RHEVM : A question of fencing
Hello
My understanding is that although the hypervisor fencing is triggered via RHEV-M , the actual fencing is handled by the hypervisors. For instance, a restart of a hypervisor A is requested via RHEV-M webadmin but the fencing of the hypervisor A node is carried out by hypervisor B.
Is that correct ?
If so, why not fence hypervisors directly from the Manager ? I'm sure there's an obvious answer.
Thank You