xfs + raid5

[Dan Ohnesorg]

Dne Fri, Mar 05, 2004 at 12:37:27PM +0000, Jan Hradil napsal:

> Mar  4 12:01:23 brain kernel: raid5: switching cache buffer size, 512
> --> 4096
> Mar  4 12:01:54 brain kernel: raid5: switching cache buffer size, 4096
> --> 512
> Mar  4 12:02:00 brain kernel: raid5: switching cache buffer size, 512
> --> 4096

Tohle je znama vec, reseni si uz nepamatuju, ale je to o tom, ze se musi xfs
zakladat se stejnou velikosti bloku jako raid.


> 
> Mar  4 18:15:46 brain kernel: xfs_inotobp: xfs_imap()  returned an error
> 22 on md(9,0).  Returning error.
> Mar  4 18:15:46 brain kernel: xfs_iunlink_remove: xfs_inotobp() 
> returned an error 22 on md(9,0).  Returning error.
> Mar  4 18:15:46 brain kernel: xfs_inactive:     xfs_ifree() returned an
> error = 22 on md(9,0)
> Mar  4 18:15:46 brain kernel: xfs_force_shutdown(md(9,0),0x1) called
> from line 1844 of file xfs_vnodeops.c.  Return address = 0xc2ae1ff0
> Mar  4 18:15:46 brain kernel: raid5: switching cache buffer size, 4096
> --> 512
> Mar  4 18:15:46 brain kernel: Filesystem "md(9,0)": I/O Error Detected. 
> Shutting down filesystem: md(9,0)
> Mar  4 18:15:46 brain kernel: Please umount the filesystem, and rectify
> the problem(s)

Tohle ukazuje na pravdepodobny problem v HW. Nebo je to proste naborene xfs,
ktere potrebuje jeden prubeh xfs_recover aby se srovnalo.

zdravim
dan