My cache drive became unmountable overnight, showing Unmountable: Unsupported or no file system.  I tried rebooting but that didn't make a difference.  Following the manual I was going to try to scrub the drive as it is BTRFS but it says Scrub is only available when array is Started even though the array is started.  Any thoughts on the next steps that I should try to remount the cache drive?  I've attached my most recent diagnostics.

tower-diagnostics-20231024-1017.zip

If the log tree is the only problem this may help:

 

btrfs rescue zero-log /dev/sde1

 

Then restart the array.

 

1 hour ago, faulty_lamp said:

it says Scrub is only available when array is Started

Scrub is only available for a mounted fileysstem, it won't help in this case

That worked, thank you!

20 minutes ago, JorgeB said:

If the log tree is the only problem this may help:

 

btrfs rescue zero-log /dev/sde1

 

Then restart the array.

 

Scrub is only available for a mounted fileysstem, it won't help in this case

 

Actually, now it looks like the cache is mounted as read-only (Drive mounted read-only or completely full).  The drive is less than 50% full.  I've attached the newest diagnostics, is there anything in that or would the next step be to run a memtest?

tower-diagnostics-20231024-1150.zip

Looks like there are more issues with the filesystem:

 

Oct 24 11:40:59 Tower kernel: BTRFS warning (device sde1): checksum verify failed on logical 188203008 mirror 1 wanted 0x8270c9db found 0x100372cb level 0
Oct 24 11:40:59 Tower kernel: BTRFS warning (device sde1): checksum verify failed on logical 188203008 mirror 2 wanted 0x8270c9db found 0x100372cb level 0
Oct 24 11:40:59 Tower kernel: BTRFS error (device sde1): failed to run delayed ref for logical 246287810560 num_bytes 8192 type 178 action 1 ref_mod 1: -5
Oct 24 11:40:59 Tower kernel: BTRFS: error (device sde1: state A) in btrfs_run_delayed_refs:2144: errno=-5 IO failure
Oct 24 11:40:59 Tower kernel: BTRFS info (device sde1: state EA): forced readonly

 

I would recommend backup and format, and yes, running memtest before would be a good idea.

3 hours ago, JorgeB said:

Looks like there are more issues with the filesystem:

 

Oct 24 11:40:59 Tower kernel: BTRFS warning (device sde1): checksum verify failed on logical 188203008 mirror 1 wanted 0x8270c9db found 0x100372cb level 0
Oct 24 11:40:59 Tower kernel: BTRFS warning (device sde1): checksum verify failed on logical 188203008 mirror 2 wanted 0x8270c9db found 0x100372cb level 0
Oct 24 11:40:59 Tower kernel: BTRFS error (device sde1): failed to run delayed ref for logical 246287810560 num_bytes 8192 type 178 action 1 ref_mod 1: -5
Oct 24 11:40:59 Tower kernel: BTRFS: error (device sde1: state A) in btrfs_run_delayed_refs:2144: errno=-5 IO failure
Oct 24 11:40:59 Tower kernel: BTRFS info (device sde1: state EA): forced readonly

 

I would recommend backup and format, and yes, running memtest before would be a good idea.

 

I've backed the drive up, but have not formatted yet.  Went ahead and started memtest and it does look like there are some errors already.  I'm assuming I'll just need to remove all memory sticks and place one at a time and test again to see which one has errors?

 

After removing the erroring stick should I still format the drive?

10 hours ago, faulty_lamp said:

I'm assuming I'll just need to remove all memory sticks and place one at a time and test again to see which one has errors?

Yep, see if you can find the culprit stick

 

10 hours ago, faulty_lamp said:

After removing the erroring stick should I still format the drive?

I would still recommend formatting, since it's likely already damaged beyond repair.