I'm having an odd issue with fencing - after I fence a node (using IPMI)
and it reboots, it becomes the cluster manager, migrates *almost* all of
the VMs to itself, and refuses to let me migrate any VMs back to other
nodes, complaining that other nodes are dead:
> Executing HA migrate for VM 102 to node pve03
> Trying to migrate pvevm:102 to pve03...Target node dead / nonexistent
> TASK ERROR: command 'clusvcadm -M pvevm:102 -m pve03' failed: exit
code 244
I then have to reboot (or fence) the other nodes before the first node
is willing to migrate anything to them again.
Presumably, this is not correct behaviour - any ideas?
--
-Adam Thompson
[email protected]
_______________________________________________
pve-user mailing list
[email protected]
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
