Started having issues today where the mover was not moving anything from the cache pool to the array so i stopped the array and put it in maintenance mode then ran the btrfs check --readonly option after clicking on the disk in the unraid gui. It had some info in there but i accidentally closed it but the gist of it was something like this:

 

block group 422051840 has wrong amount of free space

failed to load free space cache for block group 422051840

In the log when the mover is running i see this

 

May 1 16:55:11 Storage shfs/user: err: shfs_write: write: (28) No space left on device
May 1 16:55:11 Storage kernel: loop: Write error at byte offset 7035449344, length 4096.
May 1 16:55:11 Storage kernel: blk_update_request: I/O error, dev loop0, sector 13741112
May 1 16:55:11 Storage kernel: BTRFS error (device loop0): bdev /dev/loop0 errs: wr 495, rd 0, flush 0, corrupt 0, gen 0
May 1 16:56:36 Storage shfs/user: err: shfs_write: write: (28) No space left on device

 

I have ample space across all drives so i know im not out of space. I did some searching for the "wrong amount of free space" error above and saw some stuff referencing a clear_cache flag but nothing here on the lime tech forums. Any ideas on what to do here?

 

Diagnostics are attached

 

storage-diagnostics-20170501-1655.zip

Errors start immediately after boot, so something happened before that we can't see in the current logs, there are also write errors on both the docker and libvirt images.

 

I would copy everything important from the cache pool and then re-format it.

 

Also post the output of:

 

btrfs dev stats /mnt/cache

This can give an idea if there were previous pool errors.

Here are the results

 

root@Storage:~# btrfs dev stats /mnt/cache
[/dev/sdf1].write_io_errs   0
[/dev/sdf1].read_io_errs    0
[/dev/sdf1].flush_io_errs   0
[/dev/sdf1].corruption_errs 0
[/dev/sdf1].generation_errs 0
[/dev/sdk1].write_io_errs   0
[/dev/sdk1].read_io_errs    0
[/dev/sdk1].flush_io_errs   0
[/dev/sdk1].corruption_errs 0
[/dev/sdk1].generation_errs 0
 

No errors so hardware wise all looks fine, you should be OK after re-formatting.