This message generated a parse failure. Raw output follows here. Please use 'back' to navigate. From devnull@lkml.org Thu Apr 25 18:20:25 2024 Received: from entropy.muc.muohio.edu (entropy.muc.muohio.edu [134.53.213.10]) by herbie.ucs.indiana.edu (8.8.8/8.8.8) with ESMTP id TAA19722 for ; Sun, 16 Aug 1998 19:31:33 -0500 (EST) Received: from vger.rutgers.edu (vger.rutgers.edu [128.6.190.2]) by entropy.muc.muohio.edu (8.8.7/8.8.7) with ESMTP id UAA00569; Sun, 16 Aug 1998 20:24:03 -0400 Received: by vger.rutgers.edu id <154528-22291>; Sun, 16 Aug 1998 19:08:49 -0400 Received: from zero.aec.at ([195.3.98.22]:9811 "HELO zero.aec.at" ident: "qmailr") by vger.rutgers.edu with SMTP id <154583-22291>; Sun, 16 Aug 1998 18:42:17 -0400 Received: (qmail 11203 invoked by uid 573); 16 Aug 1998 22:40:04 -0000 To: Markus.Schoder@t-online.de (Markus Schoder) Cc: linux-kernel@vger.rutgers.edu Subject: Re: TCP stalls in 2.1.1xx kernels References: <35D75A35.FA20C96A@t-online.de> From: Andi Kleen Date: 17 Aug 1998 00:40:04 +0200 In-Reply-To: Markus.Schoder@t-online.de's message of "Mon, 17 Aug 1998 00:16:21 +0200" Message-Id: Lines: 26 X-Mailer: Gnus v5.6.30/Emacs 20.2 X-Orcpt: rfc822;linux-kernel@vger.rutgers.edu Sender: owner-linux-kernel@vger.rutgers.edu Precedence: bulk X-Loop: majordomo@vger.rutgers.edu In article <35D75A35.FA20C96A@t-online.de>, Markus.Schoder@t-online.de (Markus Schoder) writes: > Some people see TCP stalls with recent kernels. Me too. > After I disabled TCP time stamps (by doing an > echo 0>/proc/sys/net/ipv4/tcp_timestamps) everything worked > fine. > With TCP time stamps enabled I got lots of IpInHdrErrors (checked with > cat /proc/net/snmp) and tcpdump logged an equal number of incoming > packets with "truncated-ip - 9 bytes missing". Those packets where > then ignored by the kernel (no ack send) and caused long stalls while > waiting for retransmission. > I can't really see how this all fits together because if there is an > IpInHdrError the packet is immediately discarded and the TCP layer > shouldn't even see it. Right? So why can a TCP option make a > difference? This sounds like some terminal server whose VJ header compression function does not like TCP options. Hopefully these things will be history soon because Windows98 sends timestamps. Try starting your pppd with novj, or simply turn timestamps off like you're already discovered. There is not much that Linux can do against such crap software. -Andi - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html