ZFS RAIDZ Controller Disk Order

David Rawling djr at pdconsec.net
Wed Dec 29 13:07:41 UTC 2010


On 29/12/2010 11:08 PM, Charlie Mason wrote:
> Hi All,
>
> I have an old "server" which has FreeBSD on it. It has a cheap PCI
> raid controller in it, with the disks just set to pass through to the
> OS. Then the OS had a RAIDZ configured array form the 6 x 250gig disks
> passed through from the raid controller.
>
> I decided to upgrade to an array of 4 2TB disks running on new
> hardware (using the same server case). The disks were in a separate
> chassis to the motherboard, its quite complicated! So I copied the
> data off the old array via NFS to a separate 2TB disk. Then built the
> new array.
>
> Unfortunately I have realised I forgot to copy my old photos off the
> old array. I would really like to recover the old photos although, I
> suppose its not the end of the world if they are lost.
>
> As a fail safe I had left the old array disks as they were so, that I
> can just plug them back in if anything went wrong. Or so I thought! On
> plugging them back in zpool is complaining that they are all corrupt,
> except the first 2. It seems fairly unlikely I have got that unlucky.
> So I was wondering, does the order they were in the controller matter.
> They are all being presented in the same range of addresses da0
> through da5 but they are probably connected up to different ports, so
> da0 is now da2 and so on.
>
> Can anyone think of a way to get them back in the correct order if it
> does matter, other than trial and error. Can I find out what number
> each one was in the pool from the disk somehow?
>
> Thanks,
>
> Charlie
> _______________________________________________
> freebsd-questions at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-questions
> To unsubscribe, send any mail to "freebsd-questions-unsubscribe at freebsd.org"
>
Is it perhaps possible to disconnect the disks again, boot the system, and 
remove the ZFS pool cache (which location escapes me for the moment). Then you 
should be able to import the pool again using the -f switch (force). I think 
if you are using the cache, the order matters. If you're importing a fresh 
pool, the system simply needs to find enough member disks.

I'm not a ZFS expert though...

Dave.

-- 
David Rawling
Principal Consultant

PD Consulting And Security
20 Goodin Road
Baulkham Hills, NSW 2153
Australia
Mob: +61 412 135 513
Email: djr at pdconsec.net

Please note that whilst we take all care, neither PD Consulting and Security nor the sender accepts any responsibility for viruses and it is your responsibility to scan for viruses. The contents are intended only for use by the addressee and may contain confidential and/or privileged material and any use by other than the intended recipient is prohibited. If you received this in error, please inform the sender and/or addressee immediately and delete the material.



More information about the freebsd-questions mailing list