When opened timer Simultaneously, the process does not know he ends until the time of the timer comes to an end

[eranziz]

Hi,
I had main process and son process which been called by the main.

At the son process i had the flows displayed at the attached file(change his extension to png).

I really don't know why the process not ended if the end task of the process reached, meaning that we passed all the process until the end task but somehow the process remember that he open a timer.

how can i finish this process and ignore from the abandoned timers?
please help me
eran

[jbarrez]

The reason is because the timer is probably entered by a parallel execution of the process.
The end event in BPMN only terminates the current path. You could use the 'terminate end event' (however not yet document / fully tested) to end everything when arriving in the terminate end event.

[eranziz]

Hi,
Thanks you for replying.
Can you explain more about the 'terminate end event' , is it a java listener that make everything to be terminate?
When i add a end listener, the process doesn't execute this listener because he didn't know that he already finish.

If you didn't mean to add a end listener, so how can i terminate all the open timers\tasks of process?

[frederikherema1]

The terminate end-event is not a Java-listener but a construct in BPMN:

<endEvent id="EndEvent_2">
      <terminateEventDefinition id="TerminateEventDefinition_1"/>
<endEvent>

Add this instead of the normal end-event, it will cause all ope paths to be terminated when it's reached.

[jbarrez]

No, it is a specific BPMN element:

   <endEvent id="EndEvent_2">
      <terminateEventDefinition id="TerminateEventDefinition_1"/>
    <endEvent>

But like I said: it's a work in progress feature and not yet fully tested

[jbarrez]

Damn forum markup ….

<code>
   <endEvent id="EndEvent_2">
      <terminateEventDefinition id="TerminateEventDefinition_1"/>
    <endEvent>
</code>

[eranziz]

Hi,
Thanks for your reply..
I changed my end event to your end event suggestion and it's still not working,
The process doesn't know that he riches the end.
Any other idea?

[eranziz]

Could it be that the reason it does not work is the  version  that I use?
I'm using 5.9 version of activity…
Thanks,
Eran

[gromar]

I have found this feature too. I used simulation to predict probability of escalations for the following process:
[img]https://raw.github.com/gro-mar/activiti-crystalball/master/examples/src/main/process/org/activiti/cr...[/img]

May be another good point to use simulation for process design testing.

[jbarrez]

I don't remember which version we added it, but 5.9 might be a little bit too old.
Do you have an easy way trying it with the latest release?