Re: MQSI CSD03

2002-10-10 Thread Martin Rydman

Karen,
 
this particular bug has to do with the TDS-parser being unable to correctly identify 
and locate tagged fixed length records within looped structures in a message-set. 
There is now an official APAR (IY35724 ) out on this one, pending action from IBM.
 
But there are other things. I haven't yet reported a bug (when writing out data via 
the TDS-parser) that mistakenly throws an error claiming that a mandatory field lacks 
a value, when, in fact, it doesn't, something that's easily proved by a simple trace. 
I've run into this one twice, a bug which in effect renders fully functional 
msgsets/flows under CSD02 useless under CSD03.
 
I expect to find more problems as we investigate further. I have a number of fishy 
things that I haven't yet pinned down (sometimes *we* make mistakes, can you believe 
it...), but some of them will probably turn out to be new bugs in CSD03.
 
/Martin

-Ursprungligt meddelande- 
Från: Welch, Karen S (PLC, IT) [mailto:[EMAIL PROTECTED]] 
Skickat: on 2002-10-09 20:26 
Till: [EMAIL PROTECTED] 
Kopia: 
Ämne: Re: SV: Re: SV: MQSI CSD03



Martin,

Do you know why the parsing of the input message quit working?  Can you give
any specifics?  We were hoping that CSD03 would address some memory problems
we have been having.  Is there an ETA when things might get fixed?

Karen

-Original Message-----
    From: Martin Rydman [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, October 08, 2002 8:49 AM
To: [EMAIL PROTECTED]
Subject: Re: SV: Re: SV: MQSI CSD03


Well, going to NT won't solve you're problems, I'm afraid. We upgraded to
CSD03 to take care of a problem very similar to yours, and, while it solved
that problem, the parsing of the input message stopped working properly, an
issue now confirmed by IBM.

So, maybe we should all switch to BizTalk (double hahaha)

/Martin

-Original Message-
From: Robert Broderick [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, October 08, 2002 1:38 PM
To: [EMAIL PROTECTED]
Subject: Re: SV: Re: SV: MQSI CSD03


Thank you for the reply. We are in the process of building a safe
environment. The choise to go to CSD03 was because we were having problems
going from XML to NEON SAP iDOC with repeating formats. The repeating
formats to the output node are getting dropped. The are displaying in a
trace node prior to the output node but the serialization in the output node
isn't picking them up. Also, the Properties.MessageDomain is not in the
message on the trace although we do set it to "NEONMSG". Strange!!! We have
been working with IBM for two weeks on this and still no solution. Just
suggestion that do not work. Last one was a suggestion of new features in
CSD03. Although, IBM did say they got it to work. Except it was on an NT and
we are on Solaris. Maybe we should migrate so they don't have to fix the
issue!!! (hahaha)

          bobbee


>From: Martin Rydman <[EMAIL PROTECTED]>
>Reply-To: MQSeries List <[EMAIL PROTECTED]>
>To: [EMAIL PROTECTED]
>Subject: SV:  Re: SV:  MQSI CSD03
>Date: Tue, 8 Oct 2002 09:35:13 +0200
>
>Hi!
>
>Not like the one you're mentioning in your prevoius post, but a bunch of
>others. Messagesets won't deploy, the TDS-parser seems to parse deployed
>messagesets differently, probably due to tightened validity-checks. Just
>now I got a confirmation from IBM that a message that was correctly parsed
>under CSD02 isn't under CSD03, due to a bug.
>
>So, as I said, be cautious before deploying in a prod-environ!
>
>/Martin
>
> -Ursprungligt meddelande-
> Fren: wendy.dlamini [mailto:[EMAIL PROTECTED]]
> Skickat: ti 2002-10-08 08:25
> Till: [EMAIL PROTECTED]
> Kopia:
> Dmne: Re: SV: MQSI CSD03
>
>
>
    >     Martin,
>
> Did you have anyproblem with the message sets?
> - Original Message -
> From: "Martin Rydman" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Monday, October 07, 2002 9:33 PM
> Subject: SV: MQSI CSD03
>
>
>  

Re: SV: Re: SV: MQSI CSD03

2002-10-08 Thread Martin Rydman

Well, going to NT won't solve you're problems, I'm afraid. We upgraded to CSD03 to 
take care of a problem very similar to yours, and, while it solved that problem, the 
parsing of the input message stopped working properly, an issue now confirmed by IBM.

So, maybe we should all switch to BizTalk (double hahaha)

/Martin

-Original Message-
From: Robert Broderick [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, October 08, 2002 1:38 PM
To: [EMAIL PROTECTED]
Subject: Re: SV: Re: SV: MQSI CSD03


Thank you for the reply. We are in the process of building a safe
environment. The choise to go to CSD03 was because we were having problems
going from XML to NEON SAP iDOC with repeating formats. The repeating
formats to the output node are getting dropped. The are displaying in a
trace node prior to the output node but the serialization in the output node
isn't picking them up. Also, the Properties.MessageDomain is not in the
message on the trace although we do set it to "NEONMSG". Strange!!! We have
been working with IBM for two weeks on this and still no solution. Just
suggestion that do not work. Last one was a suggestion of new features in
CSD03. Although, IBM did say they got it to work. Except it was on an NT and
we are on Solaris. Maybe we should migrate so they don't have to fix the
issue!!! (hahaha)

