Hi all, recently my NAS was showing "degrading" on the front panel. I logged in to the NAS and found out it was Disk 1 on the way out so I replaced it with a larger 4Tb disk (from 3TB) and after it finished going through the motions, I logged in to have a look and now the volume has disappeared and an error saying "Remove inactive volumes to use the disk. Disk #2,3,4. The 3 drives are also shaded in red but not showing any errors on them. Any help would be much appreciated
Hi all, recently my NAS was showing "degrading" on the front panel. I logged in to the NAS and found out it was Disk 1 on the way out so I replaced it with a larger 4Tb disk (from 3TB) and after it finished going through the motions, I logged in to have a look and now the volume has disappeared and an error saying "Remove inactive volumes to use the disk. Disk #2,3,4. The 3 drives are also shaded in red but not showing any errors on them. Any help would be much appreciated
Are you certain you replaced the correct disk?
Try downloading the full log zip file, and look in there for clues.
Hi, while the NAS was on, I removed the 3TB drive from #1 and replaced it with a 4TB drive, both are WD red drives. The NAS went through rebuilding step (I think that's what it is called), once it was finished I noticed I couldn't access any of the files. Thats when I went looking and found the volume missing. I have also put the 3TB disk back in the NAS for now, just incase it was the 4TB drive playing up. Is it ok to send you the logs as I don't know what I would be looking at anyway.
Is it ok to send you the logs as I don't know what I would be looking at anyway.
I can take a look. Put the entire log zip into cloud storage (dropbox, google drive, icloud, etc). Then send me a PM using the envelope icon in the upper right of the forum page. Put the download link in the PM
sda (first slot) has definitely failed - kernel.log is loaded with errors
Mar 04 20:18:38 JONES kernel: sd 0:0:0:0: [sda] tag#11 Add. Sense: Unrecovered read error - auto reallocate failed
Mar 04 20:18:38 JONES kernel: blk_update_request: I/O error, dev sda, sector 9437257
sdb (second slot) is out-of-sync with the others. Is this the old disk you put back?
Mar 04 20:15:36 JONES kernel: md: kicking non-fresh sdb3 from array!
Mar 04 20:15:36 JONES kernel: md/raid:md127: device sdd3 operational as raid disk 2
Mar 04 20:15:36 JONES kernel: md/raid:md127: device sdc3 operational as raid disk 3
Mar 04 20:15:36 JONES kernel: md/raid:md127: not enough operational devices (2/4 failed)
That combination (sda failed and sdb out of sync) is why the array didn't mount.
There is also some evidence that the resync (called a resilver in volume.log) did not complete
=== maintenance history === device operation start_time end_time result details data disk test 2023-02-18 08:26:11 data disk test 2023-02-18 13:08:32 2023-02-18 13:08:42 pass data resilver 2023-03-03 08:21:29
Note that the resilver lists a start time, but not a completion time.
As far as what to do -
If you have a backup, then replacing disk 1 and doing a factory reset is the best path.
If you don't, it might be possible to forcibly add sdb to the array. Depending on how old it is, there could be some data loss/corrupted files.
Slot 1 is the disk that I replaced initially but after the volume was not showing I took the new disk out and replaced it with the original one. I can put the new disk back in and upload the logs again if that helps
@tez7171 wrote: Slot 1 is the disk that I replaced initially but after the volume was not showing I took the new disk out and replaced it with the original one. I can put the new disk back in and upload the logs again if that helps
Putting the new disk back in won't help if the resync didn't actually complete. Though you might as well remove the old failed disk.
Do you have a backup?
If not, do you want to try forcibly mounting the volume? That would require use of the linux command line - it'd be useful to know if you have experience with linux.
I would like to try and mount the volume if possible, I didn't do a backup before I added the new disk unfortunately. I haven't had any experience with Linux. I have put the new disk back in
This is risky, and you are on your own if you attempt this. Other options you have are to
use data recovery software (connecting the disks to a Windows PC)
get a data recovery contract with Netgear
use some other data recovery service
These options have less risk, but will cost.
(a) ReclaiMe data recovery software sells for about $200, and you'd likely also need 4-disk enclosure for 4 disks- at least $100 more. Plus you'd need storage to off-load the data.
When you connect to the NAS, log in as root. The password is infr8ntdebug.
Once logged in, enter these commands. (Don’t enter the # part, that is just the prompt the NAS will give you).
# rnutil chroot
# mdadm --assemble --really-force /dev/md127 /dev/sd[abcd]3
# btrfs device scan
# btrfs fi show
# mount /dev/md127 /data
# ls /data
If you get an error on any of these, post back what you are seeing on the screen before proceeding.
If this works, the last command will give you a listing of the folders in your data volume (e.g., the shares). If you see that, then power down the NAS and then power it up normally.
Update. I installed both reclaime free raid and trial version of reclaime ultimate. I made images of all 4 disks using reclaime free raid, after that I ran it through reclaime ultimate and was able to get most of the files back. A little pricey especially with the exchange rate the way it is. Lesson learned, make sure you back up then before attempting a disk swap, back up again
@tez7171 wrote: Update. I installed both reclaime free raid and trial version of reclaime ultimate. I made images of all 4 disks using reclaime free raid, after that I ran it through reclaime ultimate and was able to get most of the files back.
Thanks for updating us, and I'm glad you got most of the data back.
@tez7171 wrote: A little pricey especially with the exchange rate the way it is. Lesson learned, make sure you back up then before attempting a disk swap, back up again
Yeah. Backup is a better path - often less expensive, and certainly less stressful.
