TS420 RAID5 2nd disc write error while rebuilding

[momoxia]

Dear all,

I am having trouble with my RAID 5 that I have been running on my TS 420 since 2015.
After a first HD (slot4) showed severe errors in reading/writing I replaced it according to the instructions in the manual and rebuilding started flawlessly.
The rebuilding process ran for a while...overnight...and in the morning I recognized that it could not finish due to some error in reading of a second HD (slot2)

As the original HD was still "kind of working" my hope was then to rebuild first the secondly failed HD and then the initial erroneous HD...but that wouldn't work out either.
I had the raid running in degraded mode at a certain moment, but after doing all that swapping moves...it never came back.
So read a bit through the forum here, found some useful hints on inspecting the RAID status via SSH login to the NAS.

From what "I can understand" mdadm claims that the RAID should be active and running, while I can't see that on the web interface (it says unmounted RAID5 ).
Can anyone help me to fix it, so that I can see as well a mounted and working RAID via the web interface?

Please find info regarding my NAS below:

NAS: TS 420
RAID: RAID5, 4 slot, 4HD @3TB
Firmware/Buildnr: 4.3.3.1624
Network: single Port

Detailed HDD Information:
/dev/sda Model=ST3000VN000-1HJ166 , FwRev=SC60 , SerialNo= W6A0322T
/dev/sdb Model=ST3000VN000-1HJ166 , FwRev=SC60 , SerialNo= W7302KYE
/dev/sdc Model=WDC WD30EFRX-68EUZN0 , FwRev=82.00A82, SerialNo= WD-WCC4N4KVC590
/dev/sdd Model=WDC WD30EFRX-68EUZN0 , FwRev=82.00A82, SerialNo= WD-WCC4N4KVC77J

[dolbyman]

You should start the NAS fresh and restore from backups .. if you do not have backups, bring the NAS to a data rescue company and they might be able to rebuild it (it will cost you big time)

[momoxia]

I would do a fresh restore but I do not have a very "recent" backup, therefore I would like to try at least to recover the existing RAID.

What Bugs me is that if I log on via SSH the RAID is active and all disks are running...

Code: Select all

cat /proc/mdstat 
Personalities : [raid1] [linear] [raid0] [raid10] [raid6] [raid5] [raid4] 
md0 : active raid5 sdd3[3] sda3[0] sdc3[2] sdb3[1]
		 8786092608 blocks super 1.0 level 5, 64k chunk, algorithm 2 [4/4] [UUUU]
		 
md4 : active raid1 sdd2[4](S) sdc2[3](S) sdb2[2] sda2[0]
		 530128 blocks super 1.0 [2/2] [UU]
		 
md13 : active raid1 sdb4[0] sdc4[3] sdd4[2] sda4[1]
		 458880 blocks [4/4] [UUUU]
		 bitmap: 0/57 pages [0KB], 4KB chunk

md9 : active raid1 sda1[0] sdd1[3] sdc1[2] sdb1[1]
		 530048 blocks [4/4] [UUUU]
		 bitmap: 0/65 pages [0KB], 4KB chunk

Code: Select all

mdadm --misc -D /dev/md0                                                                                        
/dev/md0:
        Version : 01.00.03
  Creation Time : Sun Jan 11 22:32:08 2015
     Raid Level : raid5
     Array Size : 8786092608 (8379.07 GiB 8996.96 GB)
  Used Dev Size : 2928697536 (2793.02 GiB 2998.99 GB)
   Raid Devices : 4
  Total Devices : 4
Preferred Minor : 0
    Persistence : Superblock is persistent

    Update Time : Sun Jun 13 13:04:32 2021
          State : clean
 Active Devices : 4
Working Devices : 4
 Failed Devices : 0
  Spare Devices : 0

         Layout : left-symmetric
     Chunk Size : 64K

           Name : 0
           UUID : e32993b6:318d3c1f:4cb68be9:27d522df
         Events : 22841

    Number   Major   Minor   RaidDevice State
       0       8        3        0      active sync   /dev/sda3
       1       8       19        1      active sync   /dev/sdb3
       2       8       35        2      active sync   /dev/sdc3
       3       8       51        3      active sync   /dev/sdd3

[holger_kuehn]

The RAID can not be rebuild, if there is a URE on one of the drives you have the read from.

If you are lucky you can try to read the data from your degraded RAID. This should fail on the sectors that have the URE, but you might be able to get the rest of data off. So maybe only a single file gets lost.

This is actually what RAID-scrubbing is for, to find those UREs before you need the sectors for recovery.