,</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'> as reply it gets </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>an </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>HTTP response confirming </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>the </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>handshake and </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>the </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>set of new channel slots </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>in the first frame</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>.<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>Another part of spec to simplify is </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>the </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>DropChannel close codes. There are too many codes that cover protocol =
violations</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>. S</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>ince we are not designing </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>a</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'> unit-test framework for bad implementations I think most of those =
are not needed.</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'> </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>Any protocol violation should result in drop</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>ping the</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'> physical connection, there is simply no reason to continue =
communicating with a pier that does not adhere to the protocol and =
obviously is not compatible. Polite behavior in such cases can only be =
abused by </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>a </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>DoS attacker, where </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>a </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>malicious client keeps violating </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>the </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>protocol to force </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'>the </span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>remote endpoint to send &#8220;error reports&#8221;, wasting its time =
and I/O threads.</span><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497=
D'><o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>With best regards,<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'>Arman<o:p></o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif";color:#44546=
A'><o:p>&nbsp;</o:p></span></p><p class=3DMsoNormal><b><span =
style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span>=
</b><span style=3D'font-size:10.0pt;font-family:"Tahoma","sans-serif"'> =
hybi-bounces@ietf.org [mailto:hybi-bounces@ietf.org] <b>On Behalf Of =
</b>Takeshi Yoshino<br><b>Sent:</b> Thursday, May 30, 2013 10:17 =
AM<br><b>To:</b> hybi@ietf.org<br><b>Subject:</b> [hybi] Simplify =
WebSocket multiplexing<o:p></o:p></span></p><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p><div><p =
class=3DMsoNormal>(Branching from the thread &quot;Call for interest: =
multiplexing dedicated for WebSocket&quot;)<o:p></o:p></p><div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p =
class=3DMsoNormal>It seemed that we had rough consensus on accelerating =
WebSocket dedicated multiplexing by simplifying it if we keep working on =
it.<o:p></o:p></p></div><div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p =
class=3DMsoNormal>I'd like to hear your opinion on this in this separate =
thread. Below listed some points we could drop/simplify including both =
ones suggested in the thread and some I came up with =
now.<o:p></o:p></p></div><div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><div><p =
class=3DMsoNormal>1) Header delta encoding in =
AddChannelRequest<o:p></o:p></p></div><div><p class=3DMsoNormal>2) =
Header delta encoding in AddChannelResponse<o:p></o:p></p></div><div><p =
class=3DMsoNormal>3) Channel slot<o:p></o:p></p></div><div><p =
class=3DMsoNormal>4) Only one mux control in one =
message<o:p></o:p></p></div></div><div><p class=3DMsoNormal>5) Use the =
length encoding for encoding mux channel ID<o:p></o:p></p></div><div><p =
class=3DMsoNormal>6) Drop flow control<o:p></o:p></p></div><div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p =
class=3DMsoNormal>----<o:p></o:p></p></div><div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p =
class=3DMsoNormal>My opinion is to do 1, 2 and =
4.<o:p></o:p></p></div></div></div></body></html>
------=_NextPart_000_0051_01CE5D5C.7118ACE0--


From fenix@google.com  Thu May 30 11:28:54 2013
Return-Path: <fenix@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8405821F8EFC for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 11:28:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.977
X-Spam-Level: 
X-Spam-Status: No, score=-1.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001]
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 10f-a7KgBcFN for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 11:28:53 -0700 (PDT)
Received: from mail-la0-x232.google.com (mail-la0-x232.google.com [IPv6:2a00:1450:4010:c03::232]) by ietfa.amsl.com (Postfix) with ESMTP id DCBE121F8EF2 for <hybi@ietf.org>; Thu, 30 May 2013 11:28:52 -0700 (PDT)
Received: by mail-la0-f50.google.com with SMTP id ed20so574774lab.23 for <hybi@ietf.org>; Thu, 30 May 2013 11:28:51 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=poLSB5a/qI3oL9a4+5s+3+i3HwRCv+m4BNifgbm7PuM=; b=ZSkH05mbEg/hfUFoSNa5YiwNVnM0Vwg5imgETgH1iiM/xDMveQnV5W57KmUrxC1Z69 heew098BPw0t7mh79TKPw94PefyfP3SmkbOYLIgz7qu8kfdYpnUTfxwrUIDhDkiUIgFa oAVLhYeektwbzJ4egCU9FS5DgSokPtMlPBJLLJHHrHX1D3EJUMIpset/qr4YszkpJ7mZ 0TyWjRBuHB6i1E8FU67/9piMIKa+RkY+Og4BFeKdrusTHkxPnWBm975U9aFI1hkbjJNX hyyv1p27sW4DwVwfnBlJqnwHgq79VRvBdhzIJRfXeRCgKEuYmDGDrNYIR/TB3lMmm5Nc 3L+g==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:x-gm-message-state; bh=poLSB5a/qI3oL9a4+5s+3+i3HwRCv+m4BNifgbm7PuM=; b=CEVhqYDKWqlJhTP40sVbWQQ7HYpUwHx5w5WV9IgvrSzSa1JPs8CjhB+ILqzOoiuQ3K 85tro+vUUdkkm4DWCqbS1yxH+RwPLSnVLv8czQBc1uh3uwQy3TaYtX434KiHo0rAN458 2oPfxiiehs6t4zBdz0FvoX55weqFz/B/7dBAIveuKgRwOHolqhCqQXaRWX8QixXIwhKb j2fMGRu//0JpaMvLuJWCFoMhI7UrcI0jYeZ7c3eWr9E6zFEJLbTpDVZSRgBZ8UTV1AEs c6RUX+tiiBkfKR4Wi4MBBab68T9cZU8+VJg2LYOw1B/BpnZCbHbLvoTksEeUYATgRUB1 9tPQ==
MIME-Version: 1.0
X-Received: by 10.112.158.6 with SMTP id wq6mr4334241lbb.108.1369938531652; Thu, 30 May 2013 11:28:51 -0700 (PDT)
Received: by 10.112.76.231 with HTTP; Thu, 30 May 2013 11:28:51 -0700 (PDT)
In-Reply-To: <51A70DF9.5010601@tavendo.de>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net> <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net> <CAHixhFrTk79A07BjQCgvep_+bmA4rGG1ZvqmoS6gsQYNPyPoZA@mail.gmail.com> <CAH9hSJYFa+bqN=e7x87W+Xvq-st70nbzUXniQaPme2fzspCjWA@mail.gmail.com> <51A70DF9.5010601@tavendo.de>
Date: Thu, 30 May 2013 11:28:51 -0700
Message-ID: <CAGzyod6daSOvAVHU6B=Qr0xDqNca_8iXrFYr0gTGdqZPryFFpA@mail.gmail.com>
From: Roberto Peon <fenix@google.com>
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
Content-Type: multipart/alternative; boundary=001a11c268308b27ba04ddf3ab56
X-Gm-Message-State: ALoCoQnYdaufkXvidZE+Hrr5q/PBG5Yh7SLpjG79lIz1CYIctPRHgz7W6gjk1Bvd7+zZPVEKK7ewTAw7P0RxYhmJpNYn+/gigouogXYi56zJHtReYk0sj5SXvRlFBPqEFhzzS8hQa6aegdgYD5lJ1U7RYJpXEZVoHMUghaMagjsJcAUUabejzzIn7eq6obem0wBWMaUuJqDA
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 18:28:54 -0000

--001a11c268308b27ba04ddf3ab56
Content-Type: text/plain; charset=ISO-8859-1

Different schemas have nothing to do with the CRIME attack.
A pure HTTPS or WSS scheme could still be attacked with CRIME, with its
success depending on the way the compression mechanism works.
With the current HTTP/2 proposals it would be pretty difficult to attack
using the techniques in the CRIME attack.

