Hello,
I am setting up a PROXMOX HA cluster with 3 nodes N1, N2 and N3. Each node has one disk of 500G and one disk of 2TB .
I am using the 500GB for O.S and the 2TB disks with zfs.
I have distributed 4 VMs on each node. Each VM has a zfs disk of 100GB.
So on each node 400 GB of zfs space is used by the VMs
Now all VM are configured in HA. So I expect if one node crashes all the VMs are started on the other 2 nodes where we have the replica.
I am wondering now, if from N1 I make replicas of all 4 VMs to N2 and N3 (double redudancy) and N1 crashes, the cluster can choose to start the 4 VMs on N2 or N3 .
How will the cluster choose the node on which the VMs will start ?
Maybe will start 2 VMs on N2 and 2 VMs on N3 ?
Does it depend on free resources of the destination nodes ?
/Thanks
I am setting up a PROXMOX HA cluster with 3 nodes N1, N2 and N3. Each node has one disk of 500G and one disk of 2TB .
I am using the 500GB for O.S and the 2TB disks with zfs.
I have distributed 4 VMs on each node. Each VM has a zfs disk of 100GB.
So on each node 400 GB of zfs space is used by the VMs
Now all VM are configured in HA. So I expect if one node crashes all the VMs are started on the other 2 nodes where we have the replica.
I am wondering now, if from N1 I make replicas of all 4 VMs to N2 and N3 (double redudancy) and N1 crashes, the cluster can choose to start the 4 VMs on N2 or N3 .
How will the cluster choose the node on which the VMs will start ?
Maybe will start 2 VMs on N2 and 2 VMs on N3 ?
Does it depend on free resources of the destination nodes ?
/Thanks
