JobExecutor infinite loop?

Camunda 7 Topics Discussion & Questions
1

Hi,

I am facing a problem with the JobExecutor currently. My jobs are not executed although the due date is long passed (days and weeks).

I turned up the camunda logging. The Tomcat startup sequence seems fine. But soon I found that the JobAcquisition thread only sleeps for ~100ms and it seems that this thread never comes to an end. I also cannot see any “Starting command” and “Finishing command” in the log, once those rapid job acquisitions started.

I have no special configuration for my Jobexecutor like “waitTimeInMillis”.

Has anyone any idea what can cause such a behavior?

ENGINE-13006 Finishing command -------------------- AcquireJobsCmd ----------------------
ENGINE-14022 Acquired 1 jobs for process engine ‘myprocessengine’: [[5bd2302b-6eec-11ea-95e1-0050569375b1]]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd0f7aa-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd45310-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd2302b-6eec-11ea-95e1-0050569375b1]
ENGINE-13005 Starting command -------------------- AcquireJobsCmd ----------------------
ENGINE-13009 opening new command context
ENGINE-13005 Starting command -------------------- ProcessDefinitionQueryImpl ----------------------
ENGINE-13009 opening new command context
ENGINE-13011 closing existing command context
ENGINE-13006 Finishing command -------------------- AcquireJobsCmd ----------------------
ENGINE-14022 Acquired 1 jobs for process engine ‘myprocessengine’: [[5bf150fd-6eec-11ea-95e1-0050569375b1]]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd45310-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd2302b-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bf150fd-6eec-11ea-95e1-0050569375b1]
ENGINE-14022 Acquired 0 jobs for process engine ‘myprocessengine’:
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd45310-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd2302b-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bf150fd-6eec-11ea-95e1-0050569375b1]
ENGINE-14011 Job acquisition thread sleeping for 100 millis
ENGINE-14012 Job acquisition thread woke up
ENGINE-14022 Acquired 0 jobs for process engine ‘myprocessengine’:
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd45310-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd2302b-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bf150fd-6eec-11ea-95e1-0050569375b1]
ENGINE-14011 Job acquisition thread sleeping for 100 millis
ENGINE-14012 Job acquisition thread woke up
ENGINE-14022 Acquired 0 jobs for process engine ‘myprocessengine’:
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd45310-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd2302b-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bf150fd-6eec-11ea-95e1-0050569375b1]
ENGINE-14011 Job acquisition thread sleeping for 99 millis
ENGINE-14012 Job acquisition thread woke up
ENGINE-14022 Acquired 0 jobs for process engine ‘myprocessengine’:
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd45310-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bd2302b-6eec-11ea-95e1-0050569375b1]
ENGINE-14023 Execute jobs for process engine ‘myprocessengine’: [5bf150fd-6eec-11ea-95e1-0050569375b1]
ENGINE-14011 Job acquisition thread sleeping for 100 millis

Greetings,
Chris

2

Hi @TheFriedC,

I am also facing exactly the same issue that you mentioned. Did you find the reason for this and what was the fix that you applied ?

Thanks,
Siddharth

3

HI,

sorry for the late answer.

To be honest, I am not entirely sure what solved the problem. I did a few things. One thing was to upgrade to the latest Camunda version.

But I guess it was a somehow faulty workflow definition being deployed in the workflow engine, where the classname attribute of a task was set without a value.

While writing these lines, it does not occur logical to me that such an error could have been the reason for the problem.

Nevertheless - I would recommend to strip down the setup to as few as many process definitions and process instances if you have the chance.

I am sorry for not being a good help!

Chris

4

Hi Chris,
Thanks for your reply.
Due to organization restriction, we cannot upgrade to latest version.

I will follow your suggestions on the process definitions.

The isolation level in mysql is REPEATABLE-READ but in camunda sevrer.xml, the isolation level is configured as READ-COMMITTED. So I believe the transaction level isolation level of READ-COMMITTED would be used and the REPEATABLE-READ level will not be used for camunda.
Any suggestions on this please ?

Thanks,
Sidd

5