Hi, i test a deployment of PVE and i have a few questions.
Question 1:
I see that the backups are named vzdump-qemu-VMID-date-time.vma.gz (for gzip format). This naming scheme is problematic because PVE tends to "fill the gaps" when vms are deleted.
If we have a vm, back it up then delete it, and create another one which will end up with the same VMID and back up that too (using automated backup), without manual intervention i suppose the old machine's backup will be overwritten/discarded (if we have a backup limit).
I am aware that we can extract the names from command line using vma and zcat, but that will not prevent automatic backups messing up stuff.
Are there any plans to address this (such as including the vm names in the backup name)?
Question 2:
Is there any way to configure in PVE to not reuse the VM IDs (or if not, at least new machines to use the next available VM iD)?
What is the VM ID limit (if any)? It does start from 100, but does it have an upper limit?
Question 1:
I see that the backups are named vzdump-qemu-VMID-date-time.vma.gz (for gzip format). This naming scheme is problematic because PVE tends to "fill the gaps" when vms are deleted.
If we have a vm, back it up then delete it, and create another one which will end up with the same VMID and back up that too (using automated backup), without manual intervention i suppose the old machine's backup will be overwritten/discarded (if we have a backup limit).
I am aware that we can extract the names from command line using vma and zcat, but that will not prevent automatic backups messing up stuff.
Are there any plans to address this (such as including the vm names in the backup name)?
Question 2:
Is there any way to configure in PVE to not reuse the VM IDs (or if not, at least new machines to use the next available VM iD)?
What is the VM ID limit (if any)? It does start from 100, but does it have an upper limit?