The way CRIME works is to have a 3rd party include links to the
site-to-attack. Thus it doesn't matter if the schemes are http, https, ws,
wss, whatever.
So, a restriction against muxing ws and wss together makes no sense as it
provides no security benefit for CRIME.
-=R


On Thu, May 30, 2013 at 1:29 AM, Tobias Oberstein <
tobias.oberstein@tavendo.de> wrote:

>  Am 30.05.2013 09:28, schrieb Takeshi Yoshino:
>
> On Thu, May 30, 2013 at 1:12 PM, Adam Rice <ricea@google.com> wrote:
>
>>   So d) - f) cannot be multiplexed over the same physical WS as a) - c)?
>>>
>>> Or can an implementation just "silently" transport a)-c) also over wss,
>>> and hence multiplex all of a) - f) over 1 physical WS?
>>>
>>
>>  The handshake does not currently include the schema, so there would be
>> no way to communicate to the server that a)-c) were supposed to be ws:, not
>> wss:.
>>
>
>  Right. I missed that point.
>
>
>>   Even if this was amended, both client and server would have to be
>> careful that no ambient authority leaked from the wss: channels to the ws:
>> channels. For example: the client would have to be careful not to send
>> "secure" cookies with the ws: handshakes, and the server would have to be
>> careful not to apply any authority contained in a client TLS certificate to
>> the ws: logical channels.
>>
>>  For this reason, I think it would be easiest not to attempt to
>> multiplex ws: and wss: onto a single TCP/IP connection.
>>
>
>  Compression contexts, etc. also need to be isolated carefully to protect
> it from attack like CRIME.
>
>
>
> I also agree with Adam. To summarize, given the possible "issues" of
>
> - client cert based auth
> - "secure" cookies / ambient authority
> - compression contexts / CRIME
>
> (plus complexities like: what if an app opens ws://domain.com, and then
> _later_ opens wss://domain.com? Upgrade to TLS not possible anymore ..
> cannot TLS handshake in the middle of an already established non-TLS TCP
> where some traffic already traveled .. the WS handshake at least.)
>
> I think it would be good to include explicit text into the MUX draft, e.g.
>
> """
> An implementation MUST NOT multiplex multiple logical channels to
> WebSocket addresses with different schemes (ws: versus wss:) on a single
> physical WebSocket connection.
>
> For example, two WebSocket connections to ws://domain.com and
> wss://domain.com must not be shared on a single physical WebSocket
> connection to domain.com.
> """
>
> I also think it would be helpful to include explicit text describing what
> WebSocket addresses are eligible for sharing, e.g.
>
> """
> An implementation MAY multiplex multiple logical channels to different
> WebSocket addresses if and only if the WebSocket addresses share the same
> WebSocket scheme, the hostname resolves to the same IP address and the same
> port.
>
> For example, connections to all of the following WebSocket addresses may
> be multiplexed over a single physical WebSocket connection (assum
> domain.com, domain2.com and sub1.domain.com all resolve to the same IP
> address):
>
> ws://domain.com
> ws://domain.com:80
> ws://domain.com/foo
> ws://domain.com:80/foo
> ws://domain.com/bar
> ws://domain2.com
> ws://sub1.domain.com
> ws://sub1.domain.com:80
> ws://sub1.domain.com:80/bar
>
> Note: When a WS address does not specify an explicit port, the port is the
> default port for the given WS scheme (ws: = 80 and wss: = 443).
> """
>
>
>
>
> /Tobias
>
>
> _______________________________________________
> hybi mailing list
> hybi@ietf.org
> https://www.ietf.org/mailman/listinfo/hybi
>
>

