MarkView All Components – Events not being processed

Applies To

• ERP System: All
• MarkView Version: All

Behavior

• One or several managers are not processing events
• Events take a long time to process
  • It may lead to notifications being sent at incorrect times (i.e. late)
• The number of Pending Events reported in MarkView System Status is increasing
  • On MarkView Home -> Support -> MarkView System Status.
• A DBMS Jobs that are normally running is not broken
  • On MarkView Home -> Support -> Database Job Administration the last run is different from expected
  • The ‘Broken’ column is set to ‘No’
  • The Non-Production environment has the same jobs running

Known Causes

• The number of DBMS jobs running on the database has increased
  • The new DBMS jobs may not be MarkView related
• The job_queue_processes parameter in v$parameter is too small
  • The database will process as many jobs as indicated in the job_queue_processes; the rest of the DBMS jobs (MarkView or not) will be queued.

Resolution

• The job_queue_processes parameter needs to be large enough to cater for the Database DBMS jobs – some of which will be MarkView.
• Check the value of the job_queue_processes parameter in v$parameter
  • On MarkView Home -> Support -> Database Job see the current value listed towards the top of the screen
  • It should also say how many MarkView DBMS job you have
  • It does not indicate how many other non-MarkView DBMS jobs you have

Note: If the value is not listed then please use script to find the value:

• select * from v$parameter where name = 'job_queue_processes';
• If the value is too low then please request the DBA team to increase the value.
  • We normally except a value greater than 15
  • Please confirm the value against other working environment to confirm the expected value.
  • The DBA Team will be familiar with the process to change the value.
    • Example to set the parameter to 20 jobs:
    • alter system set job_queue_processes= 20;
    • commit;
  • Increasing the parameter will not automatically resolve the issue. It will allow more jobs to run simultaneously which in time should allow more DBMS jobs to process events.