Stale XFS mount for Kernel 2.6.25.14

[Dave Chinner]

On Wed, Jan 14, 2009 at 06:50:24PM -0500, Ngo, Andrew wrote:
> Dave,
> 
> From time to time I am still experiencing the stale mount.  From this
> last email, you want me to issue the 'echo t > /proc/sysrq-trigger'
> command to show all the processes in the machine.  Here is the capture
> of the /var/log/messages when I run the command. Please let me know what
> you find out.  Thanks...

> Jan 14 18:19:00 4003a6 kernel: date          R  running task        0 16073  24960
> Jan 14 18:19:00 4003a6 kernel:  ffff8101eb0f3d28 ffff8101eb0f3d6c 0000000000000000 ffff8101eb0f3d70
> Jan 14 18:19:00 4003a6 kernel:  0000000000000001 ffffffff802f7ba5 000000018020c228 0000000000001000
> Jan 14 18:19:00 4003a6 kernel:  0000000000000292 ffff8101eb0f3dc8 0000000000000005 ffff8101eb0f3de8
> Jan 14 18:19:00 4003a6 kernel: Call Trace:
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff802f7ba5>] xfs_bmap_search_extents+0x5b/0xe6
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff802fdfa9>] xfs_bmapi+0x26e/0xf89
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff803702f3>] __up_read+0x19/0x7f
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff80314887>] xfs_iunlock+0x57/0x79
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff8032ce6c>] xfs_free_eofblocks+0xb2/0x213
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff8029bc3a>] dput+0x26/0xd5
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff8028a033>] filp_close+0x5d/0x65
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff804c5581>] _write_lock_irq+0xf/0x10
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff80235e93>] do_exit+0x307/0x67a
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff80236275>] do_group_exit+0x6f/0x8a
> Jan 14 18:19:00 4003a6 kernel:  [<ffffffff8020bc9b>] system_call_after_swapgs+0x7b/0x80

That's the only stack trace that has any XFS references in it.
Do you have a 'date' process stuck on your machine or was this just
a "lucky" occurrence?

Cheers,

Dave.
-- 
Dave Chinner
david at fromorbit.com