[SOLVED] Cache Unmountable: not mounted - Failing SSD?

[Dit]

Any help would be great. Installed a new shucked disk over the weekend, and when I went to boot up the server, the cache drive was stating "Unmountable: not mounted". After doing some research I decided it would be easiest just to Unraid reformat the disk and recover all my appdata through CA Backup / Restore Appdata. This resolved the issue, but now if I stop the array and reboot the server, the cache drive states again "Unmountable: not mounted". This can be resolved again by a reformat and recovering the app data. One other issue I noticed is that my Plex DB seems to be getting corrupted over time. At first I couldnt open the plex container GUI as it would say there was no connection. If i replaced the DB files with backups it would run fine and then slowly would not be able to watch further content or would lose access to the container again.

 

 

I have am wondering if this is a failing SSD? I have tried running a SMART test on the cache drive but it doesn't seem to be running. As soon click the start button it immediately switches to say stop and immediately switches back to start. I was able to download some previous SMART data report that is attached.

 

Sabrent_FA7807030D0701356802-20211006-2146.txt

https://forums.unraid.net/applications/core/interface/file/attachment.php?id=129223&key=a9dad88a97e6eccf274705b19e57cd76

[JorgeB]

Please post the diagnostics (after array start)

[Dit]

5 hours ago, JorgeB said:

Please post the diagnostics (after array start)

Please see attached. For reference the SSD is currently unmounted due to it stating unmountable. I can reformat it and mount and then supply diagnostics if needed.

 

Someone on the unraid subreddit where I posted this said it looks like the SSD may be failing as it looks like the error count is at 71K.

 

 

tower-diagnostics-20211007-1222.zip

[JorgeB]

Oct  7 06:49:37 Tower kernel: blk_update_request: critical medium error, dev nvme0n1, sector 5125600 op 0x0:(READ) flags 0x1000 phys_seg 1 prio class 0
Oct  7 06:49:37 Tower kernel: BTRFS: error (device nvme0n1p1) in btrfs_recover_log_trees:6250: errno=-5 IO failure (Couldn't read tree log root.)
Oct  7 06:49:37 Tower kernel: BTRFS: error (device nvme0n1p1) in btrfs_replay_log:2279: errno=-5 IO failure (Failed to recover log tree)

 

1st one is a hardware error, so yes, it looks like a device problem.

[Dit]

 

1 hour ago, JorgeB said:

Oct  7 06:49:37 Tower kernel: blk_update_request: critical medium error, dev nvme0n1, sector 5125600 op 0x0:(READ) flags 0x1000 phys_seg 1 prio class 0
Oct  7 06:49:37 Tower kernel: BTRFS: error (device nvme0n1p1) in btrfs_recover_log_trees:6250: errno=-5 IO failure (Couldn't read tree log root.)
Oct  7 06:49:37 Tower kernel: BTRFS: error (device nvme0n1p1) in btrfs_replay_log:2279: errno=-5 IO failure (Failed to recover log tree)

 

1st one is a hardware error, so yes, it looks like a device problem.

I'm assuming the only thing that can throw that error is the drive itself and not anything to do with the Mobo?

[JorgeB]

Yes, I believe the board could at most drop the device, not generate a sector read error.

[Dit]

Just wanted to update this that the new SSD resolved the issue and haven't had any problems since i replaced it.