From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cuda.sgi.com (cuda2.sgi.com [192.48.176.25]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id q8BKvq8S153274 for ; Tue, 11 Sep 2012 15:57:52 -0500 Received: from ipmail06.adl2.internode.on.net (ipmail06.adl2.internode.on.net [150.101.137.129]) by cuda.sgi.com with ESMTP id BoQQNpRyRhfhHJag for ; Tue, 11 Sep 2012 13:58:56 -0700 (PDT) Date: Wed, 12 Sep 2012 06:58:27 +1000 From: Dave Chinner Subject: Re: Internal error xfs_trans_cancel at line 1467 of file fs/xfs/xfs_trans.c Message-ID: <20120911205827.GD11511@dastard> References: MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: Jacek Luczak Cc: xfs@oss.sgi.com On Tue, Sep 11, 2012 at 09:47:49AM +0200, Jacek Luczak wrote: > Hi All, > > got this during heavy IO (parallel checkout of few large projects from SVN): > > XFS (dm-8): Corruption detected. Unmount and run xfs_repair > XFS (dm-8): Internal error xfs_trans_cancel at line 1467 of file > fs/xfs/xfs_trans.c. Caller 0xffffffffa03c9974 > > Pid: 11930, comm: svn Not tainted 3.4.10-1 #1 > Call Trace: > [] ? xfs_trans_cancel+0x56/0xd7 [xfs] > [] ? xfs_create+0x467/0x4cf [xfs] > [] ? xfs_vn_mknod+0xcb/0x160 [xfs] > [] ? vfs_create+0x6e/0xc7 > [] ? do_last+0x3a5/0x745 > [] ? path_openat+0xce/0x35f > [] ? do_filp_open+0x2c/0x72 > [] ? xfs_release+0x1ac/0x1cc [xfs] > [] ? alloc_fd+0x69/0xf2 > [] ? do_sys_open+0x107/0x18e > [] ? system_call_fastpath+0x16/0x1b That won't tell us what caused the problem. FWIW: http://xfs.org/index.php/XFS_FAQ#Q:_What_information_should_I_include_when_reporting_a_problem.3F i.e. mount options, filesystem size, geometry, space usage on the filesystem, etc are all relevant. Indeed, this issue is often caused by fragmented free space andnot being able to allocate inodes, though that usually just results in ENOSPC. Use the xfs_db "freespace" command to dump the freespace histogram when the error occurs and you've unmounted the filesystem. > Is it a known issue? I have not - yet - enabled this host in CI thus > can't tell if this can be reproduced. > > I've done repair but did not saved output. Were there any errors that were fixed? Cheers, Dave. -- Dave Chinner david@fromorbit.com _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs