From owner-freebsd-net@FreeBSD.ORG Tue Aug 2 13:36:35 2011 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 79E091065675 for ; Tue, 2 Aug 2011 13:36:35 +0000 (UTC) (envelope-from prvs=119502d6cc=killing@multiplay.co.uk) Received: from mail1.multiplay.co.uk (mail1.multiplay.co.uk [85.236.96.23]) by mx1.freebsd.org (Postfix) with ESMTP id 004E68FC14 for ; Tue, 2 Aug 2011 13:36:34 +0000 (UTC) X-MDAV-Processed: mail1.multiplay.co.uk, Tue, 02 Aug 2011 14:36:02 +0100 X-Spam-Processed: mail1.multiplay.co.uk, Tue, 02 Aug 2011 14:36:02 +0100 X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on mail1.multiplay.co.uk X-Spam-Level: X-Spam-Status: No, score=-5.0 required=6.0 tests=USER_IN_WHITELIST shortcircuit=ham autolearn=disabled version=3.2.5 Received: from r2d2 ([188.220.16.49]) by mail1.multiplay.co.uk (mail1.multiplay.co.uk [85.236.96.23]) (MDaemon PRO v10.0.4) with ESMTP id md50014434374.msg for ; Tue, 02 Aug 2011 14:36:02 +0100 X-MDRemoteIP: 188.220.16.49 X-Return-Path: prvs=119502d6cc=killing@multiplay.co.uk X-Envelope-From: killing@multiplay.co.uk X-MDaemon-Deliver-To: freebsd-net@freebsd.org Message-ID: <8F1656993CD14870B678FCF7CF71C113@multiplay.co.uk> From: "Steven Hartland" To: "Jack Vogel" References: <379885BA631F4C7787C24E00A174B429@multiplay.co.uk> <00F86D895BEA493E8FB82F0CDF42AA04@multiplay.co.uk> Date: Tue, 2 Aug 2011 14:36:35 +0100 MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=response Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.5931 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.6109 Cc: freebsd-net@freebsd.org, Kevin Oberman Subject: Re: igb enable_aim or flow_control causing tcp stalls? X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 02 Aug 2011 13:36:35 -0000 Wanted to update this thread to confirm, as Jack suspected, the issue we where seeing was something much more ingrained that a driver issue. It turned out this problem in the tcp reassembly code which is likely due to insufficient size of net.inet.tcp.reass.maxsegments. The result is even with just 2 decent sized tcp streams a single lost packet can cause an unrecoverable situation resulting in the session stalling and eventually ~60 seconds later being reset by the sender. Increasing nmbclusters which in turn increases maxsegments eliminated the problem e.g. sysctl kern.ipc.nmbclusters=262144 kern.ipc.nmbclusters: 25600 -> 262144 For more information see the thread:- tcp failing to recover from a packet loss under 8.2-RELEASE? http://lists.freebsd.org/pipermail/freebsd-net/2011-August/029491.html Regards Steve ================================================ This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. In the event of misdirection, illegible or incomplete transmission please telephone +44 845 868 1337 or return the E.mail to postmaster@multiplay.co.uk.