DAE can suppress rapidly recurring dumps automatically and the support staff does not need to be aware when a dump request recurs. However, a surge of dump requests could affect system performance, even though the dumps are suppressed. The surge could go unnoticed for hours. To help the support staff take actions to avoid impact to users, the system can notify you of high-frequency dump requests.
To obtain notification, add a NOTIFY parameter to the SVCDUMP statement on the ADYSETxx parmlib member to establish a threshold for notification. The SVCDUMP statement must also specify UPDATE. The default threshold is 3 dumps requested in 30 minutes for the same symptom string. The notification time is measured from completion or suppression of dumps, rather than from initiation of dumps.
Any program can receive the ENF signal. If active, the First Failure Support Technology™ (FFST™) issues a generic alert in response to this ENF signal.
If DAE is stopped and restarted, DAE begins counting dumps again to reach the threshold.
If each system has its own DAE data set, notification is for a system. If the systems of a sysplex share the DAE data set, notification is for the sysplex. For example, with a shared DAE data set, four dumps for the same symptom string on the same or different systems in 25 minutes would cause notification if the ADYSETxx parmlib member contains NOTIFY(4,25).