From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [Bugme-new] [Bug 15507] New: kernel misses 3rd part of tcp handshake (ACK), stays in SYN_RECV state Date: Fri, 19 Mar 2010 00:12:31 +0100 Message-ID: <1268953951.2894.194.camel@edumazet-laptop> References: <20100318160103.cc28c367.akpm@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org, bugzilla-daemon@bugzilla.kernel.org, bugme-daemon@bugzilla.kernel.org, roysjosh@gmail.com To: Andrew Morton Return-path: Received: from mail-bw0-f209.google.com ([209.85.218.209]:50404 "EHLO mail-bw0-f209.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751171Ab0CRXMg (ORCPT ); Thu, 18 Mar 2010 19:12:36 -0400 Received: by bwz1 with SMTP id 1so2549978bwz.21 for ; Thu, 18 Mar 2010 16:12:34 -0700 (PDT) In-Reply-To: <20100318160103.cc28c367.akpm@linux-foundation.org> Sender: netdev-owner@vger.kernel.org List-ID: Le jeudi 18 mars 2010 =C3=A0 16:01 -0700, Andrew Morton a =C3=A9crit : > (switched to email. Please respond via emailed reply-to-all, not via= the > bugzilla web interface). >=20 > On Wed, 10 Mar 2010 14:50:51 GMT > bugzilla-daemon@bugzilla.kernel.org wrote: >=20 > > http://bugzilla.kernel.org/show_bug.cgi?id=3D15507 > >=20 > > Summary: kernel misses 3rd part of tcp handshake (ACK), = stays > > in SYN_RECV state > > Product: Networking > > Version: 2.5 > > Kernel Version: 2.6.32.9-67.fc12 > > Platform: All > > OS/Version: Linux > > Tree: Fedora > > Status: NEW > > Severity: normal > > Priority: P1 > > Component: IPV4 > > AssignedTo: shemminger@linux-foundation.org > > ReportedBy: roysjosh@gmail.com > > Regression: Yes > >=20 > >=20 > > Created an attachment (id=3D25450) > > --> (http://bugzilla.kernel.org/attachment.cgi?id=3D25450) > > tcpdump from virtual guest over ::1 showing dup synacks > >=20 > > Setup: > > 64-bit dual-core, latest F12. (I've experienced this on .31 and .3= 2 kernels.) > >=20 > > I have also experienced this on 2 other machines, all 64-bit. One = was a > > single-core virtual guest. > >=20 > > 1. Install httpd. > > 2. Start it. Don't bother configuring anything unless you need to. > > 3. tcpdump ..... > > 4. connect from localhost or (possibly) a nearby host. I'm not sur= e if this is > > timing related yet. > > 5. note duplicate SYN-ACKs sent out at the exponential backoff. So= me setups > > exhibit this more/worse than others, and will actually time out the= connection. > > (Others will only send one or two duplicate SYN-ACKs, and then ent= er the > > ESTABLISHED state.) However, if any data is sent the kernel enters > > ESTABLISHED. > >=20 > > Note that this isn't ipv4 specific. I have seen it under ipv4 and = ipv6 (via > > loopback). > >=20 >=20 > -- I would say this is expected if httpd server set DEFER_ACCEPT socket option.