I'm in the process of swapping out my cache drive for another.

 

I've got a share "appdata" (which has the docker.img file) and I have used mc to copy it to disk1 on the array. I then assigned the new cache drive, formatted it and copied the data back in.

 

I'm seeing "Your existing Docker image file needs to be recreated due to an issue from an earlier beta of unRAID 6. Failure to do so may result in your docker image suffering corruption at a later time. Please do this NOW!" within the docker settings.

 

I realize that it has to do with the NOCOW attr, so I tried

 

chattr +C docker.img

 

with no luck.

 

Btrfs knowledge is lacking at this point, was wondering what must be done? Or was copying data from the orignal cache drive to the arry and back to a new cache not correct?

 

Must I now create a new image file and loop mount the existing one and copy the data to the new image nocow file?

 

Create new docker.img:

 

touch docker.img
chattr +C docker.img
fallocate -l10g docker.img

 

mount -o loop docker_old.img /mnt/docker_image_old/

mount mount -t btrfs -o loop docker.img /mnt/docker_image_new/

You can just create a new one and install your dockers again from your templates. Your templates have all the configuration needed to get everything setup the way it was and resume. The appdata should be fine since it is stored outside docker anyway.

Is still recreating the docker image the only solution to a say "rsync -a /mnt/cache/ mnt/user/backup/cache/" and return?