Hi everyone,
If this has already been answered elsewhere, apologies—I couldn’t find it.
I understand that VMs can stream backups directly to PBS storage without needing local space.
However, for LXC containers, it seems they require a local tmpdir during backup.
Is that correct?
If so, this seems problematic.
For example, if you have an LXC with 1TB of disk, you’d need roughly 1TB of free space on every Proxmox node in the cluster just to perform backups.
That doesn’t feel practical.
Nodes shouldn’t need to reserve that much local storage just for backups.
Ideally, all backup-related storage should remain on the storage nodes, centralized for the cluster.
As it stands, we have several LXC containers that aren’t backed up because of this limitation.
Has anyone found a solution or workaround for this?
If this has already been answered elsewhere, apologies—I couldn’t find it.
I understand that VMs can stream backups directly to PBS storage without needing local space.
However, for LXC containers, it seems they require a local tmpdir during backup.
Is that correct?
If so, this seems problematic.
For example, if you have an LXC with 1TB of disk, you’d need roughly 1TB of free space on every Proxmox node in the cluster just to perform backups.
That doesn’t feel practical.
Nodes shouldn’t need to reserve that much local storage just for backups.
Ideally, all backup-related storage should remain on the storage nodes, centralized for the cluster.
As it stands, we have several LXC containers that aren’t backed up because of this limitation.
Has anyone found a solution or workaround for this?
Last edited: