fence_ipmilan fails when using method="cycle" (-M cycle or -m cycle) in RHEL 5, 6 or 7, even though it succeeds in rebooting the node in question
Issue
fence_ipmilan
works when run manually, but with cluster or usingfence_node
command it is not working.- After a fence event, node is rebooted by fence_ipmilan but it can't rejoin the cluster
- I use the method "cycle" of the fence_ipmilan fence agent (
fence_ipmilan -o reboot -m cycle
)
Environment
- Red Hat Enterprise Linux (RHEL) 5 with the High Availability Add On
cman
prior to release 2.0.115-109.el5_9.5
- Red Hat Enterprise Linux (RHEL) 6 with the High Availability Add On
fence-agents` prior to release 3.1.5-17.el6
- One or more nodes configured with
fence_ipmilan
as theirfencedevice agent
in/etc/cluster/cluster.conf
fence_ipmilan
configured withmethod="cycle"
(-M cycle
from the command line-m cycle
in RHEL 6, 7)
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.