my server (Debian Wheezy) has a failed disk. I have sda and sdb and according to the support of my hoster, sdb needs a replacement.

I send them this information with the mdstat: http://pastebin.com/GrzECixk

So they took the server down, checked the disks, said sdb is defect and I need to manually stop the software raid. I tried to use my hosters tutorial for that: http://wiki.hetzner.de/index.php/Festplattenaustausch_im_Software-RAID/en (I am not sure if that is the English version)

But the catch is that mdstat has changed since the technician is done:

# cat /proc/mdstat
Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md125 : active raid1 sdb1[1]
      2096064 blocks [2/1] [_U]

md126 : active (auto-read-only) raid1 sdb2[1]
      524224 blocks [2/1] [_U]

md127 : active (auto-read-only) raid1 sdb3[1]
      1462516672 blocks [2/1] [_U]

md2 : active raid1 sda3[0]
      1462516672 blocks [2/1] [U_]

md1 : active raid1 sda2[0]
      524224 blocks [2/1] [U_]

md0 : active raid1 sda1[0]
      2096064 blocks [2/1] [U_]

I tried removing md125, it said the device or resource is busy, so I set it to failed and tried again, still busy.

So what do I need to do now? How can I disable sdb first and give the technician the go to replace the disk?

Additional mdadm -E output for each sda partition: pastebin.com/J8ytbVgt

for sdb partitions: pastebin.com/RFyDygZj

  • Add the output of mdadm -E /dev/sd[ab][123] to your question. – psusi Jan 13 '14 at 21:07
  • added it: pastebin.com/J8ytbVgt – Maddis Jan 13 '14 at 21:35
  • Please also paste the output for /dev/sdb[123]. Does mdadm --manage /dev/md125 -S work for stopping md125? – Nils Jan 13 '14 at 22:11
  • pastebin.com/RFyDygZj And it stopped – Maddis Jan 13 '14 at 22:16
  • It is better to paste the output into the question instead of a link to pastebin. – psusi Jan 13 '14 at 23:21

I'm not sure how you got into this situation, but it looks like your array got split at some point and you may have ping ponged back and forth between the two across a few reboots, so you may now have some more updated files on one and different updated files on the other. You will need to figure out which drive is bad and may want to mount that one and see if there are any files that are newer there than on the other array and rescue them by copying them to the other before removing that drive.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.