Currently if the engine is started, for example in an application server as shared engine, the job executor is also started.
After start up the executor finds no jobs to acquire and suspends for 5 seconds, after 5 seconds
he tries to acquire new jobs. If no jobs are found he suspends for 10 seconds and so on until the max suspend time is reached, which is 60 seconds.
If now a process is started, with for example a boundary timer event, the job executor gets an hint, but since the timer is in the future he can't acquire the job and suspends for 60 seconds again.
Is the timer duration less then a minute this represents a problem because the timer is executed after 60 seconds, since the job executor suspends for this time.
AT:
- The job executor waiting time should be reseted after a hint was given to the initial waiting
time. So if again the job executor acquire no jobs he suspends for 5 seconds and tries again
not for 1 minute!