The Sun Cluster HA for BroadVision One-To-One Enterprise fault monitor checks BroadVision One-To-One Enterprise back-end and Interaction Manager process health. BroadVision One-To-One Enterprise process health impacts BroadVision One-To-One Enterprise resources' failure history, which in turn drives the fault monitor's actions. For each BroadVision One-To-One Enterprise resource, fault monitor actions include no action, restart, or failover.
For Interaction Manager resources, failover happens only when both of the following conditions are met.
The value of the desired primaries is less than the value of the maximum primaries.
One of the nodes is unavailable.
After failover, the fault monitor will not restart the resource on any cluster node if both of the following conditions occur.
The values of the maximum primaries and desired primaries of the Interaction Manager resource group are the same.
The fault monitor has completed restarting the Interaction Manager resource the number of times that the Retry_count property specifies.
The fault monitors for each BroadVision One-To-One Enterprise resource (root host, back-end host, and Interaction Manager host) monitor the following processes.
The orbix daemon (orbixd), which is common for all of the BroadVision One-To-One Enterprise resources – The probes use the ps(1) command to ensure that orbixd is functioning. If orbixd is not functioning, the probe considers the failure as complete, and the Resource Group Manager (RGM) restarts the orbix daemon.
The orbix daemon is started with the checkpoint feature. Therefore, the BroadVision One-To-One Enterprise servers, which the previous instance of orbixd started, continue to run with the new instance of orbixd.
The BroadVision One-To-One Enterprise daemons that you have configured in the resource – If orbixd is healthy, the probe uses the BroadVision command bvconf ps to ensure that the BroadVision One-To-One Enterprise daemons are functioning. If BroadVision One-To-One Enterprise daemons are not functioning, the RGM restarts the resource, which restarts all of the configured daemons.