Hello,
We have a process in which an intermediate time event (non-interrupting) is triggered at 48H which got executed tens of thousands of times (~80000) during daylight saving hour.
After the first couple of times it threw:
java.lang.NullPointerException
at org.camunda.bpm.engine.impl.db.entitymanager.cache.DbEntityCache.getCachedEntity(DbEntityCache.java:145)
at org.camunda.bpm.engine.impl.db.entitymanager.cache.DbEntityCache.setDeleted(DbEntityCache.java:362)
at org.camunda.bpm.engine.impl.db.entitymanager.DbEntityManager.delete(DbEntityManager.java:513)
at org.camunda.bpm.engine.impl.persistence.entity.EverLivingJobEntity.init(EverLivingJobEntity.java:79)
at org.camunda.bpm.engine.impl.persistence.entity.JobManager.reschedule(JobManager.java:98)
at org.camunda.bpm.engine.impl.jobexecutor.historycleanup.HistoryCleanupSchedulerCmd.execute(HistoryCleanupSchedulerCmd.java:69) at org.camunda.bpm.engine.impl.jobexecutor.historycleanup.HistoryCleanupHandler$1.execute(HistoryCleanupHandler.java:49) at org.camunda.bpm.engine.impl.jobexecutor.historycleanup.HistoryCleanupHandler$1.execute(HistoryCleanupHandler.java:42)
After that we could see in the logs:
org.camunda.commons.logging.BaseLogger.logWarn ENGINE-14006 Exception while executing job 5ac528f9-99cc-11ea-a122-001a4a140319: OptimisticLockingException. To see the full stacktrace set logging level to DEBUG.
And then the task was triggered again.
We suspect that it has something to do with the daylight saving.
Could you please advise on how we could prevent this on happening in the future?
P.S. We’re using Camunda 7.12 on Tomcat and it’s database is on a PostgreSQL.
Thank you!