Why a Process could get stuck in Aborting status

Products

• Clarity PPM On Premise

Issue/Introduction

After I cancelled a process instance, it stays in "Aborting" status for a long time. Why? I want to delete this instance so that I can make its process definition back to "Draft" mode. How long should I wait before it is in "Aborted" status? 

Environment

Release: Any

Resolution

• When you cancel a process instance on UI, the Application Service broadcasts an "Abort Process Instance" to the cluster and changes the status of the process instance to "Aborting".
• Upon receiving the message, a Process Engine will check whether the process instance is being loaded by this engine.
• If not, it ignores the message; otherwise, it will change the status of the process instance to "Aborted" and unload the process instance from the cache. 
  
  So after the process instance is cancelled on UI, if it stays in "Aborting" status for a long time, it can be one of the following: 
• There is no process engine running in the cluster The Message Server on the Process Engine that currently loads the process instance for whatever reason did not respond to the "Abort Process Instance" message.
  • In this case, the Process Controller thread will complete the process instance abortion in the next Exception Run Cycle, which is due in 10-20 minutes, or depending on the "Exception Run Interval" setting in NSA. 
• The aborting process instance is currently not being loaded by any Process Engine in the cluster. This can occur in the initialization period of a Process Engine, when process instances in the system are still being loaded by the Process Loader thread.
  • In this case, the Process Controller thread will complete the process instance abortion in the next Exception Run Cycle, which is due in 10-20 minutes, or depending on the "Exception Run Interval" setting in NSA. 
     

Additional Information

Also check Processes stuck in Aborting status for long time, how to set them to Aborted