8.2 JP1/IM - Manager health check function
JP1/IM - Manager provides a feature to detect when a process goes into an infinite loop or deadlock (hangup#) and ceases processing, and to notify the operator by issuing a JP1 event or executing a notification command. In JP1/IM - Manager, this is called the health check function.
#: Hangups are caused by a deadlock or infinite loop. This means that the process can no longer accept processing requests.
Use of the health check function enables early detection and response to process errors that can often go unnoticed, such as deadlocks and infinite loops, by issuing a message or JP1 event to report the error and prompt the operator to take recovery action.
The following figure gives an overview of the JP1/IM - Manager health check function.
|
|
![[Figure]](GRAPHICS/CS038600.GIF)
If enabled, the health check function outputs message information to the OS log (the Windows event log or UNIX syslog) and the integrated trace log when a process hangs. By making the appropriate settings in the health check definition file (jcohc.conf), you can also have the health check function issue a JP1 event or notification command when a process hangs.
JP1/Base also provides a health check function for monitoring the various JP1/Base processes. Used in conjunction, the JP1/IM - Manager and JP1/Base health check functions enable early detection and response for process errors in JP1/IM - Manager and instances of JP1/Base in the JP1/IM system.
The following describes the processes monitored by the health check function, how to enable and disable the function, and how the function monitors processes.