Hi,

I tried to upgrade my Cache disks but it was throwing me errors during that process. So I figured I'd put the original back in (sdb) but I can tell its not actually working. After this, my Docker now says "Docker Service failed to start" which I'm guessing it got corrupted. Not sure if its relevant but one month ago I removed some old disk, 3,4,5. I've also tried the Plugin Mover Tuning but that didn't appear to help either.

I did make a copy of my AppData since I saw my backups of it were not running. But now moving just doesn't want to move anything off the cache. It ran for about 12 hours and didn't move anything off cache but the array got about 100GB written to it in that time with Docker and VMs off. Ideally, I wanted to try to fix the cache situation before trying to tackle the docker issue.

I get this error for Fix Common Problems but these are SSDs so not sure how to fix this or if I can ignore it.

Attached are the diagnostics. I rebooted, then start Mover and just noticed it filled up Disk 1 real fast which doesn't have a lot of space on it, then Disk 1 got its space back and activity went to a crawl speed of 1MBs with the Mover still running.

I have no idea how best to tackle this?

Can I get the 2nd Raid1 Cache disk fixed or move my data to the array to recreate the pool?

orthanc-diagnostics-20260208-0112.zip

Edited February 8Feb 8 by elementnak
Attachment

That pool is missing a device and also going read-only on mount.

Reimport the pool with just cache2 and post new diags after array start.

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 cache2 pool devices, leave the filesystem set to auto

start the array to import the pool

OK, removed the pool, created a new pool with 1 slot.

Added the device back.

Started the array and it added a slot back and says its missing.

Is it remembering the old pool configuration from the disk. So i switch it to slot 1.

Try to start the array and get an error.

Kinda stuck now.

orthanc-diagnostics-20260208-0940.zip

12 hours ago, JorgeB said:

That pool is missing a device and also going read-only on mount.

Reimport the pool with just cache2 and post new diags after array start.

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 cache2 pool devices, leave the filesystem set to auto

start the array to import the pool

It sounds like I have a similar situation as this post where you offered these steps.

I can start the array with the sdc in slot 2 of the cache and this is what I see in the array.

Would i do this command?

btrfs balance start -f -mconvert=single /mnt/cache
btrfs dev del devid:1 /mnt/cache

17 hours ago, elementnak said:

Is it remembering the old pool configuration from the disk. So i switch it to slot 1.

You can't do that before.

Post new diags with the array started first to confirm the pool status (reimport the pool with that single device once more if needed to start)

I've attached logs but I would think they are the same as in my second post. Array is started and cache looks like this.

Thanks

orthanc-diagnostics-20260209-0929.zip

With the array running type:

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

btrfs dev del missing /mnt/cache

After that, reimport the pool again with the single device.

Hmm, I got this running the first command. I'd take a guess to use 'btrfs device remove' instead of 'btrfs balance'

The balance must be finished for the removal to work; post the diags after the balance error.

orthanc-diagnostics-20260210-0736.zip

The pool has other issues; my suggestion would be to back it up, then reformat the pool and restore the data.