drbd and high availability.

[RobFantini]

Hello
we've a dual primary drbd set up . and due to rgmanager issues [ or as likely something we set up wrong] ha cluster using drbd is not working for us.

my concern is this

suppose one node has a motherboard break.

How can we manually tell the surviving node how to take over the kvm's on drbd?

 

[dietmar]

First. make sure the other node is really offline.

The set expected votes to gain quorum:

# pvecm expected 1

Then simply move the vm config files:

# mv /etc/pve/nodes/[oldnode]/qemu-server/[vmid].conf /etc/pve/nodes/[newnode]/qemu-server/

After that you can start the VMs on [newnode].

 

[RobFantini]

Thank you, I'll copy and paste that to notes in /etc/pve/ .