--001a11c268308b27ba04ddf3ab56
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Different schemas have nothing to do with the CRIME attack=
.<br><div style>A pure HTTPS or WSS scheme could still be attacked with CRI=
ME, with its success depending on the way the compression mechanism works.<=
/div>
<div style>With the current HTTP/2 proposals it would be pretty difficult t=
o attack using the techniques in the CRIME attack.</div><div style><br></di=
v><div style>The way CRIME works is to have a 3rd party include links to th=
e site-to-attack. Thus it doesn&#39;t matter if the schemes are http, https=
, ws, wss, whatever.</div>
<div style>So, a restriction against muxing ws and wss together makes no se=
nse as it provides no security benefit for CRIME.</div><div style>-=3DR</di=
v></div><div class=3D"gmail_extra"><br><br><div class=3D"gmail_quote">On Th=
u, May 30, 2013 at 1:29 AM, Tobias Oberstein <span dir=3D"ltr">&lt;<a href=
=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_blank">tobias.oberstein@=
tavendo.de</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">
 =20
   =20
 =20
  <div text=3D"#000000" bgcolor=3D"#FFFFFF">
    <div>Am 30.05.2013 09:28, schrieb Takeshi
      Yoshino:<br>
    </div><div><div class=3D"h5">
    <blockquote type=3D"cite">
      <div dir=3D"ltr">On Thu, May 30, 2013 at 1:12 PM, Adam Rice <span dir=
=3D"ltr">&lt;<a href=3D"mailto:ricea@google.com" target=3D"_blank">ricea@go=
ogle.com</a>&gt;</span>
        wrote:<br>
        <div class=3D"gmail_extra">
          <div class=3D"gmail_quote">
            <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;bo=
rder-left:1px #ccc solid;padding-left:1ex">
              <div dir=3D"ltr">
                <div class=3D"gmail_extra">
                  <div class=3D"gmail_quote">
                    <div>
                      <blockquote class=3D"gmail_quote" style=3D"margin:0 0=
 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">So d) - f) cannot be
                        multiplexed over the same physical WS as a) -
                        c)?<br>
                        <br>
                        Or can an implementation just &quot;silently&quot;
                        transport a)-c) also over wss, and hence
                        multiplex all of a) - f) over 1 physical WS?<br>
                      </blockquote>
                      <div><br>
                      </div>
                    </div>
                    <div>The handshake does not currently include the
                      schema, so there would be no way to communicate to
                      the server that a)-c) were supposed to be ws:, not
                      wss:.</div>
                  </div>
                </div>
              </div>
            </blockquote>
            <div><br>
            </div>
            <div>Right. I missed that point.</div>
            <div>=A0<br>
            </div>
            <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;bo=
rder-left:1px #ccc solid;padding-left:1ex">
              <div dir=3D"ltr">
                <div class=3D"gmail_extra">
                  <div class=3D"gmail_quote">
                    <div>Even if this was amended, both client and
                      server would have to be careful that no ambient
                      authority leaked from the wss: channels to the ws:
                      channels. For example: the client would have to be
                      careful not to send &quot;secure&quot; cookies with t=
he ws:
                      handshakes, and the server would have to be
                      careful not to apply any authority contained in a
                      client TLS certificate to the ws: logical
                      channels.<br>
                    </div>
                    <div><br>
                    </div>
                    <div>For this reason, I think it would be easiest
                      not to attempt to multiplex ws: and wss: onto a
                      single TCP/IP connection.</div>
                  </div>
                </div>
              </div>
            </blockquote>
            <div><br>
            </div>
            <div>Compression contexts, etc. also need to be
              isolated carefully to protect it from attack like CRIME.</div=
>
            <div>=A0</div>
          </div>
        </div>
      </div>
    </blockquote>
    <br></div></div>
    I also agree with Adam. To summarize, given the possible &quot;issues&q=
uot; of<br>
    <br>
    - client cert based auth<br>
    - &quot;secure&quot; cookies / ambient authority<br>
    - compression contexts / CRIME<br>
    <br>
    (plus complexities like: what if an app opens <a>ws://domain.com</a>, a=
nd
    then _later_ opens <a>wss://domain.com</a>? Upgrade to TLS not possible
    anymore .. cannot TLS handshake in the middle of an already
    established non-TLS TCP where some traffic already traveled .. the
    WS handshake at least.)<br>
    <br>
    I think it would be good to include explicit text into the MUX
    draft, e.g.<br>
    <br>
    &quot;&quot;&quot;<br>
    An implementation MUST NOT multiplex multiple logical channels to
    WebSocket addresses with different schemes (ws: versus wss:) on a
    single physical WebSocket connection.<br>
    <br>
    For example, two WebSocket connections to <a>ws://domain.com</a> and
    <a>wss://domain.com</a> must not be shared on a single physical WebSock=
et
    connection to <a href=3D"http://domain.com" target=3D"_blank">domain.co=
m</a>.<br>
    &quot;&quot;&quot;<br>
    <br>
    I also think it would be helpful to include explicit text describing
    what WebSocket addresses are eligible for sharing, e.g.<br>
    <br>
    &quot;&quot;&quot;<br>
    An implementation MAY multiplex multiple logical channels to
    different WebSocket addresses if and only if the WebSocket addresses
    share the same WebSocket scheme, the hostname resolves to the same
    IP address and the same port.<br>
    <br>
    For example, connections to all of the following WebSocket addresses
    may be multiplexed over a single physical WebSocket connection
    (assum <a href=3D"http://domain.com" target=3D"_blank">domain.com</a>, =
<a href=3D"http://domain2.com" target=3D"_blank">domain2.com</a> and <a hre=
f=3D"http://sub1.domain.com" target=3D"_blank">sub1.domain.com</a> all reso=
lve to
    the same IP address):<br>
    <br>
    <a>ws://domain.com</a><br>
    <a>ws://domain.com:80</a><br>
    <a>ws://domain.com/foo</a><br>
    <a>ws://domain.com:80/foo</a><br>
    <a>ws://domain.com/bar</a><br>
    <a>ws://domain2.com</a><br>
    <a>ws://sub1.domain.com</a><br>
    <a>ws://sub1.domain.com:80</a><br>
    <a>ws://sub1.domain.com:80/bar</a><br>
    <br>
    Note: When a WS address does not specify an explicit port, the port
    is the default port for the given WS scheme (ws: =3D 80 and wss: =3D
    443).<br>
    &quot;&quot;&quot;<span class=3D"HOEnZb"><font color=3D"#888888"><br>
    <br>
    <br>
    <br>
    <br>
    /Tobias<br>
    <br>
  </font></span></div>

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

--001a11c268308b27ba04ddf3ab56--

From tobias.oberstein@tavendo.de  Thu May 30 11:44:05 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6792221F8F11 for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 11:44:05 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eCg79U-XAXkv for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 11:44:00 -0700 (PDT)
Received: from EXHUB020-4.exch020.serverdata.net (exhub020-4.exch020.serverdata.net [206.225.164.31]) by ietfa.amsl.com (Postfix) with ESMTP id 791CA21F8F9E for <hybi@ietf.org>; Thu, 30 May 2013 11:44:00 -0700 (PDT)
Received: from [78.42.145.191] (78.42.145.191) by smtpx20.serverdata.net (206.225.164.36) with Microsoft SMTP Server (TLS) id 8.3.264.0; Thu, 30 May 2013 11:43:58 -0700
Message-ID: <51A79DF0.5060901@tavendo.de>
Date: Thu, 30 May 2013 20:44:00 +0200
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: Roberto Peon <fenix@google.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net> <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net> <CAHixhFrTk79A07BjQCgvep_+bmA4rGG1ZvqmoS6gsQYNPyPoZA@mail.gmail.com> <CAH9hSJYFa+bqN=e7x87W+Xvq-st70nbzUXniQaPme2fzspCjWA@mail.gmail.com> <51A70DF9.5010601@tavendo.de> <CAGzyod6daSOvAVHU6B=Qr0xDqNca_8iXrFYr0gTGdqZPryFFpA@mail.gmail.com>
In-Reply-To: <CAGzyod6daSOvAVHU6B=Qr0xDqNca_8iXrFYr0gTGdqZPryFFpA@mail.gmail.com>
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 18:44:05 -0000

Am 30.05.2013 20:28, schrieb Roberto Peon:
> Different schemas have nothing to do with the CRIME attack.
> A pure HTTPS or WSS scheme could still be attacked with CRIME, with its
> success depending on the way the compression mechanism works.
> With the current HTTP/2 proposals it would be pretty difficult to attack
> using the techniques in the CRIME attack.
>
> The way CRIME works is to have a 3rd party include links to the
> site-to-attack. Thus it doesn't matter if the schemes are http, https,
> ws, wss, whatever.
> So, a restriction against muxing ws and wss together makes no sense as
> it provides no security benefit for CRIME.

What about the other issues listed?

Example:

An app first connect to a ws: address. Then the app opens a 2nd 
connection over wss: to the same target (IP:port) sending some sensitive 
information (either in cookies or in payload) relying on the presumed 
privacy/encryption provided by wss:.

An implementation of MUX might run the 2nd wss: connection as a logical 
WS connection over an _unencrypted_ physical ws: that was established in 
the first place upon the app opening the 1st ws: connection.

So at least the implementation must not mux logical wss: over physical ws:.

Transporting logical ws: over physical wss: seems less problematic.


