0

Using Linux mdadm to recover from failed disk / partition. Here are titles of the steps I took so far - I can post detailed copy if it helps. ( I am not sure how to post file link here...)

The last response to "run" is :

nov25-1@nov251-desktop:~$ sudo mdadm --run /dev/md0
mdadm: failed to start array /dev/md/0: Input/output error
nov25-1@nov251-desktop:~$

What would be my next step to "activate" the array ?

nov25-1@nov251-desktop:~$ cat /proc/mdstat  1
nov25-1@nov251-desktop:~$ cat /proc/mdstat  2
nov25-1@nov251-desktop:~$ sudo mdadm --assemble /dev/md0    3
nov25-1@nov251-desktop:~$ sudo mdadm --assemble --force /dev/md0    3
nov25-1@nov251-desktop:~$ sudo mdadm --detail /dev/md0  4
nov25-1@nov251-desktop:~$ sudo mdadm --run /dev/md0recovering
Romeo Ninov
  • 5,263
  • 4
  • 20
  • 26
Jan Hus
  • 117
  • 4
  • Have you checked the status of the members of the array? What is the output of `dmesg` after the `mdadm --run` command? It could be that two of your devices are broken, which means all your data can be lost. – Tero Kilkanen Dec 05 '22 at 20:07
  • Please read last comment - on run. Here it is again nov25-1@nov251-desktop:~$ sudo mdadm --run /dev/md0 mdadm: failed to start array /dev/md/0: Input/output error nov25-1@nov251-desktop:~$ – Jan Hus Dec 06 '22 at 00:53
  • Here is a part of dmesg [ 3193.003771] md/raid:md0: not clean -- starting background reconstruction [ 3193.003843] md/raid:md0: device sdb17 operational as raid disk 1 [ 3193.003849] md/raid:md0: device sdb4 operational as raid disk 0 [ 3193.006900] md/raid:md0: cannot start dirty degraded array. [ 3193.007721] md/raid:md0: failed to run raid set. [ 3193.007733] md: pers->run() failed ... nov25-1@nov251-desktop:~$ nov25-1@nov251-desktop:~$ sudo mdadm --run /dev/md0 mdadm: failed to start array /dev/md/0: Input/output erro – Jan Hus Dec 06 '22 at 00:54
  • it saiz "starting background reconstruction" what does that mean ? – Jan Hus Dec 06 '22 at 00:55
  • 1
    How many devices are supposed to be in your RAID array? I see only two in your dmesg output (`sdb17`, and `sdb4` .. both of which seem to be on the same physical device - which is probably a Very Bad Idea). If your array consisted of three devices - it should start up in a degraded (at risk) state. If it consisted of more than three - then it won't be able to start up until the missing devices are made available again. Review `/proc/partitions` (or `lsblk`, or `blkid`) to find out which devices are currently "visible" – JMusgrove Dec 06 '22 at 09:53
  • Please edit the original question and add additional information there with proper formatting. The information is very hard to read when it is in comments. – Tero Kilkanen Dec 06 '22 at 17:04

1 Answers1

-1

Wouldn't the next step be to mount it?

I think I've used this guide before, maybe it will help you out?

https://kb.synology.com/en-uk/DSM/tutorial/How_can_I_recover_data_from_my_DiskStation_using_a_PC

You should definitely consider whether any disk has failed because trying to recover with a faulty drive still in place may be worse than removing it. I'd at least try to check out the SMART status of each disk. The read-only nature of the setup in the article means if you get it wrong it won't damage your array, you can just start over with more or different disks.

Jonathan Rixon
  • 39
  • 1
  • I have installed lvm2 and run sudo cat /proc/mdstat lvs getting this nov25-1@nov251-desktop:~$ sudo cat /proc/mdstat lvs Personalities : [raid6] [raid5] [raid4] [linear] [multipath] [raid0] [raid1] [raid10] 26 : inactive sde24[0](S) sde25[1](S) 204666880 blocks super 1.2 md0 : inactive sdb17[1] sdb4[4] 409335808 blocks super 1.2 unused devices: cat: lvs: No such file or directory nov25-1@nov251-desktop:~$ I had to delete some mdx output to fit here – Jan Hus Dec 06 '22 at 01:10
  • Using lvs give no output nov25-1@nov251-desktop:~$ sudo lvs [sudo] password for nov25-1: nov25-1@nov251-desktop:~$ – Jan Hus Dec 06 '22 at 01:13