ZFS scrub is finished?

[Steve Bertrand]

xorquewasp at googlemail.com wrote:
> On 2009-07-11 16:37:20, Steve Bertrand wrote:
>> xorquewasp at googlemail.com wrote:
>>> On 2009-07-11 15:08:04, Steve Bertrand wrote:
>>>> Perhaps it is as simple as cron-ing something like the following (of
>>>> course, replace "zpool scrub" with my cat commands.
>>> Problem is that 'zpool scrub' immediately returns with a success
>>> error code if command line option parsing went OK and the specified
>>> pools actually exist. You have to manually check the output of 'zpool status'
>>> to see if the scrub executed with no errors (and that may be hours
>>> later).
>> Ok. What is your ultimate goal? Is it to:
>>
>> - run the first scrub
>> - search for an "ok" err status upon return
>> - begin the second scrub
> 
> Yep.
> 
>> If (in your case) the first scrub fails, is it still feasible to run the
>> second scrub anyway? If it is, then chaining the commands together
>> *should* work (however, I've never tested whether '&&' will continue
>> upon certain 'failures', but so long as the first command _exits_, you
>> should be alright).
> 
> I've just realised that the documentation is somewhat misleading.
> 
> Seems I can actually just do:
> 
> # zpool scrub backup
> # zpool scrub storage
> 
> ... and it'll run for both. The manual page made it sound as if
> only one scrub operation could be running at one time, across all
> pools.

Nice!

Now ensure you submit some text to the maintainers that may help clarify
this in the documentation ;)

Steve
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3233 bytes
Desc: S/MIME Cryptographic Signature
Url : http://lists.freebsd.org/pipermail/freebsd-questions/attachments/20090711/07f341a2/smime.bin