Activiti

Implement non interrupting boundaryEvent (cancelActivity="false")

Details

  • Type: New Feature New Feature
  • Status: Resolved Resolved
  • Priority: Major Major
  • Resolution: Fixed
  • Affects Version/s: None
  • Fix Version/s: 5.10
  • Component/s: Engine
  • Labels:
    None
  • Number of attachments :
    0

Description

see ACT-284, this is currently missing but specified in BPMN 2.0.

example:
<boundaryEvent id="escalationMail" cancelActivity="false" attachedToRef="firstLineSupport">
<timerEventDefinition>
<timeDuration>PT4H</timeDuration>
</timerEventDefinition>
</boundaryEvent>

Raised by support case https://app.camunda.com/jira/browse/SUPPORT-56

Issue Links

is duplicated by

Bug - A problem which impairs or prevents the functions of the product. ACT-1173 Non-interrupting timer boundary event suspends execution of subprocess

  • Major - Major loss of function.
  • Closed - The issue is considered finished, the resolution is correct. Issues which are not closed can be reopened.

Activity

Ascending order - Click to sort in descending order
Hide
Permalink
Bernd Ruecker (camunda) added a comment -

Implemented it, but test coverage could be improved, see notes from Joram:

I think with following process test, my confidence would be high enough to doc it.

  • concurrency, where on path is ended by catching the event
  • embedded subprocess + event
  • embedded subprocess with concurrency + event
  • call activity with concurrency + event

These last two I can do, since I'm currently testing them anyway

  • multi instance usertasks + event
  • multi instance on embedded subprocess + event

As a matter of fact, the above cases should be already defined for 95% in existing tests.

Show
Bernd Ruecker (camunda) added a comment - Implemented it, but test coverage could be improved, see notes from Joram: I think with following process test, my confidence would be high enough to doc it. concurrency, where on path is ended by catching the event embedded subprocess + event embedded subprocess with concurrency + event call activity with concurrency + event These last two I can do, since I'm currently testing them anyway multi instance usertasks + event multi instance on embedded subprocess + event As a matter of fact, the above cases should be already defined for 95% in existing tests.
Hide
Permalink
Joram Barrez added a comment -

Other todo: enhance ProcessDiagramGenerator to visualize non-interrupting events

Show
Joram Barrez added a comment - Other todo: enhance ProcessDiagramGenerator to visualize non-interrupting events
Hide
Permalink
Bernd Ruecker (camunda) added a comment -

Hi Joram. Can you enhance the ProcessDiagramGenerator? maybe you know hot to easily get the dotted line? If you prepare that I could try to add the rest to call it correctly. Or shall we move it to an own issue?

Show
Bernd Ruecker (camunda) added a comment - Hi Joram. Can you enhance the ProcessDiagramGenerator? maybe you know hot to easily get the dotted line? If you prepare that I could try to add the rest to call it correctly. Or shall we move it to an own issue?
Hide
Permalink
Joram Barrez added a comment -

I could do the dotted line, however I dont know if i will be able to get it in this release.

What rest call are you talking about? Afaics, nothing needs to be changed on that front.

Show
Joram Barrez added a comment - I could do the dotted line, however I dont know if i will be able to get it in this release. What rest call are you talking about? Afaics, nothing needs to be changed on that front.
Hide
Permalink
Bernd Ruecker (camunda) added a comment -

Sorry, I meant I do the ramining stuff But then I maybe just create a new issue, for me that's not urgent for this release, currently it is shown as interrupting, that is "just" a small flaw...

Show
Bernd Ruecker (camunda) added a comment - Sorry, I meant I do the ramining stuff But then I maybe just create a new issue, for me that's not urgent for this release, currently it is shown as interrupting, that is "just" a small flaw...
Hide
Permalink
Joram Barrez added a comment -

I added a failing test to the testcase.

Coverage is not good enough, so I'm removing the documentation. Postponing to next release.

Show
Joram Barrez added a comment - I added a failing test to the testcase. Coverage is not good enough, so I'm removing the documentation. Postponing to next release.
Hide
Permalink
Martin Pradny added a comment - - edited

Any plans to complete this? Currently it seems to work, but it can't be set in Designer.

Show
Martin Pradny added a comment - - edited Any plans to complete this? Currently it seems to work, but it can't be set in Designer.
Hide
Permalink
Nils Preusker (camunda) added a comment -

I added another failing test case in org.activiti.engine.test.bpmn.event.timer.BoundaryTimerNonInterruptingEventTest. It shows that the execution in a subprocess doesn't accept signals after a boundary timer fires, even if the cancelActivity attribute of the timer is set to 'false'.

This is because in org.activiti.engine.impl.jobexecutor.TimerExecuteNestedActivityJobHandler, the current activity of the execution in the sub-process is set to the boundary event, regardless of the value of the cancelActivity attribute.

Show
Nils Preusker (camunda) added a comment - I added another failing test case in org.activiti.engine.test.bpmn.event.timer.BoundaryTimerNonInterruptingEventTest. It shows that the execution in a subprocess doesn't accept signals after a boundary timer fires, even if the cancelActivity attribute of the timer is set to 'false'. This is because in org.activiti.engine.impl.jobexecutor.TimerExecuteNestedActivityJobHandler, the current activity of the execution in the sub-process is set to the boundary event, regardless of the value of the cancelActivity attribute.
Hide
Permalink
Daniel Meyer (camunda) added a comment -

Coverage is better now; all failing tests run. I am closing this issue. If problems arise, they can be filed as individual issues.

Show
Daniel Meyer (camunda) added a comment - Coverage is better now; all failing tests run. I am closing this issue. If problems arise, they can be filed as individual issues.
Hide
Permalink
Bernd Ruecker (camunda) added a comment -

I re-added the docs for it (was commented until it leaves experimental stage)

Show
Bernd Ruecker (camunda) added a comment - I re-added the docs for it (was commented until it leaves experimental stage)

People

  • Assignee:
    Daniel Meyer (camunda)
    Reporter:
    Bernd Ruecker (camunda)
Vote (5)
Watch (4)

Dates

  • Created:
    Updated:
    Resolved: