JobScheduler hangs if an order job has an end time in its current run time period and the task of the job needs more time

Current Situation
An order job has a run time with a begin/end interval and let_run="no" (default)

<job order="yes">
    <script  language="shell">
        <![CDATA[
@echo off
ping -n 1 -w 180000 1.1.1.1 > nul
exit 0
        ]]>
    </script>

    <run_time >
        <weekdays >
            <day  day="1 2 3 4 5 6 7">
                <period  begin="21:00" end="21:10"/>
            </day>
        </weekdays>
    </run_time>
</job>

If this example job is started by an order at 21:08 then it needs longer (3 min) than the end time 21:10 is reached.

• In this case the JobScheduler hangs. It means
  • ... no answer per TCP
  • ... no further log line.
• It seems dead but e.g. a windows service is still running
• This issue doesn't occur if let_run="yes".

Desired Behavior

• The JobScheduler doesn't hang.
• let_run doesn't have an effect for a task of a shell job which has reached the end of the run times period.
• let_run has only an effect for tasks of an api job which have multiple process steps; see http://www.sos-berlin.com/doc/en/scheduler.doc/xml/run_time.xml#attribute_let_run.

Workaround
Configure the period in the job's run time with let_run="yes".

<period  begin="21:00" end="21:10" let_run="yes"/>