zfs discuss - Jul 2011 - How to recover -- LUNs go offline, now permanent errors?

I recently had an issue with my LUNs from our storage unit going offline.  This
caused the zpool to get numerous errors on the luns.  The pool is on-line, and I
did a scrub, but one of the raid sets is
degraded:

           raidz2-3                                 DEGRADED     0     0     0
            c7t60001FF011C6F3103B000100001D1BF1d0  DEGRADED     0     0     0 
too many errors
            c7t60001FF011C6F30239000100001D1BF1d0  DEGRADED     0     0     0 
too many errors
            c7t60001FF011C6F2F537000100001D1BF1d0  DEGRADED     0     0     0 
too many errors
            c7t60001FF011C6F2E435000100001D1BF1d0  DEGRADED     0     0     0 
too many errors
            c7t60001FF011C6F2D233000100001D1BF1d0  DEGRADED     0     0     0 
too many errors
            c7t60001FF011C6F2A931000100001D1BF1d0  DEGRADED     0     0     0 
too many errors
            c7t60001FF011C6F29A2F000100001D1BF1d0  DEGRADED     0     0     0 
too many errors
            c7t60001FF011C6F2682D000100001D1BF1d0  DEGRADED     0     0     0 
too many errors
            c7t60001FF011C6F24C2B000100001D1BF1d0  DEGRADED     0     0     0 
too many errors
            c7t60001FF011C6F21929000100001D1BF1d0  DEGRADED     0     0     0 
too many errors

Also I have the following:
errors: Permanent errors have been detected in the following files:

        <0x3a>:<0x3b04>

Originally, there was a file, and then a directory listed, but I removed them. 
Now I''m stuck with
the hex codes above.  How do I interpret them?  Can this pool be recovered, or
basically how do
I proceed?

The system is Solaris 10 U9 with all recent patches.

Thanks,

David
-- 
This message posted from opensolaris.org

Hi David,

If the permanent error is in some kind of metadata, then it doesn''t
translate to a specific file name.

You might try another zpool scrub and then a zpool clear to see if
it clears this error.

Thanks,

Cindy

On 07/13/11 12:47, David Smith wrote:
> I recently had an issue with my LUNs from our storage unit going offline. 
This caused the zpool to get numerous errors on the luns.  The pool is on-line,
and I did a scrub, but one of the raid sets is
> degraded:
> 
>            raidz2-3                                 DEGRADED     0     0   
0
>             c7t60001FF011C6F3103B000100001D1BF1d0  DEGRADED     0     0    
0  too many errors
>             c7t60001FF011C6F30239000100001D1BF1d0  DEGRADED     0     0    
0  too many errors
>             c7t60001FF011C6F2F537000100001D1BF1d0  DEGRADED     0     0    
0  too many errors
>             c7t60001FF011C6F2E435000100001D1BF1d0  DEGRADED     0     0    
0  too many errors
>             c7t60001FF011C6F2D233000100001D1BF1d0  DEGRADED     0     0    
0  too many errors
>             c7t60001FF011C6F2A931000100001D1BF1d0  DEGRADED     0     0    
0  too many errors
>             c7t60001FF011C6F29A2F000100001D1BF1d0  DEGRADED     0     0    
0  too many errors
>             c7t60001FF011C6F2682D000100001D1BF1d0  DEGRADED     0     0    
0  too many errors
>             c7t60001FF011C6F24C2B000100001D1BF1d0  DEGRADED     0     0    
0  too many errors
>             c7t60001FF011C6F21929000100001D1BF1d0  DEGRADED     0     0    
0  too many errors
> 
> Also I have the following:
> errors: Permanent errors have been detected in the following files:
> 
>         <0x3a>:<0x3b04>
> 
> Originally, there was a file, and then a directory listed, but I removed
them.  Now I''m stuck with
> the hex codes above.  How do I interpret them?  Can this pool be recovered,
or basically how do
> I proceed?
> 
> The system is Solaris 10 U9 with all recent patches.
> 
> Thanks,
> 
> David

Cindy,

Thanks for the reply.  I''ll get that a try and then send an update.

Thanks,

David
-- 
This message posted from opensolaris.org

Cindy,

I gave your suggestion a try.  I did the zpool clear and then did another zpool
scrub  and all is happy now.  Thank you for your help.

David
-- 
This message posted from opensolaris.org