Hi. I’d like to copy or move an incus container instance to another one through the network. When I ran the command:
sudo incus copy --refresh --allow-inconsistent mycontainername myremotebackup
it started perfectly, but when it found a I/O in the disk the rsync got stucked and the container has never been created on destination.
Is there a way to run a custom rsync to skip failing bad blocks?
I tried to export to a file into a external disk, but it failed too.
What are your suggestions?
I run the incus version 6.6, Debian 12, the storage driver is “dir”. Thanks a lot.
You can probably try to get as much data as you can direcrtly from /var/lib/incus/containers/mycontainername/. If things get stuck in I/O wait rather than just failing, that may be a bit difficult though as anything trying to rsync from it will get similarly stuck.
Thanks Stéphane,
As I 'm new in incus, let me ask another question: how could I create a new instance from the recovery data? Incus has a “create” function from images and “import” from tarball backups. Is there a manual process?? Thanks again.
If you just have the rootfs, you could use incus-migrate to transfer that filesystem as a new instance. If you have the whole folder, you could restore the folder in the same spot and then use incus admin recover to scan the storage pool, detect that instance and re-create the DB entries from it.