[sclug] RedHat Software Raid

[Alex Butcher]

On Wed, 26 Jan 2005, Scott Rixon wrote:

> I have a RedHat 9 server kicking around with 2x60GB disk in a software RAID.
>
> I have been looking in more detail about the software raid as I haven't 
> checked this box for a long time!
>
> I have just run a 'cat /dev/mdstat' and get this report..
>
> [root at greenflag root]# cat /proc/mdstat
> Personalities : [raid1]
> read_ahead 1024 sectors
> md2 : active raid1 hda3[0]
>     57930304 blocks [2/1] [U_]
>    md0 : active raid1 hda2[0]
>     610368 blocks [2/1] [U_]
>    md1 : active raid1 hda1[0]
>     104320 blocks [2/1] [U_]
>    unused devices: <none>
>
> Looks to me like on of the disks.. /dev/hdb might have failed as it's not 
> listed!

It might have gone offline due to a soft (read) error. If you just use mdadm
to add the partitions on hdb back again, the resync should either complete
successfully (and causing the drive firmware to map out any failed blocks as
it goes) or fail if the drive /really/ is dead. If you feel nervous about
continuing to use the drive, run smartctl -a on it and see if any of the
values are <= their respective thresholds.

This has happened to me once before. Once I re-sync'ed the drive, all was
well.

> Any thoughts? Advice? Is there a way to check the second drive and then 
> re-sync the mirror?
>
> Thanks,
>
> Scott

HTH,
Alex.
-- 
Alex Butcher      Brainbench MVP for Internet Security: www.brainbench.com
Bristol, UK                      Need reliable and secure network systems?
PGP/GnuPG ID:0x271fd950                         <http://www.assursys.com/>