Health Overview

  1. From the Setup menu of the Enterprise Manager console, select Manage the Manager.

  2. Select the Health Overview sub-menu.

The Job System Status region of the Health Overview page displays the following information:

  • Step Scheduler Status

    The job step scheduler processes the job steps that are ready to run. If the status indicates that job step scheduler is running in warning or error mode, the job system is not functioning normally. In this case, the job system may run in fail-over mode, where the job dispatcher process may also run the task performed by the job step scheduler periodically. However, the job system may be running below its potential capacity, so resolving this situation would be beneficial.

    Several possible messages can appear:

    • DBMS_SCHEDULER job for step-scheduler not found

      This message is very rare and usually indicates a potentially serious issue. The job was likely removed inadvertently or due to some special processing

      (patch installation, for example, that requires recycling all DBMS_SCHEDULER jobs). No automatic resolution is possible here, and this would need to be addressed on a case by case basis.

    • Failure in checking status

      This is a rare occurrence. The error message is usually shown. The error may disappear on its own as this error indicates that the status could not be calculated.

    • DBMS_SCHEDULER is disabled

      All of the DBMS_SCHEDULER jobs are disabled in the environment. This should not occur unless a type of installation is in progress. Resolve this by starting DBMS_SCHEDULER processes.

    • All job queue processes are in use

      The DBMS_SCHEDULER processes have been expended. Increase the parameter job_queue_processes in the repository RDBMS.

    • All slave processes are in use

      The cause is similar to the above case. In this situation, you need to increase MAX_JOB_SLAVE_PROCESSES of the DBMS_SCHEDULER.

    • All sessions are in use

      No RDBMS sessions were available for the DBMS_SCHEDULER. Increase the PROCESSES for the RDBMS.

    • Reason for delay could not be established

      This usually appears because none of the above criteria were met, and is the most common warning. The dispatcher may just be overloaded because there is more work than available workers. Check the backlog in this case. The situation should resolve automatically, but if it persists, the number of workers available for the job system may be insufficient for the load the site experiences.

  • Job Backlog

    The job backlog indicates the number of job steps that have passed their scheduled time but have not executed yet. If this number is high and has not decreased for a long period, the job system is not functioning normally. This situation usually arises if job engine resources are unable to meet the inflow of jobs from system or user activity.

    A high backlog can also happen because of the abnormal processing of specific jobs because they are stuck for extended periods. For more information on stuck job worker threads, do the following:

    1. From the OMS and Repository menu of the Health Overview page, select Monitoring, then Diagnostic Metrics.

    If the jobs system has a backlog for long periods of time, or if you would like to process the backlog faster, set the following parameters with the emctl set property command. These settings assume that sufficient database resources are available to support more load. These parameters are likely to be needed in a Large configuration with 2 OMS nodes.

    Table 15-1 Large Job System Backlog Settings

    Parameter Value

    oracle.sysman.core.jobs.shortPoolSize

    50

    oracle.sysman.core.jobs.longPoolSize

    24

    oracle.sysman.core.jobs.longSystemPoolSize

    20

    oracle.sysman.core.jobs.systemPoolSize

    50

    oracle.sysman.core.conn.maxConnForJobWorkers

    144

    This setting may require an increase of the processes setting in the database of 144 * number of OMS servers.