Hi Jay,
I think you mean you want to connect the disk with a potentially damaged
ZFS BE on another system and mount the ZFS BE for possible repair
purposes.
This recovery method is complicated by the fact that changing the root
pool name can cause the original system not to boot.
Other potential options that don't involve moving disks are:
1. Boot from a local or shared CD or netinstall server in single-user
mode and import the local root pool.
2. On a Solaris 10 system, you can attempt to boot failsafe mode
and roll back the ZFS BE snapshot.
3. Extensive disaster recovery:
- Send a remote copy of the ZFS BE snapshot back to the local system
- Create a flash archive of the root pool (Solaris 10)
Thanks,
Cindy
On 06/16/10 11:24, Jay Seaman wrote:
I have multiple servers, all with the same configuration of mirrored zfs root pools. I've
been asked how to take a potentially "damaged" disk from one machine and carry
it to another machine, in the event that some hw failure prevents fixing a boot problem
in place. So we have one half of mirror of rpool that we want to take to another system
that already has an rpool and we want to import the second rpool. The second system has
enough disk slots to add the third disk.
I believe the answer is to use
zpool import
to get the id number of the non-native rpool
then use
zpool import -f <id#> -R /mnt newpool
but I don't really have a way to check this...
Thanks
Jay
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
