Subject: Re: RAID disk failure
To: Wojciech Puchar <wojtek@chylonia.3miasto.net>
From: Jukka Marin <jmarin@pyy.jmp.fi>
List: netbsd-users
Date: 12/27/2001 18:05:29
On Thu, Dec 27, 2001 at 02:19:11PM +0200, Jukka Marin wrote:
> Well, I got a 80 GB disk to get RAID fully up again.  I plan to move to
> 80 GB disks later (to replace the other IBM before it dies).
> 
> How do I reconstruct my RAID1 system?  raid0 consisted of /dev/wd4e and
> /dev/wd5e and then /dev/wd4e failed.  After replacing /dev/wd4e and
> disklabeling it, I ended up with
> 
> root@kyyhky ~ # raidctl -s raid0
> Components:
>           component0: failed
> 	             /dev/wd5e: optimal
> No spares.
> component0 status is: failed.  Skipping label.
> 
> If I try to 'raidctl -R /dev/wd4e raid0' as suggested by the raidctl manpage,
> I get
> 
> raidctl: /dev/wd4e is not a component of this device

I did 'raidctl -a /dev/wd4e raid0' and 'raidctl -F component0 raid0'

and then 'shutdown -r now' and this time, my raid0 came up as healthy as
ever.  Don't know what was the problem - maybe I'm not allowed to 'reboot'
a system with raid?  (Is 'shutdown' safer?)

Now, maybe I should get some 80 GB Seagate Barracudas for a larger RAID.. 

  -jm