Re: QRG4033 - help needed

[Scott Klement <rpg400-l@xxxxxxxxxxxxxxxx>]

This certainly SOUNDS (albeit, I haven't looked at the code) like a 
coding error on your part.  I suspect that's why IBM is offering their 
services team to fix it -- because it isn't technical support's job to 
fix your code, their job is to fix IBM's code.

IMHO, going back to OPM makes no sense.

OPM RPG hasn't been enhanced since 1993.  It was already old back then.  
That's TWENTY-SEVEN YEARS AGO.  More than a quarter of a century!  This 
"solution" is not reasonable.    You've already run into a technical 
reason why this is not so easy.  But, think about the future...  You are 
stuck in 1980's technology, with no future.

This is not a reasonable business decision.  Whomever is making it is 
stuck in the past, or doesn't really understand the situation.

If you truly can't fix the coding error with OPNQRYF (which is also a 
legacy 1980's technology that you should be moving away from) then 
rewrite it to use SQL or some other, newer, better technology.   Don't 
go backwards.  And certainly not THAT far backwards.

-SK


On 4/15/2020 9:13 AM, dfreinkel wrote:
> I have a CLP that calls and OPM RPG program. The OPM RPG calls multiple
> other OPM RPG programs.
>
> I had to rewrite one of these and did so in ILE.
>
> The CLP has OVRDBF share(*YES), OPNQRYF and CLOF.
>
> On running the CLP program, the CLP would crash with an error linked to the
> OPNQRYF. I put back the old RPG program and the CLP and all programs run
> normally. The ILE program runs normally on its own.
>
> The CLP fails before it calls the first RPG program.
>
> IBM wants us to engage one of their services teams to resolve the error.
> Management said to go back to OPM.