Solutions

IDM 5.x or 6.x loses connection to database repository which stops LiveSync working

Last updated Nov 2, 2020

The purpose of this article is to provide assistance if the scheduler stops working after IDM loses connection to the repository database. This typically affects LiveSync schedules. You will likely see "Error updating trigger in repo" errors when this happens.


Symptoms

The scheduler stops working, which may be seen as LiveSync not working.

An error similar to one of the following is shown in the IDM log when this happens (depending on the log level):

  • WARNING: WARNING: Scheduled service "scheduler-service-group.NAME" invocation reported failure: org.forgerock.json.resource.PreconditionFailedException: Update rejected as current Object revision 100 is different than expected by caller (99), the object has changed since retrieval. org.forgerock.openidm.quartz.impl.ExecutionException: org.forgerock.json.resource.PreconditionFailedException: Update rejected as current Object revision 100 is different than expected by caller (99), the object has changed since retrieval.
  • SEVERE: SEVERE: An error occured while firing trigger 'scheduler-service-group.trigger-NAME' org.quartz.JobPersistenceException: Error updating trigger in repo [See nested exception: org.forgerock.json.resource.PreconditionFailedException: Update rejected as current Object revision 100 is different than expected by caller (99), the object has changed since retrieval.]

Recent Changes

IDM lost connection to the repository database.

The repository database has an outage or was restarted.

Causes

The database repository is used to keep track of the scheduler states. Once the database comes back up or the connection is restored, IDM reconnects correctly but the outage causes the scheduler to have an invalid state (corrupted). As a result, schedules (including LiveSync) stop working.

Solution

This issue can be resolved by upgrading to IDM 7 or later; you can download this from BackStage.

Workaround

You can workaround this issue by updating the schedule with the existing configuration using REST as described in Integrator's Guide › Updating a Schedule. This effectively resets the scheduler configuration, which enables it to resume working again.

See Also

How do I configure the scheduler in IDM 5.x and OpenIDM 4.x to ensure JSON tasks are executed correctly when the time changes?

How do I monitor LiveSync activity using REST in IDM/OpenIDM (All versions)?

Integrator's Guide › Scheduling Tasks and Events

Related Training

N/A

Related Issue Tracker IDs

OPENIDM-12568 (Schedules may stop working after repository db outage)



Copyright and TrademarksCopyright © 2020 ForgeRock, all rights reserved.
Loading...