JorgeB

It's not mounting read only, you need to enable that in UD for that disk.

Cannot explain the discrepancy, but you need to run a correcting check.

There won't be anything visible, but without doing that first the disk would not mount.

Dec 16 17:53:58 Minilith kernel: critical medium error, dev nvme0n1, sector 0 op 0x1:(WRITE) flags 0x800 phys_seg 0 prio class 0 Dec 16 17:53:58 Minilith kernel: BTRFS error (device nvme0n1p1): bdev /dev/nvme0n1p1 errs: wr 0, rd 0, flush 1, corrupt 0, gen 0 Dec 16 17:53:58 Minilith kernel: BTRFS warning (device nvme0n1p1): chunk 1048576 missing 1 devices, max tolerance is 0 for writable mount Device problem, and be because it cannot write to it btrfs cannot mount read/write, you can try mounting read-only with the UD plugin to see if the problem is with writes only, if yes that would allow for data recovery.

Yes, you can change it by clicking on it. Yes, as long as there's enough space. Don't forget that you need to change XFS UUID first, UD settings.

Yes, thanks, I was thinking of rsync, the command you had won't work.

Are you sure you didn't forget the trailing slash?

Yes. Yes. Depends on the options used, but in the case you want that, you could use for example: rsync -av --delete /mnt/disks/UD_name/ /mnt/disk3/ --delete will delete any files on dest that don't exist on source, so need to be careful with that (it will leave the empty folders like lost+found but you can then delete that), and it will copy any missing files.

Yes, you can also just mount the old disk with UD (after changing the UUID) and run rsync to make both the same.

Close the terminal window

No issues for me:

Sorry, I meant without -n.

Also note that not sure Krusader shows the correct stats for btrfs, go to shares and click on compute to see if it's the same.

Just the docker image should not account for that, unless it's a large image, could be appdata, only way to know is to move elsewhere and then move back, docker image is easy to just recreate.

Run it again with -n, then look for a lost+found folder.

Check filesystem on disk3

Any vdisks or other image type files there can grow with time if not trimmed/unmapped, do you have those type of files there?

The set timeout (60 secs) is not enough, you can add 30 secs or so, or better yet stop the array and see how long it takes, then add 30 secs to that.

Then just use the GUI balance option, since 99% of the chunks are raid0 not much point in using the soft flag.

Nothings obvious, delete the share using the disk path then reboot.

Now these could be interesting, they've recently announced Gen2 with 16 and 18TB models, and now there will be a SATA version, which unlike the SAS models will present a single volume to the OS, problem will likely be the price:

This pool has both raid0 and raid1 chunks, I assume you want raid0 since most are, if yes type in the console: btrfs balance start -dconvert=raid0,soft /mnt/cache_download_one

Please post the diagnostics.

Please use the existing plugin support thread:

This sometimes happens, specially with WD drives, they are "false positives", a full disk write should get rid of them if you want.