Perform HA on all nodes

[Jeremy69]

Hello everyone.

I've tried to find a way to implement HA (High Availability) across all nodes, but I haven't found how to do it, except on NFS where I managed to. However, the issue is that when I turn off this NFS server, it makes my machines that rely on NFS unavailable. Is there a way or another method to use HA without my machines becoming unavailable ?

 

[fiona]

Hi,
yes, you need shared storage for HA: https://pve.proxmox.com/pve-docs/chapter-ha-manager.html#_requirements
But if the shared storage fails, there's nothing HA can do. Many people use Ceph, because that is more resilient than NFS: https://pve.proxmox.com/pve-docs/chapter-pveceph.html

 

[aaron]

For small clusters, ZFS + Replication can also be an option. If you are fine with some potential data loss in case of HA recovery, because it is async and the shortest possible interval is currently 1 minute.