zfs discuss - Apr 2007 - ZFS panics with dmu_buf_hold_array

Hi,

I have been wrestling with ZFS issues since yesterday when one of my disks sort
of died.  After much wrestling with "zpool replace" I managed to get
the new disk in and got the pool to resilver, but since then I have one error
left that I can''t clear:
  pool: data
 state: ONLINE
status: One or more devices has experienced an error resulting in data
        corruption.  Applications may be affected.
action: Restore the file in question if possible.  Otherwise restore the
        entire pool from backup.
   see: http://www.sun.com/msg/ZFS-8000-8A
 scrub: none requested
config:

        NAME        STATE     READ WRITE CKSUM
        data        ONLINE       0     0     0
          raidz     ONLINE       0     0     0
            c4d0s0  ONLINE       0     0     0
            c4d1s0  ONLINE       0     0     0
            c5d0s0  ONLINE       0     0     0
            c5d1s0  ONLINE       0     0     0

errors: The following persistent errors have been detected:

          DATASET  OBJECT  RANGE
          mos      15      0-524288

I am trying to backup the pool, but when I tar some of the filesystems, the
kernel panics with the following message:
Apr  3 16:37:56 columbia genunix: [ID 403854 kern.notice] assertion failed: 0 ==
dmu_buf_hold_array(os, object, offset, size, FALSE, FTAG, &numbufs,
&dbp), file: ../../common/fs/zfs/dmu.c, line: 386
Apr  3 16:37:56 columbia unix: [ID 100000 kern.notice] 
Apr  3 16:37:56 columbia genunix: [ID 353471 kern.notice] d4042ba0
genunix:assfail+5c (f975dfac, f975df90,)
Apr  3 16:37:56 columbia genunix: [ID 353471 kern.notice] d4042bf4
zfs:dmu_write+171 (d33c50d4, f, 0, f07)
Apr  3 16:37:56 columbia genunix: [ID 353471 kern.notice] d4042c74
zfs:space_map_sync+2f0 (d471b1f8, 1, d471b0)
Apr  3 16:37:56 columbia genunix: [ID 353471 kern.notice] d4042cdc
zfs:metaslab_sync+263 (d471b080, 2b2225, 0)
Apr  3 16:37:56 columbia genunix: [ID 353471 kern.notice] d4042d1c
zfs:vdev_sync+ba (d34d8680, 2b2225, 0)
Apr  3 16:37:56 columbia genunix: [ID 353471 kern.notice] d4042d6c
zfs:spa_sync+129 (d995cac0, 2b2225, 0)
Apr  3 16:37:56 columbia genunix: [ID 353471 kern.notice] d4042dc8
zfs:txg_sync_thread+1df (d44f9a80, 0)
Apr  3 16:37:56 columbia genunix: [ID 353471 kern.notice] d4042dd8
unix:thread_start+8 ()
Apr  3 16:37:56 columbia unix: [ID 100000 kern.notice] 
Apr  3 16:37:56 columbia genunix: [ID 672855 kern.notice] syncing file
systems...
Apr  3 16:37:56 columbia genunix: [ID 904073 kern.notice]  done
Apr  3 16:37:57 columbia genunix: [ID 111219 kern.notice] dumping to
/dev/dsk/c0d0s1, offset 419495936, content: kernel

I am running nv_38 and I was wondering if this error was fixed in nv_60 or not.

I tried to search the forums for something, but couldn''t find anything.
Thanks a lot in advance for your help. 
Bertrand.
 
 
This message posted from opensolaris.org

Bertrand Sirodot wrote:
> I am trying to backup the pool, but when I tar some of the filesystems, the
kernel panics with the following message:
This error is occurring because a critical piece of metadata can''t be 
read while we are trying to write out changes.  Try ensuring that you 
aren''t making any changes to the filesystems by mounting readonly
(''zfs
set readonly=on <pool>'') or at least with atime off.

--matt

Hi Matt,

I have tried both and every time the server panics with the same error message.
I guess my pool is foobar!!!!

Thanks a lot for the help,
Bertrand.
 
 
This message posted from opensolaris.org