Migrate/restart iSCSI VM on secondary node when primary is down

[beuc]

Hi,

I have 2 Proxmox nodes running VMs stored on iSCSI. It works well
Though, when one of the node is down (e.g. hardware failure), I'd like to restart the VMs on the other node.

When I try to migrate from the offline node, even if the VM was already shutdown, the system still tries to connect to the failed node, and I can't find a way to restart the VM on the other node.
Ideally a VM could be migrated offline from a node even if the node is unreachable.

Is there a way to do it?

 

[dietmar]

Is there a way to do it?

If the VM is on shared storage, and you are sure that the second node is not running, you can simply move the config file:

First, make sure the seconf node is offline!
Then set expected votes to get quorum

# pvecm expected 1

Then move the config file

# mv /etc/pve/nodes/<node1>/qemu-server/<vmid>.conf /etc/pve/nodes/<node2>/qemu-server/

 

[tom]

why don´t you use HA? see http://pve.proxmox.com/wiki/High_Availability_Cluster

 

[beuc]

Thanks, that works fine!
The HA Cluster looks quite interesting, we'll probably use it when/if we scale from 2 to 4 nodes. Is it possible to "upgrade" a normal Proxmox cluster to a HA Cluster later on?

 

[dietmar]

Is it possible to "upgrade" a normal Proxmox cluster to a HA Cluster later on?

Yes, that is no problem.