> -=R
>
>
> On Thu, May 30, 2013 at 1:29 AM, Tobias Oberstein
> <tobias.oberstein@tavendo.de <mailto:tobias.oberstein@tavendo.de>> wrote:
>
>     Am 30.05.2013 09:28, schrieb Takeshi Yoshino:
>>     On Thu, May 30, 2013 at 1:12 PM, Adam Rice <ricea@google.com
>>     <mailto:ricea@google.com>> wrote:
>>
>>             So d) - f) cannot be multiplexed over the same physical WS
>>             as a) - c)?
>>
>>             Or can an implementation just "silently" transport a)-c)
>>             also over wss, and hence multiplex all of a) - f) over 1
>>             physical WS?
>>
>>
>>         The handshake does not currently include the schema, so there
>>         would be no way to communicate to the server that a)-c) were
>>         supposed to be ws:, not wss:.
>>
>>
>>     Right. I missed that point.
>>
>>         Even if this was amended, both client and server would have to
>>         be careful that no ambient authority leaked from the wss:
>>         channels to the ws: channels. For example: the client would
>>         have to be careful not to send "secure" cookies with the ws:
>>         handshakes, and the server would have to be careful not to
>>         apply any authority contained in a client TLS certificate to
>>         the ws: logical channels.
>>
>>         For this reason, I think it would be easiest not to attempt to
>>         multiplex ws: and wss: onto a single TCP/IP connection.
>>
>>
>>     Compression contexts, etc. also need to be isolated carefully to
>>     protect it from attack like CRIME.
>
>     I also agree with Adam. To summarize, given the possible "issues" of
>
>     - client cert based auth
>     - "secure" cookies / ambient authority
>     - compression contexts / CRIME
>
>     (plus complexities like: what if an app opens ws://domain.com, and
>     then _later_ opens wss://domain.com? Upgrade to TLS not possible
>     anymore .. cannot TLS handshake in the middle of an already
>     established non-TLS TCP where some traffic already traveled .. the
>     WS handshake at least.)
>
>     I think it would be good to include explicit text into the MUX
>     draft, e.g.
>
>     """
>     An implementation MUST NOT multiplex multiple logical channels to
>     WebSocket addresses with different schemes (ws: versus wss:) on a
>     single physical WebSocket connection.
>
>     For example, two WebSocket connections to ws://domain.com and
>     wss://domain.com must not be shared on a single physical WebSocket
>     connection to domain.com <http://domain.com>.
>     """
>
>     I also think it would be helpful to include explicit text describing
>     what WebSocket addresses are eligible for sharing, e.g.
>
>     """
>     An implementation MAY multiplex multiple logical channels to
>     different WebSocket addresses if and only if the WebSocket addresses
>     share the same WebSocket scheme, the hostname resolves to the same
>     IP address and the same port.
>
>     For example, connections to all of the following WebSocket addresses
>     may be multiplexed over a single physical WebSocket connection
>     (assum domain.com <http://domain.com>, domain2.com
>     <http://domain2.com> and sub1.domain.com <http://sub1.domain.com>
>     all resolve to the same IP address):
>
>     ws://domain.com
>     ws://domain.com:80
>     ws://domain.com/foo
>     ws://domain.com:80/foo
>     ws://domain.com/bar
>     ws://domain2.com
>     ws://sub1.domain.com
>     ws://sub1.domain.com:80
>     ws://sub1.domain.com:80/bar
>
>     Note: When a WS address does not specify an explicit port, the port
>     is the default port for the given WS scheme (ws: = 80 and wss: = 443).
>     """
>
>
>
>
>     /Tobias
>
>
>     _______________________________________________
>     hybi mailing list
>     hybi@ietf.org <mailto:hybi@ietf.org>
>     https://www.ietf.org/mailman/listinfo/hybi
>
>


From fenix@google.com  Thu May 30 13:16:38 2013
Return-Path: <fenix@google.com>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C451421F88AC for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 13:16:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.144
X-Spam-Level: 
X-Spam-Status: No, score=-1.144 tagged_above=-999 required=5 tests=[AWL=-0.833, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001, SARE_HTML_USL_OBFU=1.666]
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 EMUWQLI6nDNU for <hybi@ietfa.amsl.com>; Thu, 30 May 2013 13:16:37 -0700 (PDT)
Received: from mail-la0-x233.google.com (mail-la0-x233.google.com [IPv6:2a00:1450:4010:c03::233]) by ietfa.amsl.com (Postfix) with ESMTP id EF83021F8617 for <hybi@ietf.org>; Thu, 30 May 2013 13:16:36 -0700 (PDT)
Received: by mail-la0-f51.google.com with SMTP id lx15so686035lab.10 for <hybi@ietf.org>; Thu, 30 May 2013 13:16:35 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=V50oYYKlit9FyeBwT/K3tVgv16UF3z8G1l3uF6PHmG4=; b=YVvPJdOFplecpkXefDL66vExmLJdH76Hg/tdQe6g7NAqG5rf9y8nF1Qdijfr9C8GrZ rn/XYprCaABXDhWr16hdvrTy56XEMECscWDmdwC7n5+hxFEHo3evJxfakBu1UaKwi0rB fWIkv4JcFw2uh2abnTMScvhAvho8GTMAVAsTFElIWraCWzcfj8V9fy8RAziW3N+i2ER1 3UG84mva9j+M1lhHvHC/Rl/awBQPgDeJPGBVZP9QtwLWxsBhOXO53i0WHRvPdv7p99FK zT5EjycTJzPDKt3UgZcYWkditf2eEd3TY6u7rbDXbf6ylJ+4rCKmBfOiuAllHGqf+H6j dhMQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:x-gm-message-state; bh=V50oYYKlit9FyeBwT/K3tVgv16UF3z8G1l3uF6PHmG4=; b=Z4MNn5Rxjl7DmE0y2nY0svH7CydpKipvHSn4voZ0claWfjd2dNJZrx/BPz9QovwJNx 9Z2BGknDvPOdU21OpDVgESV5FF8JO6P3ygS49QrjzPjSz9cD5Ryfx0YtPCq7rjJMA9RM L1wrlcn8bTQD97rrHBZXkaxRyLH44t0hvIFO34FCVGHp6bTWCTe9OvWospdZJLOEqNLx 7arOx4EUR2Ohr00rAeVNgBeYeXG2JQBjtLlkrtUgetxWqxXQYtoghHCKbZ+/ixCcXT9Z cIHcGYxj/MyT7GtOBuzOB6nswJ5JqNZbmBA1jKfZy8ne8FgU/eum8wBJZlWpzIamCtKA tyhQ==
MIME-Version: 1.0
X-Received: by 10.112.148.104 with SMTP id tr8mr4568498lbb.56.1369944995606; Thu, 30 May 2013 13:16:35 -0700 (PDT)
Received: by 10.112.76.231 with HTTP; Thu, 30 May 2013 13:16:35 -0700 (PDT)
In-Reply-To: <51A79DF0.5060901@tavendo.de>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net> <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net> <CAHixhFrTk79A07BjQCgvep_+bmA4rGG1ZvqmoS6gsQYNPyPoZA@mail.gmail.com> <CAH9hSJYFa+bqN=e7x87W+Xvq-st70nbzUXniQaPme2fzspCjWA@mail.gmail.com> <51A70DF9.5010601@tavendo.de> <CAGzyod6daSOvAVHU6B=Qr0xDqNca_8iXrFYr0gTGdqZPryFFpA@mail.gmail.com> <51A79DF0.5060901@tavendo.de>
Date: Thu, 30 May 2013 13:16:35 -0700
Message-ID: <CAGzyod4zX8fpJZBuOx=9CVFR1y0nNMqenzTVXuEx1ayFOdK0BQ@mail.gmail.com>
From: Roberto Peon <fenix@google.com>
To: Tobias Oberstein <tobias.oberstein@tavendo.de>
Content-Type: multipart/alternative; boundary=047d7b3a8300d345c004ddf52cd3
X-Gm-Message-State: ALoCoQn3TPHcucA9GWByrjMQBq74SG5aQyfgree2PEf5xg1QUriTS/GJH+TqrcFbeg94/q6ECmc4WtV3fQ06aAvVs+NfuQhwKkWSMdEssHPD/AHuFRyQaH9fOOiglYba8Z86VzaLgFnn/qZbxlOO+v8Och7AlIB1BQhEVODRYTVJmoC0GIt/eIWar6CCveu+cml9DtWR/FRo
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 30 May 2013 20:16:38 -0000

