MCH4227 rc22 (was: CPF4227 Reason 22 on Program Call)

The opening and next post of this thread seem not to have been received via NNTP. Hopefully the above correction to the subject line appears in the archives as it would in NNTP threading... and along with that, the following information may assist someone in the future.

Anyhow the error was in symptom keyword string format: msgMCH4227 RC22; per lack of a joblog, the context of the error with regard to messaging is not available, and the description of the scenario [also] does not include the request [message] being made; the subject said "program call", but the "resolution" would imply possibly a "procedure call" from a service program. Apparently, per the description of "resolved", the error was manifest when the system was trying to re-translate the referenced down-level IBM service program... but the joblog was not given, so that is not obvious, if in fact that was the case.

As Jon suggests in another reply, copying IBM code to a user library is generally inappropriate [and often unsupported]. Yet like all /changes/ to a system [i.e. customizations], when doing so is supported and\or functional, one must remember also to implement proper System Change Management, to ensure that the same copy\change activity is repeated again after every upgrade and after any maintenance that might have replaced\updated that copied code. Unlike user code for which the system intends the code to remain functional on a later release than where the code was compiled, the IBM code often does not effect that same support for itself; i.e. unless explicitly designed to enable /skip-ship/ support, IBM code is expected to be replaced by an install\upgrade, to match the current release that is installed. Down-level IBM code is necessarily expected to fail re-translation.

Regards, Chuck

On 20-Feb-2014 08:56 -0800, Michael Ryan wrote:

Problem resolved. Here's the issue. I had an old copy of
QHTTPSVR/QZHBCGI in a user library. The QHTTPSVR version of QZHBCGI
had this:

Conversion required . . . . . . . . . . . . . . : *NO
Conversion detail . . . . . . . . . . . . . . . : *COMMON

The copy I had in my user library had this:

Conversion required . . . . . . . . . . . . . . : *YES
Conversion detail . . . . . . . . . . . . . . . : *FORMAT

so it was in the wrong format for my version of the OS. I put
QHTTPSVR higher in my library list so the IBM object would be
grabbed. Program created fine and executed fine.

On Thu, Feb 20, 2014 at 11:05 AM, Michael Ryan wrote:

Ok...I wonder if it's combining an old service program with newer
techniques like like_ds. The service program is from 2008...didn't
have that then, right?

On Thu, Feb 20, 2014 at 10:16 AM, steveg wrote:

I think I had this once before. It was because I was compiling a
program back-level and something on a D or C spec wasn't
supported in that release.

Jon Paris on 20/02/2014 14:51 wrote:

Usually it means what it says Michael - this is a compiler
problem and you have either discovered a new problem or are a
few PTFs short of a full load.

There is rarely anything a mere user can do that would throw an
error of this type.

On 2014-02-20, at 9:28 AM, Michael Ryan wrote:

And here's the error message:

Message ID . . : MCH4227 Severity . . : 40
Message type . : Escape
Date sent . . : 02/20/14 Time sent . : 09:11:18
Message . . . . : Instruction stream not valid.
Cause . . . . . : The object was not created because the
sequence of instruction stream objects is not valid.
Recovery . . . : There is a problem in the compiler. Report
this problem to the supplier of the compiler.
Technical description . . . : The reason code is 22.
The current instruction count is 75. The offset to the
instruction stream data that is not valid is 660 bytes.
The offset to the current instruction is 640 bytes and the
operation code of the current instruction is X'00000061'.
The current statement number is 23. The LIC log note ID is
01003648.

>>>> On Thu, Feb 20, 2014 at 9:27 AM, Michael Ryan wrote:

I did some Googling on this, but I'm not sure how to find
out the problem. I think it's a mismatch with an old
version of a service program (old code) and new code, but
I'm not sure how to determine where I'm having the problem.
Any ideas on how to research this?