From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1763324AbXGXRo7 (ORCPT ); Tue, 24 Jul 2007 13:44:59 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1758899AbXGXRoj (ORCPT ); Tue, 24 Jul 2007 13:44:39 -0400 Received: from disorder.primate.net ([198.144.194.12]:45776 "EHLO disorder.primate.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758650AbXGXRoi (ORCPT ); Tue, 24 Jul 2007 13:44:38 -0400 X-Greylist: delayed 1163 seconds by postgrey-1.27 at vger.kernel.org; Tue, 24 Jul 2007 13:44:38 EDT Date: Tue, 24 Jul 2007 10:25:14 -0700 From: Aaron Porter To: linux-kernel@vger.kernel.org Subject: miserable performance of 2.6.21 under network load Message-ID: <20070724172513.GB24543@primate.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.13 (2006-08-11) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I'm in the process up upgrading a pool of apache servers from 2.6.17.8 to 2.6.21.5, and we're seeing a pretty major change in behavior. Under identical network load, 2.6.21 has a load average more than 3 times higher, cpu 0 spends well over 90% of its time in interrupts (vs ~30% under 2.6.17). When we hit 3k apache sessions, ksoftirqd eats 100% of cpu0 and our network traffic drops off rapidly. The end result is that 2.6.17 performs twice as well under this load.