--047d7b3a8300d345c004ddf52cd3
Content-Type: text/plain; charset=ISO-8859-1

inline


On Thu, May 30, 2013 at 11:44 AM, Tobias Oberstein <
tobias.oberstein@tavendo.de> wrote:

> Am 30.05.2013 20:28, schrieb Roberto Peon:
>
>  Different schemas have nothing to do with the CRIME attack.
>> A pure HTTPS or WSS scheme could still be attacked with CRIME, with its
>> success depending on the way the compression mechanism works.
>> With the current HTTP/2 proposals it would be pretty difficult to attack
>> using the techniques in the CRIME attack.
>>
>> The way CRIME works is to have a 3rd party include links to the
>> site-to-attack. Thus it doesn't matter if the schemes are http, https,
>> ws, wss, whatever.
>> So, a restriction against muxing ws and wss together makes no sense as
>> it provides no security benefit for CRIME.
>>
>
> What about the other issues listed?
>
> Example:
>
> An app first connect to a ws: address. Then the app opens a 2nd connection
> over wss: to the same target (IP:port) sending some sensitive information
> (either in cookies or in payload) relying on the presumed
> privacy/encryption provided by wss:.
>
> An implementation of MUX might run the 2nd wss: connection as a logical WS
> connection over an _unencrypted_ physical ws: that was established in the
> first place upon the app opening the 1st ws: connection.
>
> So at least the implementation must not mux logical wss: over physical ws:.
>

Agreed. That would invalidate the security guarantee of wss.


>
> Transporting logical ws: over physical wss: seems less problematic.
>

Yup!
-=R


>
>
>  -=R
>>
>>
>> On Thu, May 30, 2013 at 1:29 AM, Tobias Oberstein
>> <tobias.oberstein@tavendo.de <mailto:tobias.oberstein@**tavendo.de<tobias.oberstein@tavendo.de>>>
>> wrote:
>>
>>     Am 30.05.2013 09:28, schrieb Takeshi Yoshino:
>>
>>>     On Thu, May 30, 2013 at 1:12 PM, Adam Rice <ricea@google.com
>>>     <mailto:ricea@google.com>> wrote:
>>>
>>>             So d) - f) cannot be multiplexed over the same physical WS
>>>             as a) - c)?
>>>
>>>             Or can an implementation just "silently" transport a)-c)
>>>             also over wss, and hence multiplex all of a) - f) over 1
>>>             physical WS?
>>>
>>>
>>>         The handshake does not currently include the schema, so there
>>>         would be no way to communicate to the server that a)-c) were
>>>         supposed to be ws:, not wss:.
>>>
>>>
>>>     Right. I missed that point.
>>>
>>>         Even if this was amended, both client and server would have to
>>>         be careful that no ambient authority leaked from the wss:
>>>         channels to the ws: channels. For example: the client would
>>>         have to be careful not to send "secure" cookies with the ws:
>>>         handshakes, and the server would have to be careful not to
>>>         apply any authority contained in a client TLS certificate to
>>>         the ws: logical channels.
>>>
>>>         For this reason, I think it would be easiest not to attempt to
>>>         multiplex ws: and wss: onto a single TCP/IP connection.
>>>
>>>
>>>     Compression contexts, etc. also need to be isolated carefully to
>>>     protect it from attack like CRIME.
>>>
>>
>>     I also agree with Adam. To summarize, given the possible "issues" of
>>
>>     - client cert based auth
>>     - "secure" cookies / ambient authority
>>     - compression contexts / CRIME
>>
>>     (plus complexities like: what if an app opens ws://domain.com, and
>>     then _later_ opens wss://domain.com? Upgrade to TLS not possible
>>     anymore .. cannot TLS handshake in the middle of an already
>>     established non-TLS TCP where some traffic already traveled .. the
>>     WS handshake at least.)
>>
>>     I think it would be good to include explicit text into the MUX
>>     draft, e.g.
>>
>>     """
>>     An implementation MUST NOT multiplex multiple logical channels to
>>     WebSocket addresses with different schemes (ws: versus wss:) on a
>>     single physical WebSocket connection.
>>
>>     For example, two WebSocket connections to ws://domain.com and
>>     wss://domain.com must not be shared on a single physical WebSocket
>>     connection to domain.com <http://domain.com>.
>>
>>     """
>>
>>     I also think it would be helpful to include explicit text describing
>>     what WebSocket addresses are eligible for sharing, e.g.
>>
>>     """
>>     An implementation MAY multiplex multiple logical channels to
>>     different WebSocket addresses if and only if the WebSocket addresses
>>     share the same WebSocket scheme, the hostname resolves to the same
>>     IP address and the same port.
>>
>>     For example, connections to all of the following WebSocket addresses
>>     may be multiplexed over a single physical WebSocket connection
>>     (assum domain.com <http://domain.com>, domain2.com
>>     <http://domain2.com> and sub1.domain.com <http://sub1.domain.com>
>>
>>     all resolve to the same IP address):
>>
>>     ws://domain.com
>>     ws://domain.com:80
>>     ws://domain.com/foo
>>     ws://domain.com:80/foo
>>     ws://domain.com/bar
>>     ws://domain2.com
>>     ws://sub1.domain.com
>>     ws://sub1.domain.com:80
>>     ws://sub1.domain.com:80/bar
>>
>>     Note: When a WS address does not specify an explicit port, the port
>>     is the default port for the given WS scheme (ws: = 80 and wss: = 443).
>>     """
>>
>>
>>
>>
>>     /Tobias
>>
>>
>>     ______________________________**_________________
>>     hybi mailing list
>>     hybi@ietf.org <mailto:hybi@ietf.org>
>>     https://www.ietf.org/mailman/**listinfo/hybi<https://www.ietf.org/mailman/listinfo/hybi>
>>
>>
>>
>

