I think the QRPLOBJ may be the 'issue'. I've never run into or knew about
that library, and it seems like a good thing to know about. I went back
to my first updated program and did a refresh on the SEP and it is
debugging now. *(I tried removing/re-adding the SEP previously and that
didn't work, so I guess I need to make sure to do the refresh).
Thanks for the help.
------------------------------
message: 4
date: Wed, 5 Jul 2017 10:19:41 -0500
from: broehmer@xxxxxxxxxxxxxxx
subject: Re: [WDSCI-L] Service Point Entry not working (on RDI 9.5)
Just a thought, but have you looked to see what object it is really
using when it's running? (dspjob option 11). Sometimes I
get the QRPGOBJ library object instead of the new compiled object.
------------------------------
message: 5
date: Wed, 5 Jul 2017 12:19:53 -0400
from: Buck Calabro <kc2hiz@xxxxxxxxx>
subject: Re: [WDSCI-L] Service Point Entry not working (on RDI 9.5)
On 7/5/2017 10:18 AM,
SReeves@xxxxxxxxxxxxxxxxxx wrote:
After a bunch of testing and trying to figure out if I had the right
username I ended up going back to the original program, and that would
stop for the Service Entry Point. So I added some free code in it, and
it
still worked.
This describes my environment. I have RPG III-style code that I add
fully-free specs into. I can't ever recall a problem with the debugger
not stopping due to a mix of specs.
I then changed one line of code in the main C specs to free
form (a call subroutine), and it stopped working again. So I changed
that
line back to fixed format and it still would not go into debug.
This implies:
Save
Compile *PGM
SEP > Refresh
CALL *PGM (or SBMJOB...)
*PGM runs to completion and debugger does not stop it
Is that correct?
So, I
then went back to the original program and recompiled under a new
program
name and that new version worked again.
So, a new *PGM? This has two implications that I can think of:
1) Never been put in debug
2) Not in QRPLOBJ library (as it would be with *REPLACE)
Again I changed that call
subroutine and it stopped loading in debug.
Several things to look at:
1) Is the original program call terminating? That sounds odd, but if
it's in a named activation group and no LR, then the program may not yet
have terminated.
2) In Preferences > IBMi Debug is the box called 'Terminate session on
program termination' selected?
3) In the RDi Debug View, when you've debugged as much as you want to
(eg you want to change the source, recompile and try again) there is a
red square (Run > Terminate). If that red square is clickable, the
original debug session is running, pointing to a program that is now in
QRPLOBJ. Click that Terminate icon, then SEP > Refresh and see if that
cures this.
midrange.com
