The bane of all clustered computing systems is sympathy sickness, and sysplex is no exception. Failure to deal with an unresponsive component or an unresponsive system in a timely manner can lead to mysterious slow downs in the sysplex, even outright catastrophe. But finding the culprit and the root cause of these "sick but not dead" situations can be difficult, especially when you are under pressure to restore the sysplex to normal operation (now!). Over the years, the sysplex components have made enhancements to address this issue.
In this session, we consider the various ways in which sympathy sickness can be induced in the sysplex. With that context, we then examine the various messages, commands, and reports that we in IBM Support use to get insights into what is going on in the sysplex (or not going on as the case may be). This session will make you aware of the information that is available for diagnosing sympathy sickness problems and describe how to make correct inferences from that information. In cases where you still can't figure out the problem from the data at hand, you will be more prepared for your engagement with the IBM support team.