RE: Mysterious MCH3601 errors at V4R3M0

["Leland, David" <dleland@xxxxxxxxxx>]

Title: RE: Mysterious MCH3601 errors at V4R3M0

Well, that could explain my problem.  I'm using CEEDOD and only passing in the 6 required parameters.  Do I need to pass in *OMIT for the seventh parameter which is omissible?

Thanks,
Dave

----------
From:   David Morris[SMTP:dmorris@plumcreek.com]
Reply To:       MIDRANGE-L@midrange.com
Sent:   Tuesday, August 10, 1999 11:15 AM
To:     MIDRANGE-L@midrange.com
Subject:        RE: Mysterious MCH3601 errors at V4R3M0

Hans,

Although this doesn't anser Don's question, I have seen this error
caused by not passing *OMIT to the CEETSTA API.  I know this can
type of error can be caused by simply not passing a parameter for
any CEE API parameter that specifies that it is omissible.  To
correct this error you must specify *OMIT, even if it is the last parameter
passed.  Even the people who wrote the sample code for the RPG
manuals incorporated this bug. Failure to pass *OMIT can cause
memory to be corrupted in the most unlikely places (experience tells
me x'0000' may pop up in user spaces or static program variables.)

David Morris

>>> <boldt@ca.ibm.com> 08/10/99 08:17AM >>>




Don wrote:
>It also doesn't explain why, when converted from RPGIV to RPGIII, it
>magically starts working.  If there is no bug in the RPGIV compiler, as Hans
>claims, why does this solution work?

Sure it explains it.  Some program in the application is
corrupting some area of storage.  By changing the affected
program (not the buggy program) to an RPG III program, the
corruption is happening in some storage which does not
manifest itself in the same way.  The problem is still
there, lurking quietly, waiting for the opportunity to pop
up again when some other condition changes.

As I stated before, we've seen several reports of
"mysterious MCH3601" problems and they've all turned out to
be bugs in the application, not in the ILE RPG compiler or
run-time.  I'm not trying to deflect criticism away from
RPG, but our experience tells us that the most likely
source of the problem is somewhere in the application.  I
know from my own experience, that this type of problem is
the hardest to debug.

Cheers!  Hans

Hans Boldt, ILE RPG Development, IBM Toronto Lab, boldt@ca.ibm.com




+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---