From owner-freebsd-net@FreeBSD.ORG Mon May 10 20:49:25 2010 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id AF5051065670 for ; Mon, 10 May 2010 20:49:25 +0000 (UTC) (envelope-from ml@netfence.it) Received: from cp-out8.libero.it (cp-out8.libero.it [212.52.84.108]) by mx1.freebsd.org (Postfix) with ESMTP id 17CA68FC0A for ; Mon, 10 May 2010 20:49:24 +0000 (UTC) Received: from soth.ventu (151.51.8.163) by cp-out8.libero.it (8.5.107) id 4BDF0CE30136902E for freebsd-net@freebsd.org; Mon, 10 May 2010 22:49:23 +0200 Received: from alamar.ventu (alamar.ventu [10.1.2.18]) by soth.ventu (8.14.4/8.14.3) with ESMTP id o4AKnHHv005329 for ; Mon, 10 May 2010 22:49:17 +0200 (CEST) (envelope-from ml@netfence.it) Message-ID: <4BE8714D.8080608@netfence.it> Date: Mon, 10 May 2010 22:49:17 +0200 From: Andrea Venturoli User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; it-IT; rv:1.9.1.9) Gecko/20100402 Thunderbird/3.0.4 MIME-Version: 1.0 To: freebsd-net@freebsd.org Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: Warnings with TSO on em 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: Mon, 10 May 2010 20:49:25 -0000 Hello. For quite a while, I've been seeing in the logs a lot of messages like the following: snort: (snort_decoder) WARNING: IP dgm len < IP Hdr len! I'm not sure about this, but I suspect they started when I upgraded from 6.3 to 7.2. Today, while investigating another problem, I decided I had to move this out of the way; tcpdump showed zero-length IP packets and Google told me they might be the result of TSO on the em cards. So I checked and, yes, I had TSO enabled (by default). "ifconfig em0 -tso" made this noise go away. Now I'm looking for some more insight: if it's only a performance problem, I don't think I'll be hit, but could there be other side effects? # uname -a FreeBSD xxxxx.xxxxxxxx.xx 7.2-RELEASE-p7 FreeBSD 7.2-RELEASE-p7 #9: ... # pciconf -l|grep em em0@pci0:6:0:0: class=0x020000 card=0x34768086 chip=0x10968086 rev=0x01 hdr=0x00 em1@pci0:6:0:1: class=0x020000 card=0x34768086 chip=0x10968086 rev=0x01 hdr=0x00 I found some thread suggesting this might possibly be a bug with the em driver, but there was no follow up for more than a year. bye & Thanks av. P.S. I had another problem on this machine: when I bonded the the two "em" interfaces into a "lagg" one, carp stopped working properly. Now I wonder if the two issues might be related...