AnsweredAssumed Answered

Timer Start Event 'forgotten'

Question asked by tstephen on Jun 21, 2015
Latest reply on Jun 22, 2015 by vasile.dirla
Hi all,

I run a process every 10 minutes that GETs from one REST API and POSTs to another. For a little while now I have found that the timer start event is 'forgotten' at unpredictable intervals. These are the symptoms I observe:

- If an exception occurs during one of the REST calls then the insert of the next job seems to be skipped (and therefore the job does not run again)
- Once in this state it is not possible to delete the process definition, i.e. processEngine.getRepositoryService().deleteDeployment(id, true); 

Today, however I have noticed that the job has stopped again only this time I have this in the ACT_RU_JOB table:

| 10074293 |   16 | timer | NULL           | NULL        |          1 | NULL          | NULL                 | LogCalls:1:10046996 |       -3 | 10074305            |
### Error updating database.  Cause: java.sql.SQLException: Lock wait timeout exceeded; try restarting transaction
### The error may involve org.activiti.engine.impl.persistence.entity.JobEntity.insertTimer-Inline
### The error occurred while setting … | 2015-06-20 22:59:39 | R/2015-06-20T20:05:38/PT10M | timer-start-event | {"activityId":"LogCalls"} | tenant   |

Any suggestions of where to investigate would be appreciated.