Glad to see that this doc was updated to include a statement on HPOM clusters, however the statement leaves a lot to be desired. Here is the text that was added.
I have an issue when I attempt to run HP Operations Manager Incident Web Service SmartConnector in a cluster environment.
An issue can result if you attempt to run the HP Operations Manager Incident Web Service SmartConnector in a cluster environment. By default, the connector is not designed to work with an HP OM cluster. If installed in a cluster, and if during a failover the active node changes to a standby node within the cluster, the connector cannot accommodate this change to the standby node and does not automatically resubscribe.
To resolve this issue, do not attempt to run HP Operations Manager Incident Web Service SmartConnector in a cluster environment.
How can the resolution to a problem be "don't attempt"? You could state that this is not supported due to a limitation with the SmartConnector's web services functionality (ie. always assuming that a web service subscription is valid and not monitoring is the subscription still works).
For customers that have HPOM running in a cluster, the way that I solved this was to install a SmartConnector on each HPOM cluster node and configure the cluster to start the SmartConnector on the active node. This way, when the cluster flips from node A to node B, the SmartConnector running on node A stops and the SmartConnector on node B starts and subscribes to the HPOM Incident web service.