After successfully upgrading my 9.0-RELEASE ZFS system from a single 250GB drive to a mirror with two 500GB disks I now have a spare 250GB drive which I intend to put into another system. I upgraded by setting autoexpand to "on", adding each new drive with zpool attach sys /dev/gpt/sys0 /dev/gpt/sys1 followed by zpool attach sys /dev/gpt/sys1 /dev/gpt/sys2 and then removing the original drive with zpool detach sys /dev/gpt/sys0.
Although I'll be completely re-installing FreeBSD on the drive in the new location I thought it would be interesting to experiment to see if this would be a suitable procedure for copying a ZFS system to a different computer but when I installed the old drive in the other PC and tried to boot it failed with the following:
Then I tried booting the 9.0 installation DVD to see what I could discover about the ZFS pool on /dev/gpt/sys0
Perhaps it's not possible to re-use a drive this way after using zpool detach on it though I'd be interested to know if it can be done in case a similar need arises in the future. Perhaps I should have used zpool offline instead of zpool detach and then used the detach command after physically removing the drive?
I'm also puzzled that after installing in the other PC the pool now appears to be named zpool whereas it was originally named sys.
I'd welcome any suggestions on how the ZFS system on this drive might be recovered. Although I'll eventually be deleting the contents and re-installing I'm inclined to make the most of the opportunity of experimenting on a system where it won't matter if the contents get trashed - unless it's effectively trashed already
Although I'll be completely re-installing FreeBSD on the drive in the new location I thought it would be interesting to experiment to see if this would be a suitable procedure for copying a ZFS system to a different computer but when I installed the old drive in the other PC and tried to boot it failed with the following:
Code:
ZFS: i/o error - all block copies unavailable
ZFS: can't read MOS
ZFS: unexpected object set type 0
ZFS: unexpected object set type 0
FreeBSD/x86 boot
Default: sys:/boot/kernel/kernel
boot:
ZFS: unexpected object set type 0
FreeBSD/x86 boot
Default: sys:/boot/kernel/kernel
boot:
Code:
# [color="Blue"][B]zpool import -D[/B][/color]
pool: zpool
id: 5821244805494361020
state: FAULTED (DESTROYED)
status: One or more devices contains corrupted data.
action: The pool cannot be imported due to damaged devices or data.
The pool was destroyed, but can be imported using the '-Df' flags.
see: http://www.sun.com/msg/ZFS-8000-5E
config:
zpool FAULTED corrupted data
5933748989556188019 UNAVAIL corrupted data
# [color="Blue"][B]zpool import -Df zpool[/B][/color]
cannot import 'zpool': one or more devices is currently unavailable
# [color="Blue"][B]glabel status[/B][/color]
Name Status Components
ntfs/PRESARIO N/A ada0s1
msdosfs/PRESARIO_RP N/A ada0s2
gpt/bootcode0 N/A ada1p1
gptid/68a7d8c0-9b49-11e1-9c82-6cf0499e8897 N/A ada1p1
gpt/swap0 N/A ada1p2
gptid/68bb1608-9b49-11e1-9c82-6cf0499e8897 N/A ada1p2
gpt/sys0 N/A ada1p3
gptid/68cf7984-9b49-11e1-9c82-6cf0499e8897 N/A ada1p3
iso9660/FREEBSD_INSTALL N/A cd0
# [color="Blue"][B]gpart show ada1[/B][/color]
=> 34 488397101 ada1 GPT (232G)
34 256 1 freebsd-boot (128k)
290 8388608 2 freebsd-swap (4.0G)
8388898 480008237 3 freebsd-zfs (228G)
I'm also puzzled that after installing in the other PC the pool now appears to be named zpool whereas it was originally named sys.
I'd welcome any suggestions on how the ZFS system on this drive might be recovered. Although I'll eventually be deleting the contents and re-installing I'm inclined to make the most of the opportunity of experimenting on a system where it won't matter if the contents get trashed - unless it's effectively trashed already