From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965388AbXC1Lbw (ORCPT ); Wed, 28 Mar 2007 07:31:52 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S965522AbXC1Lbw (ORCPT ); Wed, 28 Mar 2007 07:31:52 -0400 Received: from netops-testserver-3-out.sgi.com ([192.48.171.28]:40487 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S965388AbXC1Lbv (ORCPT ); Wed, 28 Mar 2007 07:31:51 -0400 Date: Wed, 28 Mar 2007 22:31:41 +1100 From: David Chinner To: Oliver Joa Cc: linux-kernel@vger.kernel.org, xfs-oss Subject: Re: Corrupt XFS -Filesystems on new Hardware and Kernel Message-ID: <20070328113141.GQ32597093@melbourne.sgi.com> References: <46094344.4090007@j-o-a.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <46094344.4090007@j-o-a.de> User-Agent: Mutt/1.4.2.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Mar 27, 2007 at 06:16:04PM +0200, Oliver Joa wrote: > Hi, > > since some weeks i try to get my new hardware running: > > Intel(R) Core(TM)2 CPU 6300 @ 1.86GHz > Intel DP965LT Mainboard > Seagate SATA-Harddisk in AHCI-Mode > > After some hours of running or after some heavy file-i/o > (find / | cpio -padm /test) I always get a corrupted > XFS-filesystem. What is the corruption message in the log from XFS? Can you please post that? Without it we really can't help you. Also, please check to see if there are any I/O errors in the log around the time the corruption message appears. > I used already the following Kernels: > 2.6.19.2 > 2.6.19.7 > 2.6.20.2 > 2.6.20.4 > > After xfs_repair I get damaged files in lost+found. > > I read in newsgroups that the write-cache of the harddisk > should be turned of, but the messages are all very old. That's really only an issue for crashes, not runtime failures. > I also often get a sata-bus-reset with the kernels 2.6.19.2 > and 2.6.20.2. I/O errors. That's what we need to isolate first. The reports in your logs are the first thing we need to seeee. Cheers, Dave. -- Dave Chinner Principal Engineer SGI Australian Software Group