From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: TCP keepalive timer problem Date: Thu, 27 Aug 2009 16:17:10 +0200 Message-ID: <4A969566.3070606@gmail.com> References: <0939B589FC103041945B9F13274963E303B1A9D4@CORPUSMX90A.corp.emc.com> <4A93E36C.8070502@gmail.com> <0939B589FC103041945B9F13274963E303B1AD89@CORPUSMX90A.corp.emc.com> <4A967FCE.3000807@gmail.com> <87skfdl6qt.fsf@basil.nowhere.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Li_Xin2@emc.com, linux-kernel@vger.kernel.org, netdev@vger.kernel.org To: Andi Kleen Return-path: Received: from gw1.cosmosbay.com ([212.99.114.194]:40380 "EHLO gw1.cosmosbay.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751248AbZH0ORR (ORCPT ); Thu, 27 Aug 2009 10:17:17 -0400 In-Reply-To: <87skfdl6qt.fsf@basil.nowhere.org> Sender: netdev-owner@vger.kernel.org List-ID: Andi Kleen a =E9crit : > Eric Dumazet writes: >> Now, 7200 seconds might be inappropriate for special needs, and cons= idering >> there is no way to change tcp_retries2 for a given socket (only choi= ce being the global >> tcp_retries2 setting), I would vote for a change in our stack, to *r= elax* RFC, >> and get smaller keepalive timers if possible. >=20 > I think the better fix would be to just to only do that when > tcp_retries2 > keep alive time. So keep the existing behaviour > with default keep alive, but switch when the user defined > a very short keep alive. >=20 tcp_retries2 is a number of retries, its difficult to derive a time fro= m it. Also, it's not clear what behavior you are refering to. Imagine we can be smart and compute tcp_retries2_time (in jiffies) from= tcp_retries2 If keepalive_timer fires and we have packets in flight, what heuristic = do you suggest ? if (tp->packets_out || tcp_send_head(sk)) if (tcp_retries2_time < keepalive_time_when(tp)) goto resched; elapsed =3D tcp_time_stamp - tp->rcv_tstamp; =2E.. What would be the gain ? Arming timer exactly every keepalive_time_when(tp) instead of keepalive_time_when(tp) - (tcp_time_stamp - tp->rcv_tstamp) = ?