Tracing is a general way of collecting detailed information about how a process runs. Tracing consists of reporting information about events in a process; events include the calling of and return from functions, the exit of processes and other significant occurrences in the execution of a program. Trace information can be used to debug an application and to tune the operation of a product. Tracing can assist application developers and system administrators, and it is often crucial when asking for product support.
Tracing every possible event in a complex product like CICS® can lead to an unmanageable amount of trace information, so you need a way to specify the types of events to examine. In CICS, events are classified so you can selectively trace them. Additionally, you can choose the CICS processes and CICS modules within a region that you want to trace. With such information, you can specify the tracing with precision, and eliminate irrelevant data.
Each type of process in a CICS region; for example, IP listeners, application servers, and the CICS main process, is traced separately, and trace configuration information is held in the region-wide master trace area (MTA), as shown in Figure 1. The MTA holds the cold-start values for tracing, which are obtained from the trace-related Region Definitions (RD) stanza entries. These RD stanza entries describe the types of trace to collect, the process types and modules from which to collect it, and where to store the collected trace. See Summary of trace-related RD stanza entries for a brief description of each of the trace-related RD stanza entries. These configuration values can be modified as the region starts by modifying the value of the CICSTRACE environment variable, as described in CICSTRACE environment variable.
