On 5/8/15 11:29 AM, Clug wrote:
> On Fri, 8 May 2015, Linda Kateley wrote:
>
>>> 1. Add new drive (/dev/sdj)
>>> 2. zpool offline media
>> Offline is used for a disk not a whole pool
>
> Oh, ok. Wouldn't it be a good idea to take the whole pool down? I'm 
> going to shut down everything that uses that data, for sure.
This another one of those great features of zfs (and why i love it sooo 
much) you don't have to take it down.. you can add disks and do admin 
live!!!
>
>> Yea, it depends on how much data it needs to read to rebuild. If it's 
>> empty probably a couple seconds, if it is a 11+1 full 4TB drives 
>> maybe several days..
>
> The pool is made of two 3+1 3TB raidz1, and it's like 60% full. I 
> don't know what is where but when that one disk was not functioning 
> correctly I was seeing a LOT of errors. A scrub of that pool takes at 
> least 60 hours, even if I manage to not access it at all (yes, the 
> better part of a week). This thing was DEFINITELY not built for speed, 
> it was built for reliability (and space) (:
Yea, I don't think zol has delay code. The freebsd has delay code so 
that during scrub or resilver, it puts a delay in between ios and keeps 
the pool up and at almost full speed. Makes those admin tasks longer but 
it is nice


>
> -- 
>
> _______________________________________________
> TCLUG Mailing List - Minneapolis/St. Paul, Minnesota
> tclug-list at mn-linux.org
> http://mailman.mn-linux.org/mailman/listinfo/tclug-list

-- 
Linda Kateley
Kateley Company
Skype ID-kateleyco
http://kateleyco.com