zfs discuss - Jun 2008 - Proper wayto do disk replacement in an A1000 storage array and raidz2.

I''m using ZFS and a drive has failed.
I am quite new to solaris and Frankly I seem to know more about ZFS and how it
works then I do the OS.

I have the hot spare taking over the failed disk and from here, do I need to
remove the disk on the OS side (if so what is proper) or do I need to take
action on the ZFS side first?
 
 
This message posted from opensolaris.org

Hi--

You can replace the failed disk and then detach the spare using the
general scenario described below. Some steps might be optional but I''m
pretty cautious about disk replacement, even when its this easy.

Cindy

1. Physically replace the failed disk.

2. Let ZFS know that you replaced the failed disk.

If you replace the failed disk with a disk in the same position, say for 
example, c1t0d0, then use syntax similar to this command:

# zpool replace pool-name c1t0d0

2. Run zpool status to confirm the replacement is successful and the
resilvering is complete.

3. Detach the spare, like this:

# zpool detach pool-name spare-disk

4. Run zpool status again to confirm that the spare is back in the
pool.

Cindy

Demian Phillips wrote:
> I''m using ZFS and a drive has failed.
> I am quite new to solaris and Frankly I seem to know more about ZFS and how
it works then I do the OS.
> 
> I have the hot spare taking over the failed disk and from here, do I need
to remove the disk on the OS side (if so what is proper) or do I need to take
action on the ZFS side first?
>  
>  
> This message posted from opensolaris.org
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Thanks. I have another spare so I replaced with that and it put the used spare
back to spare status.

I assume at this point once I replace the failed disk I just need to let solaris
see the change and then add it back into the pool as a spare (to replace the
spare I took out and used in the replace)?

I see some odd behavior related to the FC array and controller but that is not
ZFS related so I will have to post elsewhere about that fun.
 
 
This message posted from opensolaris.org

Hi--

I''m not quite sure about the exact sequence of events here, but it
sounds like you had two spares and replaced the failed disk with one of
the spares, which you can do manually with the zpool replace command.

The remaining spare should drop back into the spare pool if you detached
it. Check your zpool status output to confirm the spare status. If you
need to add it back to the pool, then you can use zpool add.

You can review more about managing spares, here:

http://docs.sun.com/app/docs/doc/817-2271/gcvcw?a=view

Your questions have reminded me that we need a better example of the
failed disk--spare replacement scenario in this guide so I will add
one.

Thanks,

Cindy

Demian Phillips wrote:
> Thanks. I have another spare so I replaced with that and it put the used
spare back to spare status.
> 
> I assume at this point once I replace the failed disk I just need to let
solaris see the change and then add it back into the pool as a spare (to replace
the spare I took out and used in the replace)?
> 
> I see some odd behavior related to the FC array and controller but that is
not ZFS related so I will have to post elsewhere about that fun.
>  
>  
> This message posted from opensolaris.org
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss