Hello, I'm having a big problem here, disastrous maybe.
I have a zpool consisting of 4x500GB SATA drives, this pool was born on S10U4
and was recently upgraded to snv85 because of iSCSI issues with some initiator.
Last night I was doing housekeeping, deleting old snapshots. One snapshot
failed to delete because it had a dependant clone. So I try to destroy that
clone: Everything went wrong from there.
The deletion was taking an excessively long time (over 40 minutes). zpool
status hungs when I call it. zfs list too. zpool iostat showed disk activity.
Other services non dependant on the pool were running, and the iSCSI this
machine was serving was unbearably slow.
At one point, I lost all iSCSI, SSH, web, and all other services. Ping still
worked. So I go to the server and notice that the fans are running at 100%. I
try to get a console (local VGA+keyboard) but the monitor shows "no signal". No
disk activity seemed to be happening at the moment. So, I do the standard
procedure (reboot). Solaris boots but stops at "hostname: blah". I see disk
activity from the pool disks, so I let it boot. 30 minutes later, still didn't
finish. I thought (correctly) that the system was waiting to mount the ZFS
before booting, but for some reason it doesn't. I call it the day and let the
machine do its thing.
8 hours later I return. CPU is cold, disks are idle and... solaris stays at the
same "hostname: blah". Time to reboot again, this time in failsafe. zpool
import shows that the devices are detected and online. I delete
/etc/zfs/zpool.cache and reboot. Solaris starts normally with all services
running, but of course no zfs. zpool import shows the available pool, no
errors. I do zpool import -f pool... 20 minutes later I'm still waiting for the
pool to mount. zpool iostat shows activity:
capacity operations bandwidth
pool used avail read write read write
---------- ----- ----- ----- ----- ----- -----
tera 1.51T 312G 274 0 1.61M 2.91K
tera 1.51T 312G 308 0 1.82M 0
tera 1.51T 312G 392 0 2.31M 0
tera 1.51T 312G 468 0 2.75M 0
but the mountpoint /tera is still not populated (and zpool import still doesn't
exit).
zpool status shows:
pool: tera
state: ONLINE
scrub: none requested
config:
NAME STATE READ WRITE CKSUM
tera ONLINE 0 0 0
raidz1 ONLINE 0 0 0
c1d0 ONLINE 0 0 0
c2d0 ONLINE 0 0 0
c3d0 ONLINE 0 0 0
c4d0 ONLINE 0 0 0
errors: No known data errors
What's going on? Why is taking so long to import?
Thanks in advance,
Hernan
This message posted from opensolaris.org
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
