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 q2L5KVa3054102 for ; Wed, 21 Mar 2012 00:20:32 -0500 Received: from ipmail04.adl6.internode.on.net (ipmail04.adl6.internode.on.net [150.101.137.141]) by cuda.sgi.com with ESMTP id 44IjuMXVG8SHvn6B for ; Tue, 20 Mar 2012 22:20:30 -0700 (PDT) Date: Wed, 21 Mar 2012 16:20:27 +1100 From: Dave Chinner Subject: Re: kernel: XFS (dm-7): xlog_space_left: head behind tail Message-ID: <20120321052027.GX5091@dastard> References: <20120321043031.GW5091@dastard> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <20120321043031.GW5091@dastard> 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: Gregory Machin Cc: xfs@oss.sgi.com On Wed, Mar 21, 2012 at 03:30:31PM +1100, Dave Chinner wrote: > On Wed, Mar 21, 2012 at 05:00:02PM +1300, Gregory Machin wrote: > > I have a CentOS 6.2 virtual machine in a vmware ESXi 4.0 host 4G Ram 4 > > virtaul cpus and with about 4 TB of disk space formatted with XFS . > > I'm seeing a lot of : > > > > Mar 21 10:49:42 nzhmlfpr05 kernel: XFS (dm-4): xlog_space_left: head behind tail > > Mar 21 10:49:42 nzhmlfpr05 kernel: tail_cycle = 129, tail_bytes = 20163072 > > Mar 21 10:49:42 nzhmlfpr05 kernel: GH cycle = 129, GH bytes = 20162880 > > Mar 21 10:49:42 nzhmlfpr05 kernel: XFS (dm-7): xlog_space_left: head behind tail > > Mar 21 10:49:42 nzhmlfpr05 kernel: tail_cycle = 5, tail_bytes = 333417984 > > Mar 21 10:49:42 nzhmlfpr05 kernel: GH cycle = 5, GH bytes = 333417792 > > Mar 21 10:49:42 nzhmlfpr05 kernel: XFS (dm-7): xlog_space_left: head behind tail > > Mar 21 10:49:42 nzhmlfpr05 kernel: tail_cycle = 5, tail_bytes = 333417984 > > Mar 21 10:49:42 nzhmlfpr05 kernel: GH cycle = 5, GH bytes = 333417792 > > can you send the complete set of output, including the first > occurrence of it? > > What workload are you running? Are you freezing you filesystem > frequently? If you are freezing regularly, then the patch below should fix the problem. Cheers, Dave. -- Dave Chinner david@fromorbit.com xfs: Account log unmount transaction correctly From: Dave Chinner There have been a few reports of this warning appearing recently: XFS (dm-4): xlog_space_left: head behind tail tail_cycle = 129, tail_bytes = 20163072 GH cycle = 129, GH bytes = 20162880 The common cause appears to be lots of freeze and unfreeze cycles, and the output from the warnings indicates that we are leaking around 8 bytes of log space per freeze/unfreeze cycle. When we freeze the filesystem, we write an unmount record and that uses xlog_write directly - a special type of transaction, effectively. What it doesn't do, however, is correctly acocunt for the log space it uses. The unmount record writes an 8 byte structure with a special magic number into the log, and the space this consumes is not accounted for in the log ticket tracking the operation. Hence we leak 8 bytes every unmount record that is written. Signed-off-by: Dave Chinner --- fs/xfs/xfs_log.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 98a9cb5..6db1fef 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -726,8 +726,9 @@ xfs_log_unmount_write(xfs_mount_t *mp) .lv_iovecp = ®, }; - /* remove inited flag */ + /* remove inited flag, and account for space used */ tic->t_flags = 0; + tic->t_curr_res -= sizeof(magic); error = xlog_write(log, &vec, tic, &lsn, NULL, XLOG_UNMOUNT_TRANS); /* _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs