can't mount raid0 drives

D3smond · 05-01-2017, 11:38 AM

Hi,

I try to recover my synology raid0 drives. I have a 2 bay nas with two 3tb hdds which suddently stopped working. It seems like on of the disk is broken. However, I tried to mount the drives in my computer and can't figgure out why this isn't working. It seems like the second disk can't added to the array. Here is the output:

Code:

root@avira:~# mdadm --misc --examine /dev/sda3
/dev/sda3:
          Magic : a92b4efc
        Version : 1.2
    Feature Map : 0x0
     Array UUID : 1476862c:52847f5b:60f1212a:cd870058
           Name : nas:2
  Creation Time : Mon Oct 22 16:15:45 2012
     Raid Level : linear
   Raid Devices : 2

 Avail Dev Size : 5851088833 (2790.02 GiB 2995.76 GB)
  Used Dev Size : 0
    Data Offset : 2048 sectors
   Super Offset : 8 sectors
          State : active
    Device UUID : 09a00aaa:86bf43db:5c424ec5:71c9c2f9

    Update Time : Wed Apr  5 19:36:58 2017
       Checksum : b852d9eb - correct
         Events : 2

       Rounding : 64K

   Device Role : Active device 0
   Array State : A. ('A' == active, '.' == missing)
root@avira:~# mdadm --misc --examine /dev/sdb3
/dev/sdb3:
          Magic : a92b4efc
        Version : 1.2
    Feature Map : 0x0
     Array UUID : 1476862c:52847f5b:60f1212a:cd870058
           Name : nas:2
  Creation Time : Mon Oct 22 16:15:45 2012
     Raid Level : linear
   Raid Devices : 2

 Avail Dev Size : 5851088833 (2790.02 GiB 2995.76 GB)
  Used Dev Size : 0
    Data Offset : 2048 sectors
   Super Offset : 8 sectors
          State : active
    Device UUID : 92e21fb2:3defd6a7:600e7750:4a7478f5

    Update Time : Wed Apr  5 19:36:58 2017
       Checksum : 13d9c309 - correct
         Events : 2

       Rounding : 64K

   Device Role : Active device 32769
   Array State : A. ('A' == active, '.' == missing)
root@avira:~# mdadm --assemble /dev/md2 /dev/sda3 /dev/sdb3 --verbose
mdadm: looking for devices for /dev/md2
mdadm: /dev/sda3 is identified as a member of /dev/md2, slot 0.
mdadm: /dev/sdb3 is identified as a member of /dev/md2, slot 32769.
mdadm: no uptodate device for slot 1 of /dev/md2
mdadm: added /dev/sda3 to /dev/md2 as 0
mdadm: /dev/md2 assembled from 1 drive - not enough to start the array.
root@avira:~# cat /proc/mdstat 
Personalities : 
md2 : inactive sda3[0](S)
      2925544416 blocks super 1.2
       
unused devices: <none>

any ideas whats wrong here?

jefro · 05-01-2017, 08:37 PM

My guess. https://unix.stackexchange.com/quest...tart-the-array

The nas should have been able to look at all of this too assuming it was good. SMART status and almost full linux command via ssh.

"

The mdadm --examine /dev/sdc4 output shows why the raid device is not coming online:

Raid Level : linear

A linear-raid device needs all the devices in the RAID group to be able to start up.

https://raid.wiki.kernel.org/index.p...up#Linear_mode
"

D3smond · 05-06-2017, 08:09 AM

Thanks for the answer. I know that it needs the 2 disks to be able to start. But I don't understand why it isn't starting? I mean both disks are available and I still see the mdadm information for both disks. I didn't see any error, it just doesn't work.

smallpond · 05-06-2017, 08:33 AM

When a drive in a raid throws an error it is marked failed and dropped from the raid. In raid1 you would resync it with the good drive to recover. Since you can't do that in raid0 you need to just add the drive back into the array.

D3smond · 05-06-2017, 08:36 AM

Quote:

Originally Posted by smallpond

When a drive in a raid throws an error it is marked failed and dropped from the raid. In raid1 you would resync it with the good drive to recover. Since you can't do that in raid0 you need to just add the drive back into the array.

Okay, and how can add this back to the array? Is there a way to verify if this was dropped due to an error?

smallpond · 05-06-2017, 09:05 AM

Better if you read the man page for mdadm than expect someone to give you a command who can't see the state of your system.

D3smond · 05-06-2017, 10:07 AM

The man page didn't really helped and all I tried doesn't worked. Actually thats why I ended up here in the first place. If I try to "re add" the volume I got this error:

Code:

root@avira:~# mdadm --manage /dev/md2 --re-add /dev/sdc3
mdadm: cannot get array info for /dev/md2

This probably happens because /dev/md2 is inactive. Does anybody have a valuable hint except rtfm?

Pearlseattle · 05-06-2017, 01:38 PM

Shooting in the dark:

Stop the raid....
mdadm --stop /dev/md2

...then explicitly get rid of that disk/partition...
mdadm --manage /dev/md2 --fail /dev/sdc3
mdadm --manage /dev/md2 --remove /dev/sdc3
mdadm --zero-superblock /dev/sdc3

...and then re-add that disk? (after restarting the raid - and maybe you might have to completely recreate the raid - which would mean to zero as well the superblock of the other disk)

Btw., doesn't Synology use proprietary raid?

D3smond · 05-07-2017, 02:30 AM

thanks for your answer.

Regarding the proprietary raid. Actually I'm not 100% sure, but I thought it should work as I found a synology tutorial, where they decribe to get it work in a linux live system with mdadm: https://www.synology.com/de-de/knowl...ion_using_a_PC

However, I'm not sure if removing the superblock would really help. If I undestand the setup correctly it seems like the superblock it self should be fine, since I still see all the information if I execute "mdadm --examine ..."

Code:

root@avira:~#  mdadm --examine /dev/sd[a-b]3
/dev/sda3:
          Magic : a92b4efc
        Version : 1.2
    Feature Map : 0x0
     Array UUID : 1476862c:52847f5b:60f1212a:cd870058
           Name : nas:2
  Creation Time : Mon Oct 22 16:15:45 2012
     Raid Level : linear
   Raid Devices : 2

 Avail Dev Size : 5851088833 (2790.02 GiB 2995.76 GB)
  Used Dev Size : 0
    Data Offset : 2048 sectors
   Super Offset : 8 sectors
          State : active
    Device UUID : 92e21fb2:3defd6a7:600e7750:4a7478f5

    Update Time : Wed Apr  5 19:36:58 2017
       Checksum : 13d9c309 - correct
         Events : 2

       Rounding : 64K

   Device Role : Active device 32769
   Array State : A. ('A' == active, '.' == missing)
/dev/sdb3:
          Magic : a92b4efc
        Version : 1.2
    Feature Map : 0x0
     Array UUID : 1476862c:52847f5b:60f1212a:cd870058
           Name : nas:2
  Creation Time : Mon Oct 22 16:15:45 2012
     Raid Level : linear
   Raid Devices : 2

 Avail Dev Size : 5851088833 (2790.02 GiB 2995.76 GB)
  Used Dev Size : 0
    Data Offset : 2048 sectors
   Super Offset : 8 sectors
          State : active
    Device UUID : 09a00aaa:86bf43db:5c424ec5:71c9c2f9

    Update Time : Wed Apr  5 19:36:58 2017
       Checksum : b852d9eb - correct
         Events : 2

       Rounding : 64K

   Device Role : Active device 0
   Array State : A. ('A' == active, '.' == missing)

And even the partiones seems to be fine, right?

Code:

root@avira:~# parted -l
Model: ATA WDC WD30EZRX-00M (scsi)
Disk /dev/sda: 3001GB
Sector size (logical/physical): 512B/4096B
Partition Table: gpt

Number  Start   End     Size    File system     Name  Flags
 1      131kB   2550MB  2550MB  ext4                  raid
 2      2550MB  4698MB  2147MB  linux-swap(v1)        raid
 3      4832MB  3001GB  2996GB                        raid


Model: ATA WDC WD30EZRX-00M (scsi)
Disk /dev/sdb: 3001GB
Sector size (logical/physical): 512B/4096B
Partition Table: gpt

Number  Start   End     Size    File system     Name  Flags
 1      131kB   2550MB  2550MB  ext4                  raid
 2      2550MB  4698MB  2147MB  linux-swap(v1)        raid
 3      4832MB  3001GB  2996GB                        raid

I appreciate your suggestion, but before I will remove the superblock, I would like to understand why the raid isn't starting. I mean, the old superblock seems to be fine (since I still see all information, right?). The partions are also listed in parted. Is there anything else what can prevent the startup?