it may happen that a job stream is considered being "running" when in fact the job stream is "completed".
E.g. when there are two starters in the job stream and starter1 will run job1 and job2 and starter2 will run job3 and job4. When now starter1 starts the job stream and job2 is finished, the job stream is still considered as "running" as there are unresolved In Conditions in the job stream.
When starter1 does not have a configured "last job" then the resolver can not decide whether the job stream is completed.
The resolver examines all running job streams. Increasing the number of running instances will impact performance.
It should be possible to set the running state of a job stream to "completed". This should be done in the job stream history view. Also a bulk operation should be possible.