I've been battling this for going on 3 days now and have made no progress. I was able to plug one of my cache drives into another machine and make a clone image of it (now i have to go through hell to mount it), but i cannot for the life of me figure out why this happened. I woke up to a frozen Unraid box, rebooted, and when it came back up, all my VM's were mysteriously missing and my Cache drives were labeled as unmountable. Any ideas? If i lose all that data, I'm gonna freak.

 

 

 

Feb  1 07:42:00 HomeArray emhttp: shcmd (82): mkdir -p /mnt/cache

Feb  1 07:42:00 HomeArray emhttp: shcmd (83): set -o pipefail ; mount -t btrfs -o noatime,nodiratime -U 405f75bc-aafa-4c5d-ab98-0b49fdf02a45 /mnt/cache |& logger

Feb  1 07:42:00 HomeArray kernel: BTRFS info (device sdl1): disk space caching is enabled

Feb  1 07:42:00 HomeArray kernel: BTRFS: has skinny extents

Feb  1 07:42:04 HomeArray kernel: BTRFS error (device sdl1): parent transid verify failed on 475956133888 wanted 531255 found 531254

Feb  1 07:42:04 HomeArray kernel: BTRFS error (device sdl1): parent transid verify failed on 475956133888 wanted 531255 found 531254

Feb  1 07:42:04 HomeArray kernel: BTRFS warning (device sdl1): failed to read log tree

Feb  1 07:42:04 HomeArray root: mount: /dev/sdl1: can't read superblock

Feb  1 07:42:04 HomeArray emhttp: err: shcmd: shcmd (83): exit status: 32

Feb  1 07:42:04 HomeArray emhttp: mount error: No file system (32)

Feb  1 07:42:04 HomeArray emhttp: shcmd (84): umount /mnt/cache |& logger

Feb  1 07:42:04 HomeArray kernel: BTRFS: open_ctree failed

Feb  1 07:42:04 HomeArray root: umount: /mnt/cache: not mounted

Feb  1 07:42:04 HomeArray emhttp: shcmd (85): rmdir /mnt/cache

Feb  1 07:42:04 HomeArray emhttp: shcmd (86): sync

Feb  1 07:42:04 HomeArray emhttp: shcmd (87): mkdir /mnt/user0

Feb  1 07:42:04 HomeArray emhttp: shcmd (88): /usr/local/sbin/shfs /mnt/user0 -disks 510 -o noatime,big_writes,allow_other,use_ino  |& logger

Feb  1 07:42:04 HomeArray emhttp: shcmd (89): mkdir /mnt/user

Feb  1 07:42:04 HomeArray emhttp: shcmd (90): /usr/local/sbin/shfs /mnt/user -disks 511 2048000000 -o noatime,big_writes,allow_other,use_ino  -o remember=0  |& logger

Feb  1 07:42:04 HomeArray emhttp: shcmd (91): cat - > /boot/config/plugins/dynamix/mover.cron <<< "# Generated mover schedule:#0120 */1 * * * /usr/local/sbin/mover |& logger#012"

Feb  1 07:42:04 HomeArray emhttp: shcmd (92): /usr/local/sbin/update_cron &> /dev/null

Feb  1 07:42:04 HomeArray emhttp: Starting services...

Feb  1 07:42:04 HomeArray avahi-daemon[2733]: Files changed, reloading.

Feb  1 07:42:04 HomeArray avahi-daemon[2733]: Service group file /services/afp.service changed, reloading.

Feb  1 07:42:04 HomeArray emhttp: shcmd (105): set -o pipefail ; /usr/local/sbin/mount_image '/mnt/user/Dockers/docker.img' /var/lib/docker 64 |& logger

Feb  1 07:42:04 HomeArray kernel: BTRFS: device fsid d0b8624c-f5c8-480a-8f73-39dd8532b60f devid 1 transid 100 /dev/loop0Feb  1 07:42:04 HomeArray kernel: BTRFS info (device loop0): disk space caching is enabled

Feb  1 07:42:04 HomeArray kernel: BTRFS: has skinny extents

Feb  1 07:42:04 HomeArray root: Resize '/var/lib/docker' of 'max'

Feb  1 07:42:04 HomeArray kernel: BTRFS info (device loop0): new size for /dev/loop0 is 68719476736

Try btrfs restore, make a folder on an array disk or unassigned device, eg:

 

btrfs restore -v /dev/sdl1 /mnt/disk2/restore

 

If you were using raid1 better restore both devices to different folders, one of them may have more recent data.

EDIT:

Manually mkdir'd the restore folder and it seems to be running now. Getting a lot of offsetting, but it's doing it's thing.

 

 

Getting these errors from that command:

 

root@HomeArray:/mnt# btrfs restore -v /dev/sdl1 /mnt/disk3/restore

parent transid verify failed on 475956133888 wanted 531255 found 531254

parent transid verify failed on 475956133888 wanted 531255 found 531254

parent transid verify failed on 475956133888 wanted 531255 found 531254

parent transid verify failed on 475956133888 wanted 531255 found 531254

Ignoring transid failure

Restoring /mnt/disk3/restore/system

Error mkdiring /mnt/disk3/restore/system: 2

you need to create the destination folder beforehand, eg:

 

mkdir /mnt/disk2/restore

Thanks i realized that right after i made that comment. Went ahead and dropped an edit in.

 

It seems like it recovered a ton of data on both restores. Now my next question is, how do i get the VM's back? Do i need to re-create each VM and just point it at the vdisk.img files?

If the libvirt.img is ok you cant point to it (settings -> vm manager) and edit the vms to point to the vdisk location and they should start, if you don't have the old libvirt.img or it's corrupt you need to recreate the VMs and point to the vdisks.

Sorry to revive this thread but I'm having the same issue.  Moved my libvirt.img off the cache and can't start up vms anymore.  Seems to be too far gone based on the commands in this thread.  I created a new img file and set up my VM again using the existing disks however now when Windows 10 starts it fails with a System_License_Violation and won't boot at all.

 

Any idea how to get past this?  It must think it's a totally different machine.

 

Also, the VM was originally built in unraid 6.3.5 and I'm now doing this work in the latest RC for 6.4  Perhaps there is a hardware setting change I need to make so it'll boot?

 

Thanks