Techniques for investigating waits
You can investigate waits in a CICS® system by online inquiry, by tracing, or by analysis of the formatted CICS system dump. The last two techniques are, to some extent, complementary.
Online inquiry is the least powerful technique, and it can only tell you what resource a suspended user task is waiting for. This is enough information to locate the failing area, but you often need to do more investigation before you can solve the problem. The advantage of online inquiry is that you can find out about the waiting task as soon as you detect the problem, and so you capture the data early.
Tracing can give you much more detail than online inquiry, but it involves significant processing overhead. It must also be running with the appropriate options selected when the task first enters a wait state, so this usually means you need to reproduce the problem. However, the information it gives you about system activity in the period leading up to the wait is likely to provide much of the information you need to solve the problem.
A CICS system dump can give you a picture of the state of the CICS system at an instant during the wait. You can request the dump as soon as you notice that a task has entered a wait state, so it gives you early data capture for the problem. However, the dump is unlikely to tell you anything about system activity in the period leading up to the wait, even if you had internal tracing running with the correct selectivity when the task entered the wait. This is because the trace table has probably wrapped before you have had a chance to respond. However, the formatted dump might contain much of the information you need to solve the problem.
If you are able to reproduce the problem, consider using auxiliary tracing and dumping in combination.