To check the status of RAID:

$ cat /proc/mdstat

Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md2 : active raid1 sda3[1]
730202368 blocks [2/1] [U_]
md1 : active raid1 sda2[1]
264960 blocks [2/1] [U_]
md0 : active (auto-read-only) raid1 sda1[1]
2102464 blocks [2/1] [U_]

you will see [_U] or [U_] if there is a broken RAID.

If required remove the broken hardrive from RAID from all md devices.

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

# mdadm --manage /dev/md2 --fail /dev/sdb3

Shutdown the machine and replace the hard drive.

Once the server is booted, you will see the new device (either sda or sdb depends on what drive is broken)

# ls -l /dev/sd*

Now we need to replicate the partition schema on the new drive.

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

// -d     Dump the partitions of a device

We can add the partition to the RAID now, you could verify the partitions with fdisk -l.

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

# mdadm --manage /dev/md2 --add /dev/sdb3

It will start sync the data and will be ready once completed.

You may verify the mdstat

# cat /proc/mdstat
Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md2 : active raid1 sda3[0] sdb3[1]
7302023 blocks [2/2] [UU]

md0 : active (auto-read-only) raid1 sda1[0] sdb1[1]
21024 blocks [2/2] [UU]

./arun