Hi, Martin.
On Monday, 02 August 2010 08:30:03 +0200,
martin f krafft wrote:
> also sprach Daniel Bareiro <daniel-listas@gmx.net> [2010.08.02.0420 +0200]:
> > md2 : active raid5 sda3[0] sdd3[4](F) sdc3[2]
> > 2136170880 blocks level 5, 64k chunk, algorithm 2 [4/2] [U_U_]
> […]
> > That is to say, the RAID has four disks and failed both the spare
> > disk and other disk from array. What is unclear to me is why if
> > there are two active disks,
> There is no spare disk. The reason why sdd is listed as "faulty
> spare" is because it's out of sync. Remove and re-add it:
>
> mdadm --remove /dev/md2 /dev/sdd3
> mdadm --add /dev/md2 /dev/sdd3
# mdadm --remove /dev/md2 /dev/sdd3
mdadm: hot removed /dev/sdd3
# cat /proc/mdstat
Personalities : [raid1] [raid6] [raid5] [raid4]
md2 : active raid5 sdb3[4](S) sda3[0] sdc3[2]
2136170880 blocks level 5, 64k chunk, algorithm 2 [4/2] [U_U_]
md1 : active raid1 sda2[0] sdd2[3] sdc2[2]
19534976 blocks [4/3] [U_UU]
md0 : active raid1 sda1[0] sdd1[3] sdc1[2] sdb1[1]
979840 blocks [4/4] [UUUU]
unused devices: <none>
# mdadm --add /dev/md2 /dev/sdd3
mdadm: re-added /dev/sdd3
# cat /proc/mdstat
Personalities : [raid1] [raid6] [raid5] [raid4]
md2 : active raid5 sdd3[4](S) sdb3[5](S) sda3[0] sdc3[2]
2136170880 blocks level 5, 64k chunk, algorithm 2 [4/2] [U_U_]
md1 : active raid1 sda2[0] sdd2[3] sdc2[2]
19534976 blocks [4/3] [U_UU]
md0 : active raid1 sda1[0] sdd1[3] sdc1[2] sdb1[1]
979840 blocks [4/4] [UUUU]
unused devices: <none>
It draws attention to me that now both sdb3 and sdd3 are noticeable
like spare disks.
> > md2 : active raid5 sdb3[4](S) sda3[0] sdd3[5](F) sdc3[2]
> > 2136170880 blocks level 5, 64k chunk, algorithm 2 [4/2] [U_U_]
> This is indeed a bit strange, but the array might start syncing in
> the new disk (called a spare) as soon as you remove sdd3 (see
> above).
When I tried to add sdb3 was because the disk for some reason did not
appear with "cat /proc/mdstat" (not even like it was failed), although
when using "mdadm --detail /dev/md2" appear as both sdd3 and sdb3 as
removed. To what it can be due that sdb3 does not appear when doing "cat
/proc/mdstat"?
It is somewhat confusing to see a disk labeled like spare when it is
not. Is this "normal"?
If this is the case, the only thing I can think of is that after the
first disk failure, the system was in interim recovery mode, and during
it, the second disk would have failed.
Thanks for your reply.
Regards,
Daniel
--
Fingerprint: BFB3 08D6 B4D1 31B2 72B9 29CE 6696 BF1B 14E6 1D37
Powered by Debian GNU/Linux Lenny - Linux user #188.598
Attachment:
signature.asc
Description: Digital signature