To build upon the previous posting (which included the Pause.KFP file), here are more tips related to debugging a workflow. Use multiple Preflight "Pause" nodes to track Core's workflow execution.
As each Pause node gets added to the workflow, make a copy of the node after placing within the workflow.
The copy helps to track which pause has been executed. When execution stops, the last executed node will indicate which copy of pause was executed. When creating the copy, Core will add a number (1, 2, 3, etc) after each copy of the Pause, thus making it unique. However, you can change the number to a name to aid in tracking.
To create the Pause Preset, import the Pause.KFP into a Preflight Preset:
Once your workflow has been debugged, remove the Pause nodes from the workflow.
As a side point, since these are presets, it's also possible to change the nodes to not stop by unselecting the "When an Error Occurs". Personally I wouldn't as the nodes will always be executed and could confuse the workflow. But it is available if you wanted to leave always leave a debugging Pause in a workflow. I would also suggest that you delete unused presets (but leave the original). This will keep your preset list cleaner, and they are trivial to recreate.
Lastly, how does Pause.kfp work anyway?
Remember that you can create a preflight profile using Acrobat. With Acrobat DC it's under the Optimize PDF tool -> PreFlight:
In this case, the Preflight "error" is raised whenever the PDF file has more than 0 pages. Which will always happen.
The workflow does not currently have a "pause" node. However, there are two ways to emulate something that would be similar to a pause node:
Option 1
Use preflight with a profile that errors if the job has more than zero pages and enable the preflight option to pause jobs that fail preflight.
Thi will pause all jobs that reach the preflight node.
The only downside of this is that it could create unwanted email notifications.
Option 2
Use the Collect node and set the release criteria to something that will never happen. Some examples:
You can manually release the collected jobs in Job Manager.
Also, you can route jobs based on why they were released, including whether they were manually released.