Identifying and displaying alerts

The scheduler also sends alerts to the Tivoli® Enterprise Portal based on the MONALERT and MONOPER parameters of the ALERTS statement:
MONALERT
Defines the conditions under which IBM Z Workload Scheduler sends a generic alert to the Tivoli Enterprise Portal.
MONOPER
Defines if the conditions specified in the MONALERT parameter are in effect for the jobs that have the EXTERNAL MONITOR option set to YES (default) or for all jobs.
The alerts can be viewed from the Jobs view in the Tivoli Enterprise Portal.

Alerts can be sent on the following conditions:

  • For monitored jobs or for all jobs (in parentheses are the alert names that are displayed by the Tivoli Enterprise Portal interface):
    • Alerts for operations that ended in error (JobInError)
    • Alerts for late operations (LateJob)
    • Alerts for long durations (LongDurationJob )
    • Alerts for long time in the input queue (LongDurationJob)
    • Alerts for jobs promoted by WLM (WLMPromotedJob)
    • Alerts for the timeout of a special resource (ResourceTimeout). The timeout is specified in the RESOPTS CONTENTIONTIME parameter.
  • For IBM Z Workload Scheduler subtasks (in parentheses are the alert names that are displayed by the Tivoli Enterprise Portal interface):
    • Alerts for IBM Z Workload Scheduler subtasks or subsystems that ended in error (TaskInError)
    • Alerts for IBM Z Workload Scheduler subtask queues that exceed the threshold value (QueueThresholdExceeded)
    • Alerts for IBM Z Workload Scheduler controllers that lose monitoring events
  • Tivoli Enterprise Portal specific alerts:
    • Some events collected by the monitoring task are lost. A bulk discovery could be necessary to restore data consistency in the Tivoli Enterprise Portalinterface (QueueLostMonitorEvents).
    • An error occurred during a bulk discovery. A new bulk discovery could be necessary to restore data consistency in the Tivoli Enterprise Portal interface (ErrorDuringBulkDiscovery).
Note: If you recycle or stop the Z Monitoring agent, events are lost. The Z Monitoring agent might be stopped also as a result of a TCP/IP connection failure. In these cases, issue a bulk discovery to refresh the status of all monitored objects.