Removing a device in "removed" state from Linux software RAID array
- by Sahasranaman MS
My workstation has two disks(/dev/sd[ab]), both with similar partitioning. /dev/sdb failed, and cat /proc/mdstat stopped showing the second sdb partition.
I ran mdadm --fail and mdadm --remove for all partitions from the failed disk on the arrays that use them, although all such commands failed with
mdadm: set device faulty failed for /dev/sdb2: No such device
mdadm: hot remove failed for /dev/sdb2: No such device or address
Then I hot swapped the failed disk, partitioned the new disk and added the partitions to the respective arrays. All arrays got rebuilt properly except one, because in /dev/md2, the failed disk doesn't seem to have been removed from the array properly. Because of this, the new partition keeps getting added as a spare to the partition, and its status remains degraded.
Here's what mdadm --detail /dev/md2 shows:
[root@ldmohanr ~]# mdadm --detail /dev/md2
/dev/md2:
Version : 1.1
Creation Time : Tue Dec 27 22:55:14 2011
Raid Level : raid1
Array Size : 52427708 (50.00 GiB 53.69 GB)
Used Dev Size : 52427708 (50.00 GiB 53.69 GB)
Raid Devices : 2
Total Devices : 2
Persistence : Superblock is persistent
Intent Bitmap : Internal
Update Time : Fri Nov 23 14:59:56 2012
State : active, degraded
Active Devices : 1
Working Devices : 2
Failed Devices : 0
Spare Devices : 1
Name : ldmohanr.net:2 (local to host ldmohanr.net)
UUID : 4483f95d:e485207a:b43c9af2:c37c6df1
Events : 5912611
Number Major Minor RaidDevice State
0 8 2 0 active sync /dev/sda2
1 0 0 1 removed
2 8 18 - spare /dev/sdb2
To remove a disk, mdadm needs a device filename, which was /dev/sdb2 originally, but that no longer refers to device number 1. I need help with removing device number 1 with 'removed' status and making /dev/sdb2 active.
