From: Damian Lukowski Subject: Re: scp stalls mysteriously Date: Wed, 02 Dec 2009 20:17:44 +0100 Message-ID: <4B16BD58.3040802@tvk.rwth-aachen.de> References: <20091128010156.1219012a@houba> <20091130195004.38dbc319@houba> <20091130213727.2f4047d2@houba> <20091201211945.505d3c98@houba> <20091202085925.472136e2@houba> <20091202154403.GB30730@sd-11162.dedibox.fr> <20091202183451.173db5f2@houba> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: =?UTF-8?B?SWxwbyBKw6RydmluZW4=?= , Netdev , Asdo , David Miller , Eric Dumazet , Herbert Xu To: Frederic Leroy Return-path: Received: from mta-2.ms.rz.RWTH-Aachen.DE ([134.130.7.73]:42368 "EHLO mta-2.ms.rz.rwth-aachen.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755345AbZLBTRj (ORCPT ); Wed, 2 Dec 2009 14:17:39 -0500 Received: from ironport-out-1.rz.rwth-aachen.de ([134.130.5.40]) by mta-2.ms.rz.RWTH-Aachen.de (Sun Java(tm) System Messaging Server 6.3-7.04 (built Sep 26 2008)) with ESMTP id <0KU1003A6HLLU5C0@mta-2.ms.rz.RWTH-Aachen.de> for netdev@vger.kernel.org; Wed, 02 Dec 2009 20:17:45 +0100 (CET) In-reply-to: <20091202183451.173db5f2@houba> Sender: netdev-owner@vger.kernel.org List-ID: =46rederic Leroy schrieb: > Le Wed, 2 Dec 2009 18:05:24 +0200 (EET), > "Ilpo J=C3=A4rvinen" a =C3=A9crit : >=20 >>>> In one of the cases, also the sg end dies (the 4th case). I >>>> suppose that was running earlier kernel already? >>> This case was the one with tcp_frto=3D2 and tcp_timestamps=3D0 on h= ouba. >> I suppose we're confused, I was refering to .4. case, did you perhap= s >> mix that up with the latest set of tests which yields .8.? >=20 > You're right, I was talking about .8 >=20 >> In the recent work, the most suspicious things are the new timeout >> things, I'll read them through once I've some time (but so far I've >> not found anything wrong in them but I of course can miss something >> subtle). ...I've added Damian as CC if he has some idea. If you want >> you can try with a trivial revert of that stuff, I've included a >> patch for that below. >=20 > I will try it.=20 Hi, could you please printk retrans_stamp just before the return in=20 include/net/tcp.h:retransmits_timed_out()? If the value is not monotonically increasing but is reset to 0 at some point, this might lead to problems in tcp_write_timeout(). It's the only idea I have now. >=20 > I just discover something not good. I use tuxonice[1] branch on houba > for 2.6.32*. > Although my 2.6.31 kernel is vanilla[2]. >=20 > [1] git://git.kernel.org/pub/scm/linux/kernel/git/nigelc/tuxonice-hea= d.git > [2] git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.= git >=20 > Now, I'm gonna go compile kernels again and again and make more > tests ... :) >=20