you could also try to automate this using the API. Off the top of my head I could imagine:
Get all VMs through /cluster/resources
Manually run the backup on all instances which have the "running" status
this might be a bit racy though, since VMs could change their state in the time between querying and actually doing the backup. Otherwise, you could also just query the status of each VM right before you do the backup (/nodes/{node}/qemu/{vmid}/status/current).
As an alternative, create a pool "do_backup_this" and schedule the backup task to use that pool. Then change the pool membership to include/exclude the VM's that must be copied.
Check the bug reports that Neobin linked above as they expose a few corner cases were you might get changes in your VMs that won't be backed up.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.