Using logs to identify long-running escalations
In some cases, arserverd (arserver.exe) run escalations that take a long time to complete. These long-running escalations can be unexpected and can lead to escalation backlogs. Additionally, if the definition cache has been copied to accommodate definition or permission changes, long-running escalations can cause old copies of the cache to remain in the system much longer than expected. To help identify which escalations take a long time to complete, add the following parameter to the AR System Administration: AR System Configuration Generic UI form and turn on thread logging:
The value is the minimum number of seconds that it must take to complete an escalation before a line is added to the thread log to detail how long the escalation ran. In this example, the escalation must run for at least 5 minutes before it qualifies to be logged upon completion. The log entry will look like this:
/\* Thu Sep 29 2011 14:41:03.8090 \*/ Thread Id 3644 long running escalation - /RPC ID: 5/Escalation: CheckTicketSLA/Form: HPD:Help Desk/Elapsed time 315.9830 seconds/
You can set the parameter 0 to log the elapsed time of all escalations in the thread log. If the parameter does not exist or is set to a negative number, no long-running escalation logging will occur. You do not have to turn on escalation logging to add long-running escalation logging to the thread log. But you must turn on thread logging.
After you set this parameter in the AR System Administration: AR System Configuration Generic UI form, restart the BMC Remedy AR System server.
For more information on updating the AR System Administration: AR System Configuration Generic UI form, see Updating configuration settings by using the AR System Configuration Generic UI form.