--047d7b3a8300d345c004ddf52cd3
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">inline<div class=3D"gmail_extra"><br><br><div class=3D"gma=
il_quote">On Thu, May 30, 2013 at 11:44 AM, Tobias Oberstein <span dir=3D"l=
tr">&lt;<a href=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_blank">to=
bias.oberstein@tavendo.de</a>&gt;</span> wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">Am 30.05.2013 20:28, schrieb Roberto Peon:<d=
iv class=3D"im"><br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">
Different schemas have nothing to do with the CRIME attack.<br>
A pure HTTPS or WSS scheme could still be attacked with CRIME, with its<br>
success depending on the way the compression mechanism works.<br>
With the current HTTP/2 proposals it would be pretty difficult to attack<br=
>
using the techniques in the CRIME attack.<br>
<br>
The way CRIME works is to have a 3rd party include links to the<br>
site-to-attack. Thus it doesn&#39;t matter if the schemes are http, https,<=
br>
ws, wss, whatever.<br>
So, a restriction against muxing ws and wss together makes no sense as<br>
it provides no security benefit for CRIME.<br>
</blockquote>
<br></div>
What about the other issues listed?<br>
<br>
Example:<br>
<br>
An app first connect to a ws: address. Then the app opens a 2nd connection =
over wss: to the same target (IP:port) sending some sensitive information (=
either in cookies or in payload) relying on the presumed privacy/encryption=
 provided by wss:.<br>

<br>
An implementation of MUX might run the 2nd wss: connection as a logical WS =
connection over an _unencrypted_ physical ws: that was established in the f=
irst place upon the app opening the 1st ws: connection.<br>
<br>
So at least the implementation must not mux logical wss: over physical ws:.=
<br></blockquote><div><br></div><div style>Agreed. That would invalidate th=
e security guarantee of wss.</div><div>=A0</div><blockquote class=3D"gmail_=
quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1=
ex">

<br>
Transporting logical ws: over physical wss: seems less problematic.<br></bl=
ockquote><div><br></div><div style>Yup!</div><div style>-=3DR</div><div>=A0=
</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-l=
eft:1px #ccc solid;padding-left:1ex">

<br>
<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div class=3D"im">
-=3DR<br>
<br>
<br>
On Thu, May 30, 2013 at 1:29 AM, Tobias Oberstein<br></div><div class=3D"im=
">
&lt;<a href=3D"mailto:tobias.oberstein@tavendo.de" target=3D"_blank">tobias=
.oberstein@tavendo.de</a> &lt;mailto:<a href=3D"mailto:tobias.oberstein@tav=
endo.de" target=3D"_blank">tobias.oberstein@<u></u>tavendo.de</a>&gt;&gt; w=
rote:<br>

<br>
=A0 =A0 Am 30.05.2013 09:28, schrieb Takeshi Yoshino:<br>
</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-l=
eft:1px #ccc solid;padding-left:1ex"><div class=3D"im">
=A0 =A0 On Thu, May 30, 2013 at 1:12 PM, Adam Rice &lt;<a href=3D"mailto:ri=
cea@google.com" target=3D"_blank">ricea@google.com</a><br></div><div><div c=
lass=3D"h5">
=A0 =A0 &lt;mailto:<a href=3D"mailto:ricea@google.com" target=3D"_blank">ri=
cea@google.com</a>&gt;&gt; wrote:<br>
<br>
=A0 =A0 =A0 =A0 =A0 =A0 So d) - f) cannot be multiplexed over the same phys=
ical WS<br>
=A0 =A0 =A0 =A0 =A0 =A0 as a) - c)?<br>
<br>
=A0 =A0 =A0 =A0 =A0 =A0 Or can an implementation just &quot;silently&quot; =
transport a)-c)<br>
=A0 =A0 =A0 =A0 =A0 =A0 also over wss, and hence multiplex all of a) - f) o=
ver 1<br>
=A0 =A0 =A0 =A0 =A0 =A0 physical WS?<br>
<br>
<br>
=A0 =A0 =A0 =A0 The handshake does not currently include the schema, so the=
re<br>
=A0 =A0 =A0 =A0 would be no way to communicate to the server that a)-c) wer=
e<br>
=A0 =A0 =A0 =A0 supposed to be ws:, not wss:.<br>
<br>
<br>
=A0 =A0 Right. I missed that point.<br>
<br>
=A0 =A0 =A0 =A0 Even if this was amended, both client and server would have=
 to<br>
=A0 =A0 =A0 =A0 be careful that no ambient authority leaked from the wss:<b=
r>
=A0 =A0 =A0 =A0 channels to the ws: channels. For example: the client would=
<br>
=A0 =A0 =A0 =A0 have to be careful not to send &quot;secure&quot; cookies w=
ith the ws:<br>
=A0 =A0 =A0 =A0 handshakes, and the server would have to be careful not to<=
br>
=A0 =A0 =A0 =A0 apply any authority contained in a client TLS certificate t=
o<br>
=A0 =A0 =A0 =A0 the ws: logical channels.<br>
<br>
=A0 =A0 =A0 =A0 For this reason, I think it would be easiest not to attempt=
 to<br>
=A0 =A0 =A0 =A0 multiplex ws: and wss: onto a single TCP/IP connection.<br>
<br>
<br>
=A0 =A0 Compression contexts, etc. also need to be isolated carefully to<br=
>
=A0 =A0 protect it from attack like CRIME.<br>
</div></div></blockquote><div><div class=3D"h5">
<br>
=A0 =A0 I also agree with Adam. To summarize, given the possible &quot;issu=
es&quot; of<br>
<br>
=A0 =A0 - client cert based auth<br>
=A0 =A0 - &quot;secure&quot; cookies / ambient authority<br>
=A0 =A0 - compression contexts / CRIME<br>
<br>
=A0 =A0 (plus complexities like: what if an app opens ws://<a href=3D"http:=
//domain.com" target=3D"_blank">domain.com</a>, and<br>
=A0 =A0 then _later_ opens wss://<a href=3D"http://domain.com" target=3D"_b=
lank">domain.com</a>? Upgrade to TLS not possible<br>
=A0 =A0 anymore .. cannot TLS handshake in the middle of an already<br>
=A0 =A0 established non-TLS TCP where some traffic already traveled .. the<=
br>
=A0 =A0 WS handshake at least.)<br>
<br>
=A0 =A0 I think it would be good to include explicit text into the MUX<br>
=A0 =A0 draft, e.g.<br>
<br>
=A0 =A0 &quot;&quot;&quot;<br>
=A0 =A0 An implementation MUST NOT multiplex multiple logical channels to<b=
r>
=A0 =A0 WebSocket addresses with different schemes (ws: versus wss:) on a<b=
r>
=A0 =A0 single physical WebSocket connection.<br>
<br>
=A0 =A0 For example, two WebSocket connections to ws://<a href=3D"http://do=
main.com" target=3D"_blank">domain.com</a> and<br>
=A0 =A0 wss://<a href=3D"http://domain.com" target=3D"_blank">domain.com</a=
> must not be shared on a single physical WebSocket<br></div></div>
=A0 =A0 connection to <a href=3D"http://domain.com" target=3D"_blank">domai=
n.com</a> &lt;<a href=3D"http://domain.com" target=3D"_blank">http://domain=
.com</a>&gt;.<div class=3D"im"><br>
=A0 =A0 &quot;&quot;&quot;<br>
<br>
=A0 =A0 I also think it would be helpful to include explicit text describin=
g<br>
=A0 =A0 what WebSocket addresses are eligible for sharing, e.g.<br>
<br>
=A0 =A0 &quot;&quot;&quot;<br>
=A0 =A0 An implementation MAY multiplex multiple logical channels to<br>
=A0 =A0 different WebSocket addresses if and only if the WebSocket addresse=
s<br>
=A0 =A0 share the same WebSocket scheme, the hostname resolves to the same<=
br>
=A0 =A0 IP address and the same port.<br>
<br>
=A0 =A0 For example, connections to all of the following WebSocket addresse=
s<br>
=A0 =A0 may be multiplexed over a single physical WebSocket connection<br><=
/div>
=A0 =A0 (assum <a href=3D"http://domain.com" target=3D"_blank">domain.com</=
a> &lt;<a href=3D"http://domain.com" target=3D"_blank">http://domain.com</a=
>&gt;, <a href=3D"http://domain2.com" target=3D"_blank">domain2.com</a><br>
=A0 =A0 &lt;<a href=3D"http://domain2.com" target=3D"_blank">http://domain2=
.com</a>&gt; and <a href=3D"http://sub1.domain.com" target=3D"_blank">sub1.=
domain.com</a> &lt;<a href=3D"http://sub1.domain.com" target=3D"_blank">htt=
p://sub1.domain.com</a>&gt;<div class=3D"im">
<br>
=A0 =A0 all resolve to the same IP address):<br>
<br>
=A0 =A0 ws://<a href=3D"http://domain.com" target=3D"_blank">domain.com</a>=
<br>
=A0 =A0 ws://<a href=3D"http://domain.com:80" target=3D"_blank">domain.com:=
80</a><br>
=A0 =A0 ws://<a href=3D"http://domain.com/foo" target=3D"_blank">domain.com=
/foo</a><br>
=A0 =A0 ws://<a href=3D"http://domain.com:80/foo" target=3D"_blank">domain.=
com:80/foo</a><br>
=A0 =A0 ws://<a href=3D"http://domain.com/bar" target=3D"_blank">domain.com=
/bar</a><br>
=A0 =A0 ws://<a href=3D"http://domain2.com" target=3D"_blank">domain2.com</=
a><br>
=A0 =A0 ws://<a href=3D"http://sub1.domain.com" target=3D"_blank">sub1.doma=
in.com</a><br>
=A0 =A0 ws://<a href=3D"http://sub1.domain.com:80" target=3D"_blank">sub1.d=
omain.com:80</a><br>
=A0 =A0 ws://<a href=3D"http://sub1.domain.com:80/bar" target=3D"_blank">su=
b1.domain.com:80/bar</a><br>
<br>
=A0 =A0 Note: When a WS address does not specify an explicit port, the port=
<br>
=A0 =A0 is the default port for the given WS scheme (ws: =3D 80 and wss: =
=3D 443).<br>
=A0 =A0 &quot;&quot;&quot;<br>
<br>
<br>
<br>
<br>
=A0 =A0 /Tobias<br>
<br>
<br>
=A0 =A0 ______________________________<u></u>_________________<br>
=A0 =A0 hybi mailing list<br></div>
=A0 =A0 <a href=3D"mailto:hybi@ietf.org" target=3D"_blank">hybi@ietf.org</a=
> &lt;mailto:<a href=3D"mailto:hybi@ietf.org" target=3D"_blank">hybi@ietf.o=
rg</a>&gt;<br>
=A0 =A0 <a href=3D"https://www.ietf.org/mailman/listinfo/hybi" target=3D"_b=
lank">https://www.ietf.org/mailman/<u></u>listinfo/hybi</a><br>
<br>
<br>
</blockquote>
<br>
</blockquote></div><br></div></div>

