Hi, I had a cache drive fail (device 2) so I replaced it with a spare I had. After replacing, it's now showing degraded. I found several posts with similar issues as mine and it was suggested to follow the commands below. Before I run these commands, can someone please confirm if this is the correct process given my situation? Also, is this process safe for the cache pool data? I am backing up my cache data to my non-cache disks as I type, but just want to understand the commands and process.

https://forums.unraid.net/topic/200867-cache-degraded-1-drive-not-working/

Then you can fix the pool. With the array running, type:

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

when that's done type:

btrfs device remove missing /mnt/cache

Then stop the array and re-import the pool with the remaining 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 ADATA NVMe, leave the filesystem set to auto

- start the array to import the pool

Edited Monday at 04:41 AM4 days by TacomaTy

Please post the diagnostics.

@JorgeB sorry about that - I downloaded them but forgot to attach. Here they are.

tower-diagnostics-20261108-2105.zip

The commands from that other thread won't work here. Unraid already ran them automatically when you first started the array with the pool set to 1 slot. The balance failed with "enospc errors during balance" because the remaining device is completely full: about 1.05TiB of data on a 1.09TiB device, with no unallocated space left to convert the data.

The spare also wasn't used when you assigned it. It was dropped from the pool at array start, most likely because it's smaller than the device it's replacing. The failed device was a 2TB Intel and the spare is 1.2TB, and a pool replacement device must be the same size or larger.

The good news is that all your data is on the remaining device. The pool is just running without redundancy, and the two scrubs you ran found no errors.

Since you're already backing up the pool, the simplest and safest option is to recreate it with both 1.2TB devices:

1. Go to Settings and disable the Docker and VM services, so nothing on the pool is in use. Then finish the backup and confirm everything is there, including appdata, system and domains.

2. Stop the array, click the first pool device and use "Remove pool".

3. With the old pool device now unassigned, use Unassigned Devices to remove its partitions, like you did with the spare. Only do this after confirming the backup, because this erases the old pool.

4. Create a new pool named cache with 2 slots and assign both devices. Click the first device and set the file system to "btrfs - encrypted". Start the array, check the format box and format the pool.

5. Copy the data back, then re-enable Docker and VMs.

Note that the new pool will have about the same usable space as before, about 1.2TB, because a RAID1 pool is limited by its smallest device. It was 97% full, so I would recommend keeping less data on it, since btrfs doesn't behave well when nearly full.