http://defect.opensolaris.org/bz/show_bug.cgi?id=1016
Darren J Moffat <darrenm at opensolaris.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |INCOMPLETE
--- Comment #1 from Darren J Moffat <darrenm at opensolaris.org> 2008-04-08
06:03:31 ---
Slightly different stack trace but I managed to reproduce this by doing:
while true ; do
zfs create tank/clear
cp hamlet.txt /tank/clear
sync
zfs destroy tank/clear
done
panic[cpu2]/thread=ffffff001fcabc80: assertion failed: spa_version(dp->dp_spa)
< 9ULL || used == ds->ds_phys->ds_unique_bytes, file:
../../common/fs/zfs/dsl_dataset.c, line: 1493
ffffff001fcab8b0 genunix:assfail+7e ()
ffffff001fcaba40 zfs:dsl_dataset_destroy_sync+a9c ()
ffffff001fcaba90 zfs:dsl_sync_task_group_sync+fc ()
ffffff001fcabb00 zfs:dsl_pool_sync+fc ()
ffffff001fcabbb0 zfs:spa_sync+21a ()
ffffff001fcabc60 zfs:txg_sync_thread+2b0 ()
ffffff001fcabc70 unix:thread_start+8 ()
It must have been the second part of the || that failed so for some reason used
wasn't equal the ds_unique_bytes.
Interestingly this has been seen, and supposedly fixed, before see bugster bug#
6448999. However that fix doesn't seem to be relevant to us since I had no
clones or snapshots involved in reproducing this problem.
I'm running the same stress test on sparc using vanilla snv_84 just to be sure
this isn't a generic ZFS bug.
--
Configure bugmail: http://defect.opensolaris.org/bz/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