--047d7b3a8300d345c004ddf52cd3--

From tobias.oberstein@tavendo.de  Fri May 31 01:27:44 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 31FC021F901A for <hybi@ietfa.amsl.com>; Fri, 31 May 2013 01:27:44 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id CQjE6v6Jb7M4 for <hybi@ietfa.amsl.com>; Fri, 31 May 2013 01:27:37 -0700 (PDT)
Received: from EXHUB020-2.exch020.serverdata.net (exhub020-2.exch020.serverdata.net [206.225.164.29]) by ietfa.amsl.com (Postfix) with ESMTP id C764321F91CB for <hybi@ietf.org>; Fri, 31 May 2013 01:27:36 -0700 (PDT)
Received: from [149.172.148.89] (149.172.148.89) by smtpx20.serverdata.net (206.225.164.34) with Microsoft SMTP Server (TLS) id 8.3.264.0; Fri, 31 May 2013 01:27:35 -0700
Message-ID: <51A85EF9.3010907@tavendo.de>
Date: Fri, 31 May 2013 10:27:37 +0200
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130509 Thunderbird/17.0.6
MIME-Version: 1.0
To: Roberto Peon <fenix@google.com>
References: <CAH9hSJZxr+aG7GZa4f-dUOTGj4bnJ+3XxivUX4jei5CMyqN4LQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C319646@EXVMBX020-12.exch020.serverdata.net> <CAH9hSJYrrbSM3TTSKCQ=AMcwCfE4zqNAa1kuAvecrXZTLqy2gQ@mail.gmail.com> <634914A010D0B943A035D226786325D4422C3DA774@EXVMBX020-12.exch020.serverdata.net> <CAHixhFrTk79A07BjQCgvep_+bmA4rGG1ZvqmoS6gsQYNPyPoZA@mail.gmail.com> <CAH9hSJYFa+bqN=e7x87W+Xvq-st70nbzUXniQaPme2fzspCjWA@mail.gmail.com> <51A70DF9.5010601@tavendo.de> <CAGzyod6daSOvAVHU6B=Qr0xDqNca_8iXrFYr0gTGdqZPryFFpA@mail.gmail.com>
In-Reply-To: <CAGzyod6daSOvAVHU6B=Qr0xDqNca_8iXrFYr0gTGdqZPryFFpA@mail.gmail.com>
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "hybi@ietf.org" <hybi@ietf.org>
Subject: Re: [hybi] Call for interest: multiplexing dedicated for WebSocket
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 31 May 2013 08:27:44 -0000

> The way CRIME works is to have a 3rd party include links to the
> site-to-attack. Thus it doesn't matter if the schemes are http, https,
> ws, wss, whatever.

Does above refer to the general requirement for CRIME: the attacker must 
be able to inject data into the source material before it is compressed 
and then encrypted?

If so, how would that be possible with WebSocket and the proposed 
WebSocket per-message compression?


random thoughts:

a)
permessage-deflate only compresses WS data messages, not control frames.

So the attacker would need to inject data into the app-level payload.

The only way to send app-level payload in browsers if via the JavaScript 
WebSocket API (socket.send()).

So the attacker would need to somehow modify this JS code. However, the 
origin of the JS running is (at least with browsers) sent during the WS 
opening handshake as a HTTP header, and a WS server can check and 
decline an incoming connection.

b)
With WS HTTP headers are only sent _once_: during the initial WS opening 
handshake, and by the client to the server.

CRIME requires repeated injection of attacker data.

More so: permessage-deflate does not touch the opening handshake, which 
remains uncompressed.

Side Q: Do browsers reuse an established TLS/TCP connection for multiple 
WS connections?

