Job activation delayed by 10 seconds under stable network conditions #9658
Labels
kind/bug
Categorizes an issue or PR as a bug
version:8.1.0-alpha3
Marks an issue as being completely or in parts released in 8.1.0-alpha3
version:8.1.0
Marks an issue as being completely or in parts released in 8.1.0
Describe the bug
Sometimes jobs are activated exactly 10 seconds after the corresponding service task is activated. In general this can happen for various reasons, for example connection issue between worker, gateway and broker. 10 seconds is also the default request timeout for clients.
We've recently started to observe this in our QA runs, where these delays cause test failures. Since the network should be in a stable condition and this has only recently become an issue, it seems likely that this is actually a regression that we need to fix.
To Reproduce
With a current snapshot version of Zeebe and a spring-zeebe
8.0.5
this was at least not trivially reproducible.Expected behavior
Under stable network conditions and barring any other issues, there should be virtually no delay between service task activation and job activation.
Log/Stacktrace
Environment:
The text was updated successfully, but these errors were encountered: