Re: RAID5 (mdadm) array hosed after grow operation
On Fri, 16 Jan 2009 04:44:11 +1100, "Alex Samad" <alex@samad.com.au>
said:
> On Wed, Jan 14, 2009 at 07:45:24PM -0800, whollygoat@letterboxes.org
> wrote:
> >
> > I wonder if that would have helped with the larger drives. Too late:)
> > The smaller drives shouldn't have been bad. All I did to them was fail
>
> once you fail them they are marked as failed, plus if you did them one
> by one they would have different event id and be out of sync :(
Hmm, I've already recreated the array with the larger drives and
restored
the filesystem from backups, but I think I will try your --assume-clean
on the smaller drives to see if I can recover them. Could be a useful
little thing to have some practice with.
Thanks for the info. Bit by bit it's helping me to understand how md
works (and how to interpret the man page:)
Cheers
--
whollygoat@letterboxes.org
--
http://www.fastmail.fm - A fast, anti-spam email service.
Reply to: