IJ35491: 7.6.1.1-IFIX20200824-CRONMONITOR SHOULD NOT TREAT ALL UNRUN TASKS AS FAILED

APAR status

• Closed as program error.

Error description

• PROBLEM DESCRIPTION:
  The CRONMONITOR cron task considers any task instance that has
  not run yet to be overdue.
  It does not check to see if the instance is on a "run once"
  scheduled report.
  We have many users who schedule reports to run once at specific
  times and then end up getting
  cron monitor alerts like this one:
  
  1) Reports scheduled to run once (not recurring) are added as
  new instances of the REPORTSCHEDULE cron task.
  They're set up as yearly for the day/time the user specified to
  run.
  The REPORTSCHEDULE cron task has special logic to delete the
  instance after it runs.
  From the way the data looks on the cron task, though, it appears
  like it's supposed to run every year and just hasn't run yet.
  
  2) The CRONMONITOR considers any cron task instance that hasn't
  run yet as overdue.
  That logic is incompatible with the schedules of REPORTSCHEDULE
  instances for run-once reports.
  It means it will always determine that a REPORTSCHEDULE instance
  for a run-once report is overdue.
  
  Everything is running when it should. The CRONMONITOR is just
  doing the wrong thing when it runs.
  
  "The following cron tasks are overdue:
  REPORTSCHEDULE.1614095302239,
  REPORTSCHEDULE.1627419674031,
  REPORTSCHEDULE.1627419711392,
  REPORTSCHEDULE.1627419752956,
  REPORTSCHEDULE.1627419787864,
  REPORTSCHEDULE.1627419827256,
  REPORTSCHEDULE.1627419904275,
  REPORTSCHEDULE.1627419939542,
  REPORTSCHEDULE.1627419977139,
  REPORTSCHEDULE.1627420014671,
  REPORTSCHEDULE.1627420049141,
  REPORTSCHEDULE.1627420125846,
  REPORTSCHEDULE.1627420165441,
  REPORTSCHEDULE.1627420200302,
  REPORTSCHEDULE.1627907166365,
  REPORTSCHEDULE.1627418367539,
  REPORTSCHEDULE.1627418409025,
  REPORTSCHEDULE.1627418449541,
  REPORTSCHEDULE.1627418493668,
  REPORTSCHEDULE.1627418538076,
  REPORTSCHEDULE.1627418622126,
  REPORTSCHEDULE.1627418659283,
  REPORTSCHEDULE.1627418696347,
  REPORTSCHEDULE.1627418738348,
  REPORTSCHEDULE.1627418781428,
  REPORTSCHEDULE.1627418855024,
  REPORTSCHEDULE.1627418889791,
  REPORTSCHEDULE.1627418934168
  ...."
  
  None of those instances is overdue. They're scheduled to run
  once in the future and so haven't run yet, which is the expected
  behavior.
  
  This extra noise in the cron monitor alert first forced us to
  reschedule the monitor to a "daily"
  instead of "hourly" check just to cut down on the extra emails.
  But since we get the email every day now
  and it has so much incorrect data, it makes it very likely an
  actual failure is going to be missed among all
  incorrectly-flagged instances.
  It's impossible to look at that email content and distinguish a
  daily-scheduled report that failed from a future-scheduled
  report that isn't actually overdue.
  
  Action Taken:
  Found similar case TS003219360 that suggest to restart Crontask
  JVM to fix the issue. This didnot works.
  
  
  STEPS TO REPLICATE:
  
  There's no steps to replicate.
  
  System Information:
  
  App Server: IBM WebSphere Application Server 9.0.5.8
  
  Version: Tivoli's process automation engine
  7.6.1.1-IFIX20200824-1011 Build 20190514-1348 DB Build V7611-365
  HFDB Build HF7611-35
  
  IBM Maximo Asset Management Work Centers 7.6.0.4 Build $build$
  DB Build V7604-119
  
  IBM Tpae Integration Framework
  7.6.1.1-MIF_7611_IFIX.20200803-1640 Build 20190422-1259 DB Build
  V7611-01
  
  IBM Maximo Asset Management 7.6.1.1 Build 20190514-1348 DB Build
  V7611-01
  
  IBM Maximo for Transportation 7.6.2.0 Build 20160114-1557 DB
  Build V7620-12
  
  IoT Connection Utility 7.6.0.2 Build 20190426-2206 DB Build
  V7602-07
  
  Server OS: Windows Server 2016 10.0
  
  Server DB: Microsoft SQL Server 13.0 (13.00.5103)
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * MAXIMO                                                       *
  ****************************************************************
  

Problem conclusion

• CRONMONITOR SHOULD NOT TREAT ALL UNRUN TASKS AS FAILED
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  MAXIMO ASST MGM

• Fixed component ID

  5724R46AM

Applicable component levels