Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754844AbYGRHGT (ORCPT ); Fri, 18 Jul 2008 03:06:19 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752831AbYGRHGJ (ORCPT ); Fri, 18 Jul 2008 03:06:09 -0400 Received: from mail.lang.hm ([64.81.33.126]:34111 "EHLO bifrost.lang.hm" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752440AbYGRHGI (ORCPT ); Fri, 18 Jul 2008 03:06:08 -0400 Date: Fri, 18 Jul 2008 00:07:15 -0700 (PDT) From: david@lang.hm X-X-Sender: dlang@asgard To: linux-kernel cc: jesse.brandeburg@intel.com, jeffrey.t.kirsher@intel.com, bruce.w.allan@intel.com, john.ronciak@intel.com, auke-jan.h.kok@intel.com, davem@davemloft.net Subject: wierd networking problem on 2.6.26 with e1000 Message-ID: User-Agent: Alpine 1.10 (DEB 962 2008-03-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; format=flowed; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1423 Lines: 27 apologies for the poor report, I'll learn more tomorrow. I've got a pair of boxes dual opteron dual-core with 16G ram LSI raid, and several quad E1000 cards. I had an emergancy requiring me to setup the boxes quickly for a new purpose, on one of them I left my old 2.6.22.1 32 bit build, but on the other I tried a 2.6.26 64 bit build. both boxes have the same 32 bit userspace and are running heartbeat to move a VIP from one to the other. when I logged into the 2.6.22.1 box and ssh'd over to the 2.6.26 box I had no communications problems at all, but if I moved the VIP (eth0:0) onto the 2.6.26 box and connected in to that I would occasionally see delays of a min or two where I lost all communications with the box. At the time this breakdown happened I was still connected in from the 2.6.22 box to the real IP of the second one with no problems, but attempts to connect to that real IP from other systems failed (unable to get an ARP response) I will be able to post my configs tomorrow, but the version difference between the two kernels will limit the usability of that. any suggestions on other things I should do when I setup a new test? David Lang -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/