Re: What's the proper way to "reraise" in a CLP?

I know that you are a software house, and as such may not be at liberty
to share code; could you perhaps describe how your web-based back end
handles the unexpected?

I can share code, and often do so when I believe it adds clarity. But if I
understand correctly, this discussion is more about message handling in
CLPs; whereas our web back-end is written in ILE RPG.

In the case of our web back-end, most message handling applies to to data
validation, RI constraints, and business rules pertaining to database I/O.
We generally have one service program that implements browser I/O, another
that implements database I/O, where both bind to a generic service program
that shares (exports) an array of messages which can be referenced by all
programs and service programs that are running in an activation group.

We have some CLPs that may implement logic pertaining to MONMSG events
generated by IBM i commands. In the event that a CLP calls another program,
a message may be returned via a call parameter.

My response to your question feels like we may be going off on a tangent,
though.