Symptom
In a two-server cluster, when the primary server fails or is shut down which causes the hierarchies to fail over to the backup server, and the backup server also fails or is shut down before the hierarchies are entirely failed over to the backup server, the following behavior has been detected:
When both servers are rebooted, some of the resources in the hierarchies will be in service on one server and some will be in service on the other server. Some of the higher-level parent resources may not be in service on either server.
Solution
After both servers have been restarted and have completed LifeKeeper for Windows initialization, select the parent resource in a hierarchy that did not come in-service from the Hierarchy Administration interface and bring it in-service manually. Repeat this task until all hierarchies are in-service.
Post your comment on this topic.