Re: [hybi] Fwd: New Version Notification for draft-thomson-hybi-http-timeout-00

Mark Nottingham <mnot@mnot.net> Thu, 10 March 2011 17:38 UTC

Return-Path: <mnot@mnot.net>
X-Original-To: hybi@core3.amsl.com
Delivered-To: hybi@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 2810A3A6810 for <hybi@core3.amsl.com>; Thu, 10 Mar 2011 09:38:17 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -105.599
X-Spam-Level:
X-Spam-Status: No, score=-105.599 tagged_above=-999 required=5 tests=[AWL=-3.000, BAYES_00=-2.599, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id W4sWrDX0Kbab for <hybi@core3.amsl.com>; Thu, 10 Mar 2011 09:38:15 -0800 (PST)
Received: from mxout-07.mxes.net (mxout-07.mxes.net [216.86.168.182]) by core3.amsl.com (Postfix) with ESMTP id 917373A65A6 for <hybi@ietf.org>; Thu, 10 Mar 2011 09:38:15 -0800 (PST)
Received: from [10.10.1.131] (unknown [67.111.52.130]) (using TLSv1 with cipher AES128-SHA (128/128 bits)) (No client certificate requested) by smtp.mxes.net (Postfix) with ESMTPSA id 84FFC22E257; Thu, 10 Mar 2011 12:39:26 -0500 (EST)
Mime-Version: 1.0 (Apple Message framework v1082)
Content-Type: text/plain; charset="us-ascii"
From: Mark Nottingham <mnot@mnot.net>
In-Reply-To: <4D776669.20300@ericsson.com>
Date: Thu, 10 Mar 2011 09:39:25 -0800
Content-Transfer-Encoding: quoted-printable
Message-Id: <98F23E74-AF00-4F38-968A-3B4CD40B8B3B@mnot.net>
References: <4D776669.20300@ericsson.com>
To: Salvatore Loreto <salvatore.loreto@ericsson.com>, Greg Wilkins <gregw@webtide.com>, Martin Thomson <Martin.Thomson@andrew.com>
X-Mailer: Apple Mail (2.1082)
Cc: "hybi@ietf.org HTTP" <hybi@ietf.org>
Subject: Re: [hybi] Fwd: New Version Notification for draft-thomson-hybi-http-timeout-00
X-BeenThere: hybi@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Server-Initiated HTTP <hybi.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/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, 10 Mar 2011 17:38:17 -0000

Hey,

A couple of thoughts --

I regularly am asked by devs for something like Request-Timeout, usually for enforcement on an intermediary, rather than the origin, but it's applicable there too.

I always resist it, because there's a race condition between the client's declared request time and its enforcement (as you know it). It's better to just close the connection and enforce the timeout where it's declared; no matter how clearly you document that this is just advisory, many developers will look at request-timeout and immediately use it in place of a client-side timeout. 

Do you have use cases for Request-Timeout that aren't covered by a client-side timeout?

Regarding Connection-Timeout - I still think a simple indication that it's a long-lived request is more useful (and truthful), but if you are going to go down this road, you'll need to define 'idle' more carefully. 

For example, Squid had the concept of a read timeout <http://www.squid-cache.org/Versions/v2/2.HEAD/cfgman/read_timeout.html>, which applies to connections with an outstanding request on them, but it also has a persistent request timeout <http://www.squid-cache.org/Versions/v2/2.HEAD/cfgman/persistent_request_timeout.html> which applies to those that don't. How do they interact with Connection-TImeout?

Cheers,	


On 09/03/2011, at 3:37 AM, Salvatore Loreto wrote:

> (as individual)
> 
> 
> Martin, Greg and I have submitted the following draft:
> http://tools.ietf.org/id/draft-thomson-hybi-http-timeout-00.txt
> 
> The draft only defines HTTP headers that can be (re)used for discovery idle time to use in
> both LongPolling/Comet and WebSocket connection lifetime management (keep alive process).
> 
> Note, the draft does not discuss the actual WebSocket keep alive process;
> and this thread is only to discuss the mechanism for discovery idle time.
> 
> 
> Comments and opinions on the draft are very welcome and appreciate.
> 
> cheers
> /Sal
> 
> -- 
> Salvatore Loreto
> 
> www.sloreto.com
> 
> 
> 
> -------- Original Message --------
> Subject:	New Version Notification for draft-thomson-hybi-http-timeout-00
> Date:	Mon, 7 Mar 2011 00:09:52 +0100
> From:	IETF I-D Submission Tool <idsubmission@ietf.org>
> To:	martin.thomson@andrew.com <martin.thomson@andrew.com>
> CC:	Salvatore Loreto <salvatore.loreto@ericsson.com>, "gregw@intalio.com" <gregw@intalio.com>
> 
> A new version of I-D, draft-thomson-hybi-http-timeout-00.txt has been successfully submitted by Martin Thomson and posted to the IETF repository.
> 
> Filename:	 draft-thomson-hybi-http-timeout
> Revision:	 00
> Title:		 Hypertext Transfer Protocol (HTTP) Timeouts
> Creation_date:	 2011-03-07
> WG ID:		 Independent Submission
> Number_of_pages: 12
> 
> Abstract:
> A Request-Timeout header is defined for Hypertext Transfer Protocol
> (HTTP).  This end-to-end header informs an origin server and any
> intermediaries of the maximum time that a client will await a
> response to its request.  A server can use this header to ensure that
> a timely response is generated.  This also identifies requests as
> being potentially long-lived, and allows for better resource
> allocation for these requests.
> 
> A Connection-Timeout header is defined for HTTP.  This hop-by-hop
> header informs the entity at the other end of a connection of the
> maximum time that an idle connection is kept open.  This header
> improves reliability by providing better information about the idle
> connection management policy of HTTP hosts.
>                                                                                   
> 
> 
> The IETF Secretariat.
> 
> 
> 
> _______________________________________________
> hybi mailing list
> hybi@ietf.org
> https://www.ietf.org/mailman/listinfo/hybi

--
Mark Nottingham   http://www.mnot.net/