Description of problem: if using two nodes for HA nfs-ganesha the corosync conf file is updated within the quorum section with two_node: 1, the correct directive is two_nodes: 1 Version-Release number of selected component (if applicable): How reproducible: always Steps to Reproduce: 1. create two node HA nfs-ganesha 2. share a volume with nfs-ganesha 3. stop one of the two gluster nodes Actual results: corosync halts share Expected results: corosync is is in two node and fails IP over, service continues Additional info:
Please identify whether this is on RHEL7 or RHEL6. Nothing in RHGS makes direct changes to /etc/corosync/corosync.conf. All updates to it are made by running pacemaker command line tools. Any resulting typos are due to the tools or by pacemaker itself.
RHEL 7.1 and 7.2
I don't understand what exactly is problem. Both corosync.conf (RHEL 7) and cman cluster (RHEL 6) has "two_node" directive. Where you found two_nodeS? It's not supported in any version. Closing it as a not a bug.
Hey Kaleb, you reopened this bug. Does that mean there is a fix needed in out glusterfs-ganesha scripts?
I don't remember why I reopened it. closing, clear needinfo