How to detect long-running interactions (PEGA0019 alert)

Benefits

Through prconfig.xml file settings, you can cause the system to add a line to the Alert log when a requestor interaction requires a long time to process an input. You can set the interval; the default setting is ten minutes.

This alert can help you identify failed operations, infinite loop conditions, database server failures, and deadlock issues.

When a requestor interaction exceeds this time limit, the system adds a message to the alert log of this form, with additional details:

PEGA0019*Long running interaction detected.

Procedure

This facility is enabled by default, with a threshold of 10 minutes (600 seconds). To establish another threshold:

1. Update the alerts section of the prconfig.xml file, setting a longer or shorter desired threshold in milliseconds:

<env name="alerts/longrunningrequests/enabled" value="true" />
<env name="alerts/longrunningrequests/requesttime"
      value="{LIMIT]" />
<env name="alerts/longrunningrequests/notifications" value="[COUNT]" />

where [LIMIT] is a positive integer controlling the time to first alert in seconds and [COUNT] is a positive integer indicating maximum number of times the system reports an alert for one requestor. If not specified, the default threshold is 600 seconds; the default count is 3.

2. Stop and restart the server.

3. Interactions that trigger this alert appear in the Alert log.

As an alternative to the prconfig.xml file, you can use Dynamic System Settings to configure your application.
See How to create or update a prconfig setting.

Notes

By default, the master agent examines requestors every 60 seconds. In the requesttime entry, specify a limit of at least 120 seconds. If you change the wake-up period for the master agent, choose an interval that is at least twice as long as the wake-up period.

Other alert types — for example, the PEGA0007 database operation threshold — execute within a requestor context and so are recorded only when a response occurs. As a result, other alert types cannot report indefinite looping or failures. For example, if the database server fails completely, or the server looses connectivity to the database server, this PEGA0019 alert appears after 10 minutes, but the PEGA0007 alert does not appear.

	alert log, master agents, prconfig.xml file
	Understanding alerts

SysAdmin category