Followings are the basic eHealth cluster info you need to have prior to troubleshooting an issues:
Number of RFE and backend pollers, it’s normal to have more than one RFE
What is the OS for each RFE and backend pollers?
Has each member of the cluster run the same eHealth version and patch level?
Has each member of the cluster up and running in term of Oracle and eHealth? The both needs to be running for each member in order for a cluster to be functioning.
What is the hostname and machine_id for each member?