Disk replace in Linux raid software

Posted by Matteo Mattei / 28th December 2010

Sometimes happen that, after a long time of usage, one disk is going to be damage and starts to give some troubles… What’s happened when the disk is part of a software raid? If the failing disk is only one, all data are safe but you have to replace the disk as soon as possible in order to avoid very ugly surprises! I know that is always a very frustrating thing to change a disk from a raid software…

However, when a disk in RAID 5 or in RIAD 1 should be replaced, you have to follow these steps:

1. Take a look at /proc/mdstat, if is all ok (i.e. you have to replace a disk that contains some corrupted sectors but the raid is yet integer) mark the bad partition as failed (sdb1 in my case), otherwise continue from step 2:

   mdadm --manage /dev/md0 --fail /dev/sdb1

2. Remove that partition from the Raid array:

   mdadm --manage /dev/md0 --remove /dev/sdb1

3. Shutdown the pc, change the disk and power on the pc again.
4. At this point if you type

   cat /proc/mdstat

   you should see [U_U].

5. Copy now the partition table from a working disk (sda) to the new inserted disk (sdb):

   sfdisk -d /dev/sda | sfdisk /dev/sdb

6. Add the new partition to the array:

   mdadm --manage /dev/md0 --add /dev/sdb1

Done!

When you have finished, type

cat /proc/mdstat

and wait array rebuilding.

At the end of the process (that can take also some hours, depending on the size of the partition) you should have all “U” ([UUU]).