Thinking of using ZFS/FBSD for a backup system

[aslam_mohamed at wordbank.com]

Something interesting, I just slot couple of 1TB SATA in to the same 
FreeBSD machine and re-created the zfs pool (this is just to make sure 
that without firewire the ZFS can  work well in FeeBSD) . When I reboot 
the machine I see the following appears on booting... This is the same 
thing appeared before with Firewire!!

ad4: 76324MB <HDS728080PLA380 40Y9028LEN PF2OA65A> at ata2-master SATA150
ad8: 953869MB <Seagate ST31000340NS SN05> at ata4-master SATA150
ad10: 953869MB <Seagate ST31000340NS SN05> at ata5-master SATA150
**GEOM: ad8: corrupt or invalid GPT detected.**
**GEOM: ad8: GPT rejected -- may not be recoverable.**
**GEOM: ad10: corrupt or invalid GPT detected.**
**GEOM: ad10: GPT rejected -- may not be recoverable.**
SMP: AP CPU #1 Launched!
Trying to mount root from ufs:/dev/ad4s1a
WARNING: ZFS is considered to be an experimental feature in FreeBSD.
ZFS filesystem version 6
ZFS storage pool version 6

Now I am running Rsync to backup all the file servers and it had done up 
to 1TB without complaining.. Currently, the Zpool status looks pretty OK..

  pool: backup
 state: ONLINE
 scrub: none requested
config:

        NAME        STATE     READ WRITE CKSUM
        backup      ONLINE       0          0              0
          ad8       ONLINE          0            0             0
          ad10      ONLINE        0           0              0

errors: No known data errors

I don't know in the future there will be any issue because of this Geom 
messages... is there any known problem like zfs/geom/gpt 
incompatibility?.. is it the ZFS causing this error(!) messages?..any 
thoughts?...

Thanks


Kris Kennaway wrote:
> aslam_mohamed at wordbank.com wrote:
>> Thanks for your reply. I did get some weired messages like ' vdev 
>> failure'  just two days before the ZFS pool disappeared, but I don't 
>> know whether this is from ZFS or another GEOM issue. I shut the 
>> machine down and restarted everything back and got it to do the sync 
>> again. Later I noticed zpool status was showing some big numbers like 
>> (83, 90 ...) on checksum column. I did the zpool clear and it cleared 
>> the numbers. Then I think couple days later the sync process stopped 
>> in the middle. When I investigated it, I found that zpool status is 
>> not there. I think the firewire/FreeBSD could be a potential 
>> contributor to this problem. Now I am trying to recover the backup 
>> data (if there is any in external disks!!) by connecting the external 
>> disks to SUN Solaris and getting the SUN/ZFS to see the pool!!!.. I 
>> cant think of any other ways to recover the backup!!!..
>
> It probably should have set off alarms for you when you first started 
> seeing checksum failures.  That indicates a serious problem!  At this 
> point, it sounds like your data might be highly corrupted, and 
> unfortunately, perhaps too corrupted to recover.
>
> Kris
>