XFS shutdown in xfs_iunlink_remove() (was Re: 2.6.25: swapper: page allocation failure. order:3, mode:0x4020)

[David Chinner <dgc@sgi.com>]

[Please cc the XFS list (xfs@oss.sgi.com) on bug reports or put "XFS" in the
subject line so ppl know to pay attention to your report.]

On Mon, May 05, 2008 at 03:41:29PM +0200, Marco Berizzi wrote:
> Hi.
> Just few minutes ago an xfs filesystem
> was shutdown with these errors:
> 
> May  5 14:31:38 Pleiadi kernel: xfs_inotobp: xfs_imap()  returned an
> error 22 on hda8.  Returning error.
> May  5 14:31:38 Pleiadi kernel: xfs_iunlink_remove: xfs_inotobp()
> returned an error 22 on hda8.  Returning error.
> May  5 14:31:38 Pleiadi kernel: xfs_inactive:^Ixfs_ifree() returned an
> error = 22 on hda8
> May  5 14:31:38 Pleiadi kernel: xfs_force_shutdown(hda8,0x1) called from
> line 1737 of file fs/xfs/xfs_vnodeops.c.  Return address = 0xc01e6fde
> May  5 14:31:38 Pleiadi kernel: Filesystem "hda8": I/O Error Detected.
> Shutting down filesystem: hda8
> May  5 14:31:38 Pleiadi kernel: Please umount the filesystem, and
> rectify the problem(s)
> May  5 14:36:43 Pleiadi kernel: xfs_force_shutdown(hda8,0x1) called from
> line 420 of file fs/xfs/xfs_rw.c.  Return address = 0xc01eaf21

Is it reproducable? What were you doing at the time the problem occurred?

Cheers,

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