Hi all,
I'm trying to setup a two node high availability cluster and so far everything seems ok. The trouble is that when I test fencing with:
it is successful but I have trouble re-adding the fenced node back into the cluster. I'm fencing using a HP 1910-16G switch and when I re-enable the fenced port, the fenced node immediately fences the first node! I have to move the fenced node to another switch port and restart rgmanager.
Now, I've not added DRBD or the quorum disk yet. Could this be the trouble? At the moment I'm just trying to test fencing. I don't have an VMs yet.
So, is this expected behaviour and am I trying to test too soon?
Many thanks,
NTB
I'm trying to setup a two node high availability cluster and so far everything seems ok. The trouble is that when I test fencing with:
Code:
[COLOR=#000000][FONT=monospace]fence_node NODENAME -vv[/FONT][/COLOR]
Now, I've not added DRBD or the quorum disk yet. Could this be the trouble? At the moment I'm just trying to test fencing. I don't have an VMs yet.
So, is this expected behaviour and am I trying to test too soon?
Many thanks,
NTB