pgsql resource monitor operations time out in a RHEL 6 High Availability cluster with pacemaker

Issue

• pacemaker on the primary node for some reason tried to restart postgres after getting a number of unknown errors from the monitoring agent.
• Our pgsql resource timed out in a monitor operation
• Why might postgres become unresponsive to monitoring checks from the cluster and time out?

  Jun 29 00:05:06 [19486] node1.example.com       lrmd:  warning: child_timeout_callback:   postgres-service_monitor_4000 process (PID 3908) timed out
  Jun 29 00:05:06 [19486] node1.example.com       lrmd:  warning: operation_finished:   postgres-service_monitor_4000:3908 - timed out after 60000ms
  Jun 29 00:05:06 [19489] node1.example.com       crmd:    error: process_lrm_event:    Operation postgres-service_monitor_4000: Timed Out (node=node1.example.com call=41, timeout=60000ms)