faulty mdstat

faulty mdstat

[Eric BENARD / Free]

Hi,

here is what /proc/mdstat gives to me after a crash of the server.
2 drives (one on md1 and one on md4 have de (F).

This means they are faulty.

Should I do :
mdadm /dev/md1 -a /dev/ide/host2/bus0/target0/lun0/part5
mdadm /dev/md4 -a /dev/ide/host2/bus0/target0/lun0/part7

or is there an other way to solve the problem ?

Concerning md5 (raid 5), 3 drives are used. I assume the one marked [3] is not 
used. What can it be used for ?

Best regards
Eric

>Personalities : [raid0] [raid1] [raid5] 
>read_ahead 1024 sectors
>md0 : active raid1 ide/host2/bus1/target0/lun0/part1[1] 
ide/host2/bus0/target0/lun0/part1[0]
>      15936 blocks [2/2] [UU]
>      
>md1 : active raid1 ide/host2/bus1/target0/lun0/part5[1] 
ide/host2/bus0/target0/lun0/part5[0](F)
>      851328 blocks [2/1] [_U]
>      
>md3 : active raid0 ide/host2/bus1/target0/lun0/part6[1] 
ide/host2/bus0/target0/lun0/part6[0]
>      802944 blocks 64k chunks
>      
>md4 : active raid1 ide/host2/bus1/target0/lun0/part7[1] 
ide/host2/bus0/target0/lun0/part7[0](F)
>      1686720 blocks [2/1] [_U]
>      
>md2 : active raid1 ide/host2/bus1/target1/lun0/part1[1] 
ide/host2/bus0/target1/lun0/part1[0]
>      2554176 blocks [2/2] [UU]
>      
>md5 : active raid5 ide/host2/bus1/target1/lun0/part6[2] 
ide/host2/bus1/target0/lun0/part8[1] ide/host2/bus0/target1/lun0/part6[3] 
ide/host2/bus0/target0/lun0/part8[0]
>      228524416 blocks level 5, 64k chunk, algorithm 0 [3/3] [UUU]
>      
>unused devices: <none>
>
>

Re: faulty mdstat

[Neil Brown]

On Thursday August 7, ebenard@free.fr wrote:
> Hi,
> 
> here is what /proc/mdstat gives to me after a crash of the server.
> 2 drives (one on md1 and one on md4 have de (F).
> 
> This means they are faulty.
> 
> Should I do :
> mdadm /dev/md1 -a /dev/ide/host2/bus0/target0/lun0/part5
> mdadm /dev/md4 -a /dev/ide/host2/bus0/target0/lun0/part7
> 
> or is there an other way to solve the problem ?

Depends.  If the drive is really faulty, you want to replace it.
However it is was a transient error you need to remove and then re-add
the devices:

 mdadm /dev/md1 -r /dev/ide/host2/bus0/target0/lun0/part5
 mdadm /dev/md4 -r /dev/ide/host2/bus0/target0/lun0/part7

 mdadm /dev/md1 -a /dev/ide/host2/bus0/target0/lun0/part5
 mdadm /dev/md4 -a /dev/ide/host2/bus0/target0/lun0/part7

> 
> Concerning md5 (raid 5), 3 drives are used. I assume the one marked [3] is not 
> used. What can it be used for ?

Hot spare.

NeilBrown