[SOLVED] Software Raid

czezz · 09-02-2012, 10:22 AM

Assumptions:
- Linux software RAID1
- /dev/md0 = /dev/sdb1 + /dev/sdc1
- /dev/md0 mount point /data
- no spare disk

It is very easy to replace fail disk of the array when the system is up and running. Lets say disk /dev/sdc fail. A few steps need to be done to replace a disk:

Code:

mark bad disk:
# mdadm --manage /dev/md0 --fail /dev/sdc1

remove bad disk:
# mdadm --manage /dev/md0 --remove /dev/sdc1

reboot and replace bad disk

copy a disk layout for a new disk
# sfdisk -d /dev/sdb | sfdisk /dev/sdc

add a new disk to array and wait for resynchronization
# mdadm --manage /dev/md0 --add /dev/sdc1

Done

But, lets say we have a situation where one of disk fail and just after this machine has a power failure.
When it starts again there is no way to use /dev/md0 again and it means that all files at file system /data are not available.

Is there any procedure to restore it ?

macemoneta · 09-02-2012, 12:16 PM

In the scenario you proposed, the system will start with md0 in degraded mode following the fsck during initialization. You can then replace the failed drive as usual.

czezz · 09-02-2012, 12:45 PM

I have just tried to use fsck. I dont really how can it helps ?
Please, have a look at output:
Note - I have changed scenerio. In fact fail disk is /dev/sdc

Code:

root@cslack:~# cat /proc/mdstat
Personalities : [linear] [raid0] [raid1] [raid10] [raid6] [raid5] [raid4] [multipath]
md0 : inactive sdb1[2](S)
      203648 blocks super 1.2

unused devices: <none>

root@cslack:~# fsck -t ext4 /dev/md0
fsck from util-linux 2.21.2
e2fsck 1.42.5 (29-Jul-2012)
/sbin/e2fsck: Invalid argument while trying to open /dev/md0

The superblock could not be read or does not describe a correct ext2
filesystem.  If the device is valid and it really contains an ext2
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
    e2fsck -b 8193 <device>

root@cslack:~#  e2fsck -b 8193 /dev/md0
e2fsck 1.42.5 (29-Jul-2012)
e2fsck: Invalid argument while trying to open /dev/md0

The superblock could not be read or does not describe a correct ext2
filesystem.  If the device is valid and it really contains an ext2
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
    e2fsck -b 8193 <device>

macemoneta · 09-02-2012, 01:31 PM

The init software used by your distribution doesn't start degraded raid arrays (inactive). You will have to start the array manually.

Code:

mdadm -S /dev/md0
mdadm --assemble --scan --run

You can then fsck the array.

czezz · 09-02-2012, 01:56 PM

Code:

root@cslack:~# mdadm -S /dev/md0
mdadm: stopped /dev/md0

root@cslack:~# mdadm --assemble --scan --run

root@cslack:~# fsck.ext4 /dev/sdb1
e2fsck 1.42.5 (29-Jul-2012)
ext2fs_open2: Bad magic number in super-block
/sbin/e2fsck: Superblock invalid, trying backup blocks...
/sbin/e2fsck: Bad magic number in super-block while trying to open /dev/sdb1

The superblock could not be read or does not describe a correct ext2
filesystem.  If the device is valid and it really contains an ext2
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
    e2fsck -b 8193 <device>

root@cslack:~# e2fsck -b 8193 /dev/sdb1
e2fsck 1.42.5 (29-Jul-2012)
e2fsck: Invalid argument while trying to open /dev/sdb1

The superblock could not be read or does not describe a correct ext2
filesystem.  If the device is valid and it really contains an ext2
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
    e2fsck -b 8193 <device>

I have just tried to simply mount disk to see is there are any data yet... no luck

Code:

root@cslack:~# mount /dev/sdb1 /mnt/tmp/
NTFS signature is missing.
Failed to mount '/dev/sdb1': Invalid argument
The device '/dev/sdb1' doesn't seem to have a valid NTFS.
Maybe the wrong device is used? Or the whole disk instead of a
partition (e.g. /dev/sda, not /dev/sda1)? Or the other way around?

macemoneta · 09-02-2012, 02:36 PM

You assembled the raid array; /proc/mdstat should show it active. You must then fsck the array, not the underlying device.

czezz · 09-02-2012, 03:22 PM

Now I got this!

Thank you.

Code:

root@cslack:~# mdadm -S /dev/md0
mdadm: stopped /dev/md0
root@cslack:~# mdadm --assemble --scan --run
mdadm: /dev/md/0 has been started with 1 drive (out of 2).
root@cslack:~# cat /proc/mdstat
Personalities : [linear] [raid0] [raid1] [raid10] [raid6] [raid5] [raid4] [multipath]
md0 : active raid1 sdb1[0]
      200640 blocks super 1.2 [2/1] [U_]
root@cslack:~# sfdisk -d /dev/sdb | sfdisk /dev/sdc
root@cslack:~# mdadm --manage /dev/md0 --add /dev/sdc1
mdadm: added /dev/sdc1

root@cslack:~# cat /proc/mdstat
Personalities : [linear] [raid0] [raid1] [raid10] [raid6] [raid5] [raid4] [multipath]
md0 : active raid1 sdc1[2] sdb1[0]
      200640 blocks super 1.2 [2/2] [UU]

unused devices: <none>