Transfer zfs pool to new hardware

mailinglists mailinglists at debank.tv
Thu Dec 17 02:27:05 UTC 2009


All,

My machine had a failing motherboard so it was decided to put the drives
of the failed machine in a different machine to be able to continue
service.

The old server had 2 drives in a mirror setup (gmirror) and these were
recovered fine in the new server, these disks contain the OS so everything
is up and running again.

In addition to the mirrored drives there was a zfs (raidz) volume to
contain bulk data, this pool was built using 4 750G drives (all the same
disks) but this hasn't come up properly. I'm not sure if the disks were put
in the server in the same order and the device naming in the new machine
has changed also. One thing that stands out (for me) is that the /dev/ad12
comes up twice in the 'zpool status' output (see below), I'm not sure if
this is expected......

Old situation:
zfs raidz containing /dev/ad10, /dev/ad12, /dev/ad14 and /dev/ad16
(working fine)

In the new machine /dev/ad16 doesn't exist and the drives are numbered
different:
/dev/ad8, /dev/ad10, /dev/ad12 and /dev/ad14


My question is; can the data on this zfs pool be recovered?

Some more detailed information is below, but I'd be happy to provide
additional information if needed.

Current machine specs:

FreeBSD flappie.debank.tv 7.2-STABLE FreeBSD 7.2-STABLE #8: Mon Dec  7
00:45:32 CET 2009     root at flappie.debank.tv:/usr/obj/usr/src/sys/FLAPPIE
amd64

Dual core AMD processor, 6G memory, Tyan motherboard


>From the dmesg on the new box (all disk are recognized):

root at flappie:...home/rob#dmesg | grep "ad[0-9]"
ad4: 286188MB <Maxtor 6L300S0 BANC1G10> at ata2-master SATA150
ad6: 286188MB <Maxtor 6L300S0 BANC1G10> at ata3-master SATA150
ad8: 715404MB <SAMSUNG HD753LJ 1AA01108> at ata4-master SATA300
ad10: 715404MB <SAMSUNG HD753LJ 1AA01108> at ata5-master SATA300
ad12: 715404MB <SAMSUNG HD753LJ 1AA01108> at ata6-master SATA300
ad14: 715404MB <SAMSUNG HD753LJ 1AA01108> at ata7-master SATA300

Some more output from the various zfs commands:

root at flappie:...home/rob#zpool list
NAME   SIZE   USED  AVAIL    CAP  HEALTH  ALTROOT
data      -      -      -      -  FAULTED  -


root at flappie:...home/rob#zpool status data
  pool: data
 state: UNAVAIL
status: One or more devices could not be used because the label is missing
        or invalid.  There are insufficient replicas for the pool to
continue
        functioning.
action: Destroy and re-create the pool from a backup source.
   see: http://www.sun.com/msg/ZFS-8000-5E
 scrub: none requested
config:

        NAME        STATE     READ WRITE CKSUM
        data        UNAVAIL      0     0     0  insufficient replicas
          raidz1    UNAVAIL      0     0     0  insufficient replicas
            ad10    FAULTED      0     0     0  corrupted data
            ad12    FAULTED      0     0     0  corrupted data
            ad14    FAULTED      0     0     0  corrupted data
            ad12    ONLINE       0     0     0


root at flappie:...home/rob#zdb
data
    version=13
    name='data'
    state=0
    txg=4552247
    pool_guid=10110325244913303479
    hostid=376236222
    hostname='unset'
    vdev_tree
        type='root'
        id=0
        guid=10110325244913303479
        children[0]
                type='raidz'
                id=0
                guid=15844282688259604426
                nparity=1
                metaslab_array=14
                metaslab_shift=34
                ashift=9
                asize=3000606130176
                is_log=0
                children[0]
                        type='disk'
                        id=0
                        guid=12803254643974387344
                        path='/dev/ad10'
                        devid='ad:S13UJ1KQ236306'
                        whole_disk=0
                children[1]
                        type='disk'
                        id=1
                        guid=9558028561673012056
                        path='/dev/ad12'
                        devid='ad:S13UJ1KQ236305'
                        whole_disk=0
                children[2]
                        type='disk'
                        id=2
                        guid=12771778087909498297
                        path='/dev/ad14'
                        devid='ad:S13UJ1KQ236308'
                        whole_disk=0
                children[3]
                        type='disk'
                        id=3
                        guid=9647122164658652650
                        path='/dev/ad16'
                        devid='ad:S13UJ1KQ236309'
                        whole_disk=0



Thanks for your help,
Rob Evers

***deze e-mail is gescand door Onlinespamfilter.nl***


More information about the freebsd-questions mailing list