From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <xfs-bounce@oss.sgi.com>
Received: with ECARTIS (v1.0.0; list xfs); Sun, 10 Sep 2006 17:55:52 -0700 (PDT)
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130])
	by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id k8B0tODW029447
	for <xfs@oss.sgi.com>; Sun, 10 Sep 2006 17:55:35 -0700
Date: Mon, 11 Sep 2006 10:54:29 +1000
From: David Chinner <dgc@sgi.com>
Subject: Re: Critical xfs bug in 2.6.17.11?
Message-ID: <20060911005429.GK10950339@melbourne.sgi.com>
References: <4504151F.6050704@rapidforum.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <4504151F.6050704@rapidforum.com>
Sender: xfs-bounce@oss.sgi.com
Errors-to: xfs-bounce@oss.sgi.com
List-Id: xfs
To: Christian Schmid <webmaster@rapidforum.com>
Cc: xfs@oss.sgi.com

On Sun, Sep 10, 2006 at 03:37:35PM +0200, Christian Schmid wrote:
> Hello.
> 
> Instead of a tmpfs, I use a raid 10 softraid. Unfortunately it crashed 
> after 10 hours of extreme activities (read/block-writes with up to 250 
> streams/deletes)
> 
> 12 gb memory-test successful. 2 cpu xeon smp system.
> 
> Tell me if this helps you:
> 
> Sep  9 18:08:49 inode430 kernel: [87433.143498] 0x0: 58 41 47 46 00 00 00 
> 01 00 00 00 00 00 04 34 a0 
> Sep  9 18:08:49 inode430 kernel: [87433.143672] Filesystem "md5": XFS 
> internal error xfs_alloc_read_agf at line 2176 of file fs/xfs/xfs_alloc.c.  
> Caller 0xffffffff80314069 

Hmm - bad read of the AGF during allocation, which causes a shutdown
due to cancelling a dirty transaction.

If you run "xfs_check -n /dev/md5" and "xfs_repair -n /dev/md5" do
they warn about any AGF related corruption?

Cheers,

Dave.
-- 
Dave Chinner
Principal Engineer
SGI Australian Software Group