Cache drive "Unavailable: Wrong or no filesystem" error

[iotarho]

Unraid was running solid for several months while I was away on travel.  Came back to a few VMs needing updated, and noticed one of the VMs wouldn't come back up after I updated.  After a bit of troubleshooting, rebooted my Unraid server and came back up to the Cache drive "Unmountable: Wrong or no filesystem" error.  Other drives all seem to be intact.  Attaching diagnostics.

 

 

tower-diagnostics-20230820-1607.zip

[JorgeB]

Post the output of:

xfs_repair -v /dev/sdf1

 

[iotarho]

Thanks for the assist.  Took about 90 minutes, here's the result of that command:

 

Phase 1 - find and verify superblock...
bad primary superblock - bad magic number !!!

attempting to find secondary superblock...
................................................................................ ##repeated a few thousand times

Sorry, could not find valid secondary superblock
Exiting now.

[JorgeB]

You cache has the filesystem set to xfs, but there's no xfs filesystem there, looks like it's btrfs:

Aug 20 16:00:45 Tower kernel: BTRFS: device fsid 965f0868-d1ec-428d-ba39-03557435741d devid 1 transid 463170 /dev/sdf1 scanned by udevd (1123)

Stop the array, click on cache, change the fs to btrfs and start the array, I don't see how that could have changed on its own.

[iotarho]

Okay went in and double checked, "File system type" is listed as btrfs, not xfs, in the GUI.  Just to be sure, I reselected btrfs and hit "Done" (Apply was grayed out), then took the array offline and did a reboot, still listed as btrfs when everything boots back up.  Still the same "Unavailable" error.

[JorgeB]

Post new diags.

[iotarho]

Attaching new diagnostics here

tower-diagnostics-20230823-1040.zip

[JorgeB]

Now it's trying to mount btrfs, but looks like a log tree problem, but there are many ATA errors before with the cache device, replace cables (both power and SATA) and post new diags, if the errors are gone and it still doesn't mount we can try to fix the log tree.

[iotarho]

Okay, sorry so slow, took a few days to get new SATA power and data cables here!

 

Still visually the same error on bootup, attaching the latest diagnostics here.

 

 

tower-diagnostics-20230826-1543.zip

[JorgeB]

Still the same ATA errors, if you replaced the cables the device may be failing, if you have a spare see if you can clone it with ddrescue.

[iotarho]

Thanks.. just have the 512MB drive that the 1TB replaced.  However I'll get a new 1TB asap since it sounds like this one might be on the way out.  Will post a followup here asap.

[iotarho]

One quick followup JorgeB... I pulled out my previous (512MB) cache drive and will try to install it to see if it still has backups on it (I upgraded the 512MB cache drive to this 1TB WD Blue that you've been helping troubleshoot... the 512 worked fine but I wanted to get a larger cache drive, so in theory still has intact backups).

 

In the process I installed all new SATA power and data cables.  The 1TB is still not mountable, but I wanted to see if you'd  look at the latest diagnostics (attached) to confirm you're indeed still seeing the ATA errors suggesting an imminent drive failure?

 

Thank you

 

tower-diagnostics-20230830-1129.zip

[JorgeB]

10 hours ago, iotarho said:

to confirm you're indeed still seeing the ATA errors

Still there, if the cables have been replaced it's likely a device problem.