From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: Network hangs with 2.6.30.5 Date: Thu, 19 Nov 2009 15:40:38 -0800 (PST) Message-ID: <20091119.154038.194307613.davem@davemloft.net> References: <20090907072143.GA5966@ff.dom.local> <20091001.154913.88345178.davem@davemloft.net> Mime-Version: 1.0 Content-Type: Text/Plain; charset=iso-8859-15 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: jarkao2@gmail.com, holger.hoffstaette@googlemail.com, netdev@vger.kernel.org, eric.dumazet@gmail.com, zbr@ioremap.net To: ilpo.jarvinen@helsinki.fi Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:59005 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758155AbZKSXkW convert rfc822-to-8bit (ORCPT ); Thu, 19 Nov 2009 18:40:22 -0500 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: =46rom: "Ilpo J=E4rvinen" Date: Fri, 2 Oct 2009 11:11:55 +0300 (EEST) > On Thu, 1 Oct 2009, David Miller wrote: >=20 >> From: Jarek Poplawski >> Date: Mon, 7 Sep 2009 07:21:43 +0000 >>=20 >> I also looked through all the TCP commits in 2.6.29 to 2.6.30 >> and I could not find anything else that might cause stalls with >> time-wait recycled connections. >=20 > What about the more than 64k connections change a9d8f9110d7e953c2f2 (= or=20 > its fixes), it might be another possibility? ...It certainly does=20 > something related to reuse and happens to be in the correct time fram= e...=20 > (I've added Evgeniy). So I've been studying this one quite a bit. This change could only cause problems in timewait recycling if it: 1) Would accept using a bind bucket that previously it would not 2) Caused socket corruption due to bad locking And I can find neither problem with the Evgeniy's bind changes. So we've back to square one I think.