/Tobias


From tobias.oberstein@tavendo.de  Fri May 31 03:51:16 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A0BBF21F94E1 for <hybi@ietfa.amsl.com>; Fri, 31 May 2013 03:51:16 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.834
X-Spam-Level: 
X-Spam-Status: No, score=-1.834 tagged_above=-999 required=5 tests=[AWL=0.765,  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 TeBPsLaGr0q2 for <hybi@ietfa.amsl.com>; Fri, 31 May 2013 03:51:11 -0700 (PDT)
Received: from EXHUB020-2.exch020.serverdata.net (exhub020-2.exch020.serverdata.net [206.225.164.29]) by ietfa.amsl.com (Postfix) with ESMTP id 54D0421F93DA for <hybi@ietf.org>; Fri, 31 May 2013 03:51:07 -0700 (PDT)
Received: from EXVMBX020-12.exch020.serverdata.net ([169.254.3.90]) by EXHUB020-2.exch020.serverdata.net ([206.225.164.29]) with mapi; Fri, 31 May 2013 03:51:07 -0700
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
To: Takeshi Yoshino <tyoshino@google.com>, "hybi@ietf.org" <hybi@ietf.org>
Date: Fri, 31 May 2013 03:51:09 -0700
Thread-Topic: [hybi] I-D Action: draft-ietf-hybi-permessage-compression-09.txt
Thread-Index: Ac5bc75BydJsDfCVSP+lKrMvspJ0PgCeGEeg
Message-ID: <634914A010D0B943A035D226786325D4422DA06FA0@EXVMBX020-12.exch020.serverdata.net>
References: <20130425140626.10027.9016.idtracker@ietfa.amsl.com> <CAH9hSJYDH47Ya1FNp80xjeD=pwYJAqcx=XDr=SnBbNDD+f-r4Q@mail.gmail.com> <CAH9hSJYa8QA9VkOwS6GEr0+8iJcnQcpMy3X_fKwGQOuJRr=sEw@mail.gmail.com>
In-Reply-To: <CAH9hSJYa8QA9VkOwS6GEr0+8iJcnQcpMy3X_fKwGQOuJRr=sEw@mail.gmail.com>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: de-DE, en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Subject: Re: [hybi] I-D Action: draft-ietf-hybi-permessage-compression-09.txt
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 31 May 2013 10:51:16 -0000

> Please post your opinions and experience (implemented, faced any difficul=
ties, found issues, etc.).

Hi Takeshi,

We are implementing the extension now in Autobahn - testing against latest =
Chrome Canary which provides the extension.

Does Chrome have any (user accessible) switch/options to

1. force automatic message fragmentation
2. set compress window sizes
3. context takeover

?

Rgd 1.: I am asking since Autobahn provides APIs for message, frame and str=
eaming processing and we want to make sure permessage-deflate works for tho=
se also.

Thanks,
Tobias

From tobias.oberstein@tavendo.de  Fri May 31 13:54:51 2013
Return-Path: <tobias.oberstein@tavendo.de>
X-Original-To: hybi@ietfa.amsl.com
Delivered-To: hybi@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6F40721F8F41 for <hybi@ietfa.amsl.com>; Fri, 31 May 2013 13:54:49 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9QX-FBT9aC8a for <hybi@ietfa.amsl.com>; Fri, 31 May 2013 13:54:42 -0700 (PDT)
Received: from EXHUB020-2.exch020.serverdata.net (exhub020-2.exch020.serverdata.net [206.225.164.29]) by ietfa.amsl.com (Postfix) with ESMTP id EA29E21F8E89 for <hybi@ietf.org>; Fri, 31 May 2013 13:54:39 -0700 (PDT)
Received: from EXVMBX020-12.exch020.serverdata.net ([169.254.3.90]) by EXHUB020-2.exch020.serverdata.net ([206.225.164.29]) with mapi; Fri, 31 May 2013 13:54:39 -0700
From: Tobias Oberstein <tobias.oberstein@tavendo.de>
To: Takeshi Yoshino <tyoshino@google.com>, "hybi@ietf.org" <hybi@ietf.org>
Date: Fri, 31 May 2013 13:54:41 -0700
Thread-Topic: [hybi] I-D Action: draft-ietf-hybi-permessage-compression-09.txt
Thread-Index: Ac5bc75BydJsDfCVSP+lKrMvspJ0PgCv6OVw
Message-ID: <634914A010D0B943A035D226786325D4422DC20D62@EXVMBX020-12.exch020.serverdata.net>
References: <20130425140626.10027.9016.idtracker@ietfa.amsl.com> <CAH9hSJYDH47Ya1FNp80xjeD=pwYJAqcx=XDr=SnBbNDD+f-r4Q@mail.gmail.com> <CAH9hSJYa8QA9VkOwS6GEr0+8iJcnQcpMy3X_fKwGQOuJRr=sEw@mail.gmail.com>
In-Reply-To: <CAH9hSJYa8QA9VkOwS6GEr0+8iJcnQcpMy3X_fKwGQOuJRr=sEw@mail.gmail.com>
Accept-Language: de-DE, en-US
Content-Language: de-DE
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: de-DE, en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Subject: Re: [hybi] I-D Action: draft-ietf-hybi-permessage-compression-09.txt
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/hybi>, <mailto:hybi-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/hybi>
List-Post: <mailto:hybi@ietf.org>
List-Help: <mailto:hybi-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/hybi>, <mailto:hybi-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 31 May 2013 20:54:51 -0000

> Please post your opinions and experience (implemented, faced any difficul=
ties, found issues, etc.).

1) Some first testing ..

http://code.google.com/p/chromium/issues/detail?id=3D245732
http://code.google.com/p/chromium/issues/detail?id=3D245719

2)
If client offers multiple PMCE choices to the server by including multiple =
elements in the "Sec-WebSocket-Extensions" header and the server supports m=
ore than one of the PMCEs offered, is there any preference which PMCE the s=
erver SHOULD choose? E.g. the first PMCE in the header that is supported by=
 the server?

The description of the examples in "5.1. Negotiation Examples" seem to sugg=
est that the _ordering_ of alternative PMCE offered designates _preference_=
.

3)
"""
   A server MUST decline a "permessage-deflate" offer if any of the
   following conditions is met:

   o  The offer has any extension parameter not defined for use in an
      offer.

   o  The offer has any extension parameter with an invalid value.

   o  The offer has multiple extension parameters with the same name.

   o  The server doesn't support the offered configuration.
"""

So if multiple permessage-deflate PMCEs are offered by the client, the serv=
er silently skips any invalid until the first valid or none if there is no =
single valid one?

The server does NOT fail the WS connection upon encountering the first inva=
lid PMCE in the list of offers?

4)
"""
  A client MUST _Fail the WebSocket Connection_ if the server accepted
   a "permessage-deflate" offer with a response meeting any of the
   following condition:

   o  The response has any extension parameter not defined for use in a
      response.

   o  The response has any extension parameter with an invalid value.

   o  The response has multiple extension parameters with the same name.

   o  The client doesn't support the configuration the response
      represents.
"""

This list does not include: "The response corresponds to a PMCE configurati=
on not offered by the client originally".
Is this on purpose?
(A client may support a configuration which it did not offer .. for whateve=
r reasons)

