I have a 8 disk raidz2 pool (ZP02) that I am having some issues with.
The pool is using WD20EADS 2TB drives connected to a Intel SASUC8I
controller (LSI 1068E chip).
The pool was originally created when the machine was running SXDE 1/08.
I later installed OpenSolaris 2009.06 and imported the pool. I did not
upgrade the pool itself so it is still
using the older format.
Last week I noticed any commands that accessed the pool or filesystem
within would hang. I noticed unrecovered SCSI read errors for one drive
in the messages log. I was unable to gracefully reboot the machine and
had to power cycle it. After the reboot it would just hang when trying
to process the ZFS config. I booted off the 2009.06 Live CD and was able
to login via maintenance mode. I ran "zpool import" and it showed
the pool as being online, but in use by another system. It showed all
the drives as being online. At this point I ran "zpool -f import ZP02".
So far it has been almost 90 hours and there has been no output
to the terminal (just a blinking cursor). At this point I cannot tell if
the import command is doing anything.
Are there different steps I can try to recover this pool? Is there a way
to open another terminal when booted into system maintenance mode from
the Live CD? I'd like to see what the command is doing.
The machine has 4GB of ram. I will increase it to 8GB next week to see
if that helps. I understand that these WD drives may not be ideal for
ZFS. The pool has some important database backups on it.
I'd like to recover them if at all possible.
Thanks,
Peter.
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
