From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnd Hannemann Subject: Re: [PATCH] TCP_FAILFAST: a new socket option to timeout/abort a connection quicker Date: Tue, 24 Aug 2010 10:04:37 +0200 Message-ID: <4C737D15.5060400@nets.rwth-aachen.de> References: <1282630819-23104-1-git-send-email-hkchu@google.com> <1282632262.2378.1681.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: "H.K. Jerry Chu" , ilpo.jarvinen@helsinki.fi, davem@davemloft.net, netdev@vger.kernel.org To: Eric Dumazet Return-path: Received: from mail-i4.nets.RWTH-Aachen.DE ([137.226.12.21]:53704 "EHLO MAIL-i4.nets.rwth-aachen.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752227Ab0HXINW (ORCPT ); Tue, 24 Aug 2010 04:13:22 -0400 In-Reply-To: <1282632262.2378.1681.camel@edumazet-laptop> Sender: netdev-owner@vger.kernel.org List-ID: Am 24.08.2010 08:44, schrieb Eric Dumazet: > Le lundi 23 ao=C3=BBt 2010 =C3=A0 23:20 -0700, H.K. Jerry Chu a =C3=A9= crit : >> From: Jerry Chu >> >> This is a TCP level socket option that takes an unsigned int to spec= ify >> how long in ms TCP should resend a lost data packet before giving up >> and returning ETIMEDOUT. The normal TCP retry/abort timeout limit st= ill >> applies. In other words this option is only meant for those applicat= ions >> that need to "fail faster" than the default TCP timeout. The latter >> may take upto 20 minutes in a normal WAN environment. >> >> The option is disabled (by default) when set to 0. Also it does not >> apply during the connection establishment phase. >> >> Signed-off-by: H.K. Jerry Chu >=20 > TCP_FAILFAST might be misleading. It reads as a boolean option, while > its an option to cap the timeout, with a time unit, instead of the us= ual > "number of retransmits". Why not call it TCP_USERTIMEOUT? Later you can also send it via the TCP user timeout option... (RFC5482) Hmm... is the ms granularity really needed? Does it make sense to abort a connection below a second? > Its also funny you dont ask for a default value, given by a sysctl > tunable ;) Well retries1/2 would be the tunables, no? Best regards, Arnd