I have a proxmox cluster where each node has two tiers of storage: fast and small (=SSD or NVME) named "localSSD" and slow and large (=HDD) named "localHDD". Everything is ZFS. The VMs running are not HA configured. They can go down for some time, but preferably not hours/days. Each VM has a preferred node where it usually runns (due to different hardware).
Now I want to be prepared if a node goes down, therefore I thought of replicating all VMs to at least two other nodes. The SSD storage is not large enough for that, the HDD storage is.
Is it possible to use replication in such a way that the disks are replicated to a storage with a different ID?
I have not used the new proxmox backup yet, would it be able to accomplish that? It is very important to have incremental updates since some VMs are rather large but do not change that much.
Additional challenge: some VMs might have 2 disks, one on the SSD and one on the HDD.
Now I want to be prepared if a node goes down, therefore I thought of replicating all VMs to at least two other nodes. The SSD storage is not large enough for that, the HDD storage is.
Is it possible to use replication in such a way that the disks are replicated to a storage with a different ID?
I have not used the new proxmox backup yet, would it be able to accomplish that? It is very important to have incremental updates since some VMs are rather large but do not change that much.
Additional challenge: some VMs might have 2 disks, one on the SSD and one on the HDD.
Last edited: