Using CICS trace

CICS® trace is a debugging aid for application programmers, system programmers, and IBM® field engineers. General CICS tracing is handled by the CICS trace domain. It traces the flow of execution through CICS code, and through your applications as well. You can see what functions are being performed, which parameters are being passed, and the values of important data fields at the time trace calls are made. This type of tracing is also useful in first failure data capture, if an exception condition is detected by CICS.

Trace points are included at specific points in CICS code. From these points, trace entries can be written to any currently selected trace destination. Some trace points are used to make exception traces when exception conditions occur, and some are used to trace the mainline execution of CICS code.

CICS provides different levels of tracing to assist with problem determination. Standard trace level 1 is the default setting for each component to be traced within CICS. The user can use CETR to specify what trace levels are set for each component of CICS. By default, INTTR, SYSTR, and USERTR are set ON. This means the main system and user trace flags default to be set on, and internal tracing is active. STNTR defaults to 1, as do all the STNTRxx values, and as a result standard trace component tracing defaults to level 1 for all CICS trace components. The consequence of this is that by default a CICS system incurs the CPU usage to provide this level of internal CICS trace data.

There is a trade-off between the CPU cost in capturing trace data for problem determination, set against the ability to diagnose problems if they occur. Some customers elect to run with limited levels of tracing active on their system. While choosing to use CICS trace does increase processing requirements, not using CICS tracing reduces the amount of problem determination information that is available for the CICS region.

Note: CICS always performs exception tracing when it detects an exception condition, so a minimum of first failure data capture is provided regardless of your CICS trace setting. However, exception tracing by its nature is limited in what diagnostic data it can provide. It is difficult to perform initial problem determination without CICS tracing being active and all components capturing their trace data, as it is the trace information that helps to identify the flow of system activity, and the events in chronological order, leading up to a failure. For this reason, to assist with problem determination it is recommended that the default settings of all CICS domains and components is used when tracing is active. This is standard trace level 1 tracing.

All CICS trace points are listed in alphabetical sequence in Trace entries.