Cannot remove raid1 array (mdadm)

Question

Hi,
The system is Oracle Linux 6.3 x86_64 running under VirtualBox. In order to remove the Linux RAID root volume I have started the system using the OL 6.3 UEK boot CD.
For some reason I'm unable to remove the /dev/md127 RAID device. Please see below:
<pre>
# mdadm -As
mdadm: /dev/md/vm009.example.com:1 has been started with 2 drives.
# cat /proc/mdstat
Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] [linear]
md127 : active raid1 sda2 [0] sdb2 [1]
20457404 blocks super 1.1 [2/2] [UU]
bitmap: 0/1 pages [0KB], 65536KB chunk
md126 : active raid1 sda1[0] sdb1[1]
511988 blocks super 1.0 [2/2] [UU]
</pre>
My first idea was to stop the device and remove it, which does not work:
<pre>
# *mdadm --stop /dev/md127*
mdadm: stopped /dev/md127
# *mdadm --remove /dev/md127*
mdadm: error opening /dev/md127: No such file or directory
</pre>
However, there is no error when I do not stop /dev/md127. But also the device is not removed:
<pre>
# mdadm -As
mdadm: /dev/md/vm009.example.com:1 has been started with 2 drives.
# *mdadm --remove /dev/md127*
# cat /proc/mdstat
Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] [linear]
md127 : active raid1 sda2 [0] sdb2 [1]
20457404 blocks super 1.1 [2/2] [UU]
bitmap: 0/1 pages [0KB], 65536KB chunk
md126 : active raid1 sda1[0] sdb1[1]
511988 blocks super 1.0 [2/2] [UU]
</pre>
I thought about to use -zero-superblock:
<pre>
# *mdadm --stop /dev/md127*
mdadm: stopped /dev/md127
# *mdadm --zero-superblock /dev/sdb2*
# mdadm -As
mdadm: /dev/md/vm009.example.com:1 has been started with 1 drive (out of 2).
# cat /proc/mdstat
Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] [linear]
md127 : active raid1 sda2 [0]
20457404 blocks super 1.1 [2/1] [U_]
bitmap: 0/1 pages [0KB], 65536KB chunk
md126 : active raid1 sda1[0] sdb1[1]
511988 blocks super 1.0 [2/2] [UU]
</pre>
However, if I do the same for the other drive, then I loose the complete enchilada.
<pre>
# *mdadm --stop /dev/md127*
mdadm: stopped /dev/md127
# *mdadm --zero-superblock /dev/sda2*
# mdadm -As
mdadm: No arrays found in config file or automatically
</pre>
Here is the detail of the /dev/md127 device:
<pre>
# mdadm --detail /dev/md127
/dev/md127:
Version : 1.1
Creation Time : Wed Nov 7 01:44:42 2012
Raid Level : raid1
Array Size : 20457404 (19.51 GiB 20.95 GB)
Used Dev Size : 20457404 (19.51 GiB 20.95 GB)
Raid Devices : 2
Total Devices : 2
Persistence : Superblock is persistent
Intent Bitmap : Internal
Update Time : Thu Nov 8 01:23:00 2012
State : active
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Name : vm009.example.com:1
UUID : 006a3ad6:c0590b94:5fddf632:5704f209
Events : 70
Number Major Minor RaidDevice State
0 8 2 0 active sync /dev/sda2
1 8 18 1 active sync /dev/sdb2
</pre>
Any ideas what I might be doing wrong? Thanks!

Answer 1

I found confirmation of the procedure to remove a RAID device in chapter 5.3.5. Removing a RAID Device at https://access.redhat.com/knowledge/docs/en-US/Red_Hat_Enterprise_Linux/5/html/Deployment_Guide/s2-raid-manage-removing.html. But apparently, there is no more reference in the documentation for version 6.
Anway, after some experimenting I finally I seem to have found the solution:
<pre>
# mdadm -S /dev/md126
mdadm: stopped /dev/md126
# mdadm -S /dev/md127
mdadm: stopped /dev/md127
mdadm --zero-superblock /dev/sdb2
mdadm --zero-superblock /dev/sda2
# mdadm -As
mdadm: /dev/md/vm009.example.com:0 has been started with 2 drives.
# cat /proc/mdstat
Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] [linear]
md127 : active raid1 sda1[0] sdb1[1]
511988 blocks super 1.0 [2/2] [UU]
</pre>
Apparently the trick is to stop all of the RAID devices and not just only the one to be deleted (md127). Something might get confused between md126 and md127. I have no explanation. I guess I do not need to remove the md127 RAID device once the RAID superblock is removed, but I wonder how this will work with initramfs, which should still have references to the RAID device.