In RHEL3, there is a section about "Diagnosing and Correcting Problems in a Cluster": http://www.redhat.com/docs/manuals/enterprise/RHEL-3-Manual/cluster-suite/s1-admin-problems.html and another one about "Failover and Recovery Scenarios": http://www.redhat.com/docs/manuals/enterprise/RHEL-3-Manual/cluster-suite/s1-swinfo-fail.html Those sections are missing from the RHEL4 documentation.
At the time of publishing the RHCS4 adming guide (Red Hat Cluster Suite: Configuring and Managing a Cluster, rh-cs(EN)-3-HTML-RHI (2004-06-04T17:42), the information for those sections was either not available or not applicable to the RHCS4 release. We will check to see if information is now available or is applicable. Thank you.
All the stuff in there is mostly not applicable. The SCSI bus termination information is for multi-initiator buses and/or hardware configurations we don't support. We no longer have separate heartbeat channels. We no longer require quorum partitions, and as of U1, we didn't support serial power switches. The Failover and recovery Scenarios section is empty in the RHEL3 manual, it looks like.