RE: *INQ message sent to *EXT queue via QMHSNDPM not received via QMHRCVPM

["Dennis Lovelady" <iseries@xxxxxxxxxxxx>]

Thanks, Scott, but:

# rcvPgmMsg(msgRpyDS: %Size(msgRpyDS): 'RCVM0200'
#                   : '*EXT': *Zero
#                   : '*RPY': QMH_msgKey: -1: '*OLD': myUSEC) ;

Parameter 6 is *RPY.  Wouldn't that be the correct message type for
receiving a reply?

After beating my head against the wall on this, I have landed on the answer
(or an answer at least).  The problem that program has is that it's trying
to receive the response from *EXT rather than from the program message
queue.  Once I changed the QMHRCVPM parameter 4 from *EXT to * it began
working.  In my module-based version of this, then, I had to put the
QMHRCVPM in the same procedure as QMHSNDPM since I have no (known) way of
finding the name or address of the message queue of one procedure from
another procedure.  I have only one major recurring complaint about IBM's
ILE implementation, and that is the huge stumbling block that "program
message queue" and message management in general inevitably becomes.

I do appreciate the reply.

Dennis Lovelady
http://www.linkedin.com/in/dennislovelady
--
"Too bad all the people who know how to run the country are busy driving
taxis and cutting hair."
        -- George Burns 


> When a user replies to an *INQ message, the reply is a whole new
> message
> of type *RPY.
>
> The flaw in your code is that you are receiving the same message you
> sent.  You aren't looking for a new message of type *RPY...
>
> If it helps, I wrote a tip about this back in 2004
> http://systeminetwork.com/node/60943
>
>
>
> On 7/13/2010 2:47 PM, Dennis Lovelady wrote:
> > Hi, Folks:
> >
> >
> >
> > I'm trying to send a message to a program's *EXT queue for user
> response,
> > and then process the response.  But QMHRCVPM is not picking up the
> response.
> > (I get CPF2410 (Msg key not found in message queue.)).  I'm pretty
> sure I've
> > done everything by the book - using the msg key generated by QMHSNDPM
> as
> > parm to QMHRCVPM, et cetera - but the issue eludes me.  I have posted
> > minimal code to recreate at http://code.midrange.com/71e3be69e2.html
> and I'd
> > appreciate assistance!
> >
> >
> >
> > Note: "By the book" in this case means: according to the "Inquiry
> Messages"
> > discussion in the QMHSNDPM reference at
> http://publib.boulder.ibm.com/iseries/v5r2/ic2924/index.htm?infor/apis/
> QMHSN
> > DPM.htm.  I compared against V5R3 (current version here) and don't
> detect
> > any differences.