I have my cache pool set as degraded after an unclean shutdown. Both drives are shown as fully healthy and have no errors. But nvme1n1 lost the partition completely. The data is still there on the first drive. What is the best way to restore the pool?

unraid-diagnostics-20260612-1851.zip

Please post new diags with the array started.

2 hours ago, JorgeB said:

Please post new diags with the array started.

unraid-diagnostics-20260612-2256.zip

Pool has some single profile chunks, so it's not possible to just re-add the missing device. You can do this:

• with the array running type

  btrfs balance start -f -dconvert=single,soft -mconvert=single,soft /mnt/cache

• once that's done type

  btrfs device remove missing /mnt/cache

• stop the array, and reimport he pool wit the singe device

  on main click on the first device for that pool and then "remove pool"

  back on main, create a new pool with the same name and 1 slot

  assign the good pool device, leave the filesystem set to auto

  start the array to import the pool

• once the pool is imported as single, stop the array

• change the slots to 2

• assign the other device

• start array to begin balance

So this has happened again, the cache is degraded. The system was fully unresponsive and I had to do an unclean reset. How can I diagnose what's causing it and prevent this from happening again. It might be related to my other thread, but if it's ssd failing why all SMART is good? https://forums.unraid.net/topic/199527-btrfs-mount-hang/

Do I repeat the same steps from above or should I wait?

unraid-diagnostics-20260708-1936.zip unraid-diagnostics-20260708-1932.zip

The pool device is already missing in both diags posted, do you have the diags from when the device dropped? Those could help see what's happening.

10 minutes ago, JorgeB said:

The pool device is already missing in both diags posted, do you have the diags from when the device dropped? Those could help see what's happening.

No, I came home to an unresponsive machine and after a hard reset it was like this already.

In that case you should be able to recover the pool the same as before, then recommend setting up the syslog server to have the syslog if it happens again.

On 7/8/2026 at 9:13 PM, JorgeB said:

In that case you should be able to recover the pool the same as before, then recommend setting up the syslog server to have the syslog if it happens again.

So it (unresponsive system) happened again after I set up syslog server, and the logs show nothing when it died. This time, however, btrfs mount was not degraded, but a scrub showed 113 124 errors.

syslog-192.168.1.2.log

btrfs is detecting data corruption; recommend running memtest