I have created one large process containing many different sub-processes.
The way it is designed is that initially the process fetches different kinds of data to be used later on in the main process.
It retrieves a list of all employees first, then the current payroll period, a list of all departments and finally a list of all functions.
All four subs are basically the same:
The list is retrieved, via the map shape the results (XML) are transferred to a flat file. These in turn are then split into single documents (data process shape) and finally added to a document cache.
This works fine and has been for quite a while using test mode, yet whenever I execute the deployed process it always ends up in an error saying:
No data produced from map "Main: Department List Flat Lookup", please check source profile and make sure it matches source data.
I have created a support ticket (#67475) before regarding this issue, but it somehow managed to go away for some time but it has reared its ugly head again now that we're going live.
Is there some way to determine why there is a difference in outcome between test and deployed execution? And even better still, would anyone have any suggestions as to how to avoid this?