Critical Issues encountered

Hi Team

Request your help to understand and way to fix the following issues encountered

Background :

a. Flowable 6.4.0 - Spring boot engine includes listener class for posting message via listeners) used. Application deployed in both nodes - flowable.async-executor-activate=true in both nodes
b. Application Server- Jboss EAP 7.1 [Clustered] & MS SQL DB 2016
c. We are using triggerable= true , async=true in bPMN for service task. For task completion we are using runtimeService.triggerAsync(executionId, varaiables)… api

Issues encountered

  1. Workitem Not active : Cases are stuck in BPM and not moving to next designated flow
  2. Duplicate Workitem :
    a. Certain time when user takes decision duplicate Workitem gets created
    b. During process execution, duplicate Workitem got created in BPM at various stages in
    workflow at same time
  3. Tasks skipped : During process execution, certain steps in the flow has skipped (call activity) and has taken default path

Alle those points should of course not happen, but it’s hard to say anything without the process xml or diagram. Can you share this with us?

Also, could you share some more info about your use of the triggerable service task? Does it send out a message? Is this to a regular message queue or something else?

Hi

“Alle those points should of course not happen, but it’s hard to say anything without the process xml or diagram. Can you share this with us?”

I tried to attach , but restricted to attach bpmn file - instead sharing the link Common Service Executor

“Also, could you share some more info about your use of the triggerable service task? Does it send out a message? Is this to a regular message queue or something else?”

The service task is bound to delegate - which uses JMS Template to post message to a Active MQ queue.

What’s the scenario where you see the problems? Which tasks are stuck, duplicated or skipped? Does it happen all the time?

Regards
Paul.

Hi… received an email and have shared the updates. Please let me know if you require additional info

Also have observed in 6.4.0, when triggerAsync api is triggered , in ACT_HI_ACTINST table 3 records automatically gets created and end time is updated for the latest record and duration is measured for first and last record

We found the duplicate task getting created in single node.also

Shared link of BPMN file updated https://drive.google.com/file/d/1Ii1lNd4YPlI4nRSP4C-PKtAkVKwP9Oz8/view?usp=sharing