Camunda waits an hour between messageStartEvent and the next call activity

Hi, we’ve been encountering an issue seemingly at random where processes stop and wait for a period of time between activities. The most recent occurrence happened between a message start event and the following call activity. Usually this would continue almost instantly, but this time there was a wait for almost an hour between the message start event ending and the call activity starting.

Here is an the process from the history table:

The message start event is standard, with an async after.

Has anyone encountered this issue, and found a resolution? We have not been able to find any cause for this. This has happened on multiple test environments.

Hi @confusedCamunda,

I published a blog post today that may help you understand and diagnose this behavior: https://blog.camunda.com/post/2019/10/job-executor-what-is-going-on-in-my-process-engine/

Cheers,
Thorben

1 Like

We managed to solve the issue. It was a problem with a third party random number generator that was a blocking generator rather than an unblocking one. We found this out by doing a thread dump which showed the cause.
Hope this helps anyone who has an issue like this in future.

1 Like