We have a process to cancel all of the active process instances related to a particular work item in an application. We are achieving it with the help of a task report whereby we are passing the list of context process models under that App. Around 90% of our processes are activity chained and the majority of our sub-processes are called synchronously. The issue that we are seeing and haven't been able to figure out, happens intermittently. We receive an error notification (but the process instance doesn't show any error when we open it up) with the following verbiage -
"Problem: The sub-process started by this activity has been canceled.
Recommended Action: This node cannot complete. Please start the next node if necessary"
The cancellation happens using the 'Cancel Process' smart service which is configured to cancel the child processes first and then the parent process. Any insight or help would be highly appreciated. Thanks so much!
I'm a little fuzzy on the details, but in my experience this happens any time a running process has a subprocess that gets cancelled (like even if you go into the monitor and cancel it by hand). I've found that this happens every time. I believe the key reasoning behind this is that under normal circumstances, the parent process of the cancelled process might not necessarily ever proceed forward if the subprocess is cancelled, and they want to warn system admins and prevent permanently running, stalled parent process instances.
If you absolutely have to use the cancel node, then you can probably just ignore the errors. I'd suggest though that you look into maybe figuring out a way to terminate the parent process instead of cancelling -- terminating at the parent will also terminate any synchronous subprocesses, and without the error message.
Thanks for your insight, Mike I do agree that this may be a system-level inherent behavior and the reasoning that you provided behind it also makes sense.
Yup, the approach to Terminate the parent process also seems a logical way out to avoid getting this error. Thanks, we will look into it.
Do you have any batch process running to clear out the active instances? if yes, there could be chances of happening the issue. As I came across similar issue where it was a batch job.
This is not part of any batch process as such. It's a related action that tries to cancel all the active instances associated with that entity.
Were you able to figure out a solution for the issue that you were facing?
Yes, the issue was due to the next batch job picking the id's which were already been cancelled by previous batch job, so i changed the time interval and it worked for me.
Discussion posts and replies are publicly visible
© 2020 Appian. All rights reserved.