Raid 1 "clean / degraded" after power outage and disk remova

Hi everyone!

On Friday evening we had a power outage and everything went down for hours. As my colleague had saved an important file to server which needed to be sent to the client, I took one of the raid disks out to access it via a docking station at home to finalize the report and send it on time.

Now, when coming back to the office some colleague had switched the NAS on already, before I could put the disk back (not sure if this had an impact on my problem).

The disk is back in the NAS, but seems like the NAS doesn't recognize the disk and says raid 1 "clean / degraded":

When clicking on "Details", this is what appears:

Version : 1.2
  Creation Time : Fri Dec 21 20:09:10 2012
     Raid Level : raid1
     Array Size : 976761424 (931.51 GiB 1000.20 GB)
  Used Dev Size : 976761424 (931.51 GiB 1000.20 GB)
   Raid Devices : 2
  Total Devices : 1
    Persistence : Superblock is persistent


    Update Time : Thu Mar 13 10:10:01 2014
          State : clean, degraded
 Active Devices : 1
Working Devices : 1
 Failed Devices : 0
  Spare Devices : 0


           Name : openmediavault:raid1
           UUID : 85c070e3:0215692c:f667e7dc:78abd5b2
         Events : 1712


    Number   Major   Minor   RaidDevice State
       0       0        0        0      removed
       1       8       32        1      active sync   /dev/sdc

So from what I can see it is missing /dev/sdb which was the disk that has been removed.

When I click on "Recover" the disk doesn't show up:

How do I get OMV get to recognize the second disk for the raid1 without having to delete everything on the disk? Is there any chance?

My setup is the following:

HP Microserver N36L
Disk1: 250GB (OMV installation only)
Disk2: 1TB (Raid1)
Disk3: 1TB (Raid1)
Disk4: 1TB (Backup disk for Raid1, which will be mirrored via Rsync daily or at least 2-3 times a week).

Then for additional Backup I have two external HD of 1TB which will mirror the Raid1 in turns, always on Friday.

So there is no chance to get this running without wiping the disk?

Is there a way to say what caused this? Is it the removal itself (the server was switched off), the accessing via a docking station (only reading, no writing) or the fact that the NAS was started before the disk was back?

Does this mean, whenever I take out one of the raid disks (for whatever reason) this will happen? Or shouldn't this be happening but it happened for whatever other reason?

Hi ryecoaaron,

I will try your tips.

Thank you also for the explanation. The "moving" around of the disk basically only happened because of the power outage. The colleague has only a PC without USV and our NAS doesn't have a USV either. As I needed to send the report on the same day and the file was on my colleagues computer (which was without electricity) and the NAS (which was without electricity), there was only one chance to access the file or do everything my colleague had done again myself. So I took a chance and brought the Raid1 disk home and accessed it directly. In this case I had basically no other chance. OK, I could have waited to see if electricity would come back, but experience has shown that this could have taken a few hours, so I decided to risk it...

I'll try your tips now and see if it helps in any way. But before it is probably clever to make a backup on the 3rd internal data drive.

Hi,

So I ran the first command "mdadm --stop /dev/md127" and this is the result:

mdadm: failed to stop array /dev/md127: Device or resource busy
Perhaps a running process, mounted filesystem or active volume group?

Obviously the Raid is still mounted and I can't unmount it via the interface, because there are shared folders on it. Does this mean I have to delete all the shared folders to do this and then recreate them? That would be a real bummer...

Ahh, that is a great idea. Have to check if this can be run from a usb stick, because my NAS does not have a DVD/CD drive.

Ohhh, I need to use pxe boot? I just wanted to ask if I can do it with a Ubuntu Live USB as well, because the created USB drive with RescueCD just wouldn't boot.

But I was worried that the commands would not be the same or wouldn't work when done with Ubuntu (maybe not recognized by OMV).

Will shut down the Ubuntu LiveCD which booted just fine and try it again with the Rescue one.

Tried to boot the usb stick with SystemRescueCD but even when choosing the option pxe boot, it won't boot from the stick, which the Ubuntu stick just boots fine. I will do some research if the commands are the same and if it is save to use the ubuntu live cd.

OK, then there is something wrong with the USB stick, because the only thing that I can see when booting from the stick with SystemRescueCD is a blinking underscore...

Well, somehow when I use the command it says that the package "mdadm" is not installed and then it wanted to insall the Postfix mail server... ohhh man. hehehehe.

trying the mdadm command with and without sudo:

ubuntu@ubuntu:~$ sudo mdadm --stop /dev/md127
mdadm: error opening /dev/md127: No such file or directory
ubuntu@ubuntu:~$ mdadm --stop /dev/md127
mdadm: must be super-user to perform this action

Hi! Tried

ls -la /dev/md*

as well, but it says that no such file available.

ubuntu@ubuntu:~$ ls -la /dev/md*
ls: cannot access /dev/md*: No such file or directory

I think I will have to rebuild the Raid like in the first suggestion. I wanted to avoid this, but at the moment it doesn't look like I will get this running.

When looking at the disk, the raid disks don't even show up in Ubuntu. Only the small OMV system disk and the backup disk...

looked with gparted into the disks and they are recognized, but unfortunately ubuntu now uses sdb for the Ubuntu Live memory stick.