BTRFS volume keeps dropping to read-only


I’m running 3.9.2-24 in a KVM virtual machine, and am having problems with a volume. I shutdown the Rockstor guest and checked the host filesystems containing the qcow2 files with no errors.

However, I’m still seeing errors on the Rockstor system. From dmesg:

I tried doing a btrfs scrub command, but it exits right away.
scrub status for e6e5e639-b75e-40d1-8355-ce9c2db9763f
scrub started at Thu May 31 15:45:55 2018 and was aborted after 00:00:00
total bytes scrubbed: 0.00B with 0 errors

Thanks for any help,

Looks like your FS tree is screwed.

Noting that BTRFS recovery is a bit of a shitstorm, don’t blame me if the following fails, destroys data, or mutilates your favourite pet:

# btrfs-find-root <</path/to/disk>>
Super think's the tree root is at <<ADDRESS>>, chunk root 20971520
Found tree root at <<ADDRESS>>

# btrfs-restore -t <<ADDRESS>>

Thank you for your reply.

I got lucky. I was able to boot single user, and did a
btrfs clean /dev/sdb
and then
btrfs clean --repair /dev/sdb

I was able to mount it and it was all there. Lucky indeed! Though, I did have backups.