          bobbee


>From: Martin Rydman <[EMAIL PROTECTED]>
>Reply-To: MQSeries List <[EMAIL PROTECTED]>
>To: [EMAIL PROTECTED]
>Subject: SV:  Re: SV:  MQSI CSD03
>Date: Tue, 8 Oct 2002 09:35:13 +0200
>
>Hi!
>
>Not like the one you're mentioning in your prevoius post, but a bunch of
>others. Messagesets won't deploy, the TDS-parser seems to parse deployed
>messagesets differently, probably due to tightened validity-checks. Just
>now I got a confirmation from IBM that a message that was correctly parsed
>under CSD02 isn't under CSD03, due to a bug.
>
>So, as I said, be cautious before deploying in a prod-environ!
>
>/Martin
>
> -Ursprungligt meddelande-
> Fren: wendy.dlamini [mailto:[EMAIL PROTECTED]]
> Skickat: ti 2002-10-08 08:25
> Till: [EMAIL PROTECTED]
> Kopia:
> Dmne: Re: SV: MQSI CSD03
>
>
>
> Martin,
>
> Did you have anyproblem with the message sets?
> - Original Message -
> From: "Martin Rydman" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Monday, October 07, 2002 9:33 PM
> Subject: SV: MQSI CSD03
>
>
> Hi!
>
> We actually have a bunch of problems with CSD03. I'd strongly
>recommend you
> try it out in a safe environment first. Lots of subtle and not so
>subtle
> stuff going on, I'm afraid.
>
> /Martin Rydman
> Zipper AB, Sweden
>
> -Ursprungligt meddelande-
> Fren: Robert Broderick
>[mailto:[EMAIL PROTECTED]]
> Skickat: me 2002-10-07 20:11
> Till: [EMAIL PROTECTED]
> Kopia:
> Dmne: MQSI CSD03
>
>
>
> Anybody have problems with MQSI CSD03. We need to install
>this to
> address
> some NEON issues and are fast aproaching production
>rollout. I would
> rather
> not open up a can of worms if I can avoind it.
>
>  bobbee
>
>
>
>
>_
> Chat with friends online, try MSN Messenger:
> http://messenger.msn.com
>
> Instructions for managing your mailing list subscription
>are
> provided in
> the Listserv General Users Guide available at
>http://www.lsoft.com
> Archive: http://vm.akh-wien.ac.at/MQSeries.archive
>
>
> Instructions for managing your mailing list subscription are
>provided in
> the Listserv General Users Guide available at http://www.lsoft.com
> Archive: http://vm.akh-wien.ac.at/MQSeries.archive
>
>
>
>
>
>***
>
> This message may contain information which is confidential and
>subject to legal privilege. If you are not the intended recipient, you may
>not peruse, use, disseminate, distribute or copy this message. If you have
>received this message in error, please notify the sender immediately by
>email, facsimile or telephone and return and/or destroy the original
>message.
>
>
>**

SV: Re: SV: MQSI CSD03

2002-10-08 Thread Martin Rydman

Hi!
 
Not like the one you're mentioning in your prevoius post, but a bunch of others. 
Messagesets won't deploy, the TDS-parser seems to parse deployed messagesets 
differently, probably due to tightened validity-checks. Just now I got a confirmation 
from IBM that a message that was correctly parsed under CSD02 isn't under CSD03, due 
to a bug.
 
So, as I said, be cautious before deploying in a prod-environ!
 
/Martin

-Ursprungligt meddelande- 
Från: wendy.dlamini [mailto:[EMAIL PROTECTED]] 
Skickat: ti 2002-10-08 08:25 
Till: [EMAIL PROTECTED] 
Kopia: 
Ämne: Re: SV: MQSI CSD03



Martin,

Did you have anyproblem with the message sets?
- Original Message -
    From: "Martin Rydman" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, October 07, 2002 9:33 PM
Subject: SV: MQSI CSD03


Hi!

We actually have a bunch of problems with CSD03. I'd strongly recommend you
try it out in a safe environment first. Lots of subtle and not so subtle
stuff going on, I'm afraid.

/Martin Rydman
Zipper AB, Sweden

-Ursprungligt meddelande-
Från: Robert Broderick [mailto:[EMAIL PROTECTED]]
Skickat: må 2002-10-07 20:11
Till: [EMAIL PROTECTED]
Kopia:
Ämne: MQSI CSD03



Anybody have problems with MQSI CSD03. We need to install this to
address
some NEON issues and are fast aproaching production rollout. I would
rather
not open up a can of worms if I can avoind it.

 bobbee



_
Chat with friends online, try MSN Messenger:
http://messenger.msn.com

Instructions for managing your mailing list subscription are
provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive


Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive




***

This message may contain information which is confidential and subject to 
legal privilege. If you are not the intended recipient, you may not peruse, use, 
disseminate, distribute or copy this message. If you have received this message in 
error, please notify the sender immediately by email, facsimile or telephone and 
return and/or destroy the original message.

***

Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive


Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive



SV: MQSI CSD03

2002-10-07 Thread Martin Rydman

Hi!
 
We actually have a bunch of problems with CSD03. I'd strongly recommend you try it out 
in a safe environment first. Lots of subtle and not so subtle stuff going on, I'm 
afraid.
 
/Martin Rydman
Zipper AB, Sweden

-Ursprungligt meddelande- 
Från: Robert Broderick [mailto:[EMAIL PROTECTED]] 
Skickat: må 2002-10-07 20:11 
Till: [EMAIL PROTECTED] 
Kopia: 
Ämne: MQSI CSD03



Anybody have problems with MQSI CSD03. We need to install this to address
some NEON issues and are fast aproaching production rollout. I would rather
not open up a can of worms if I can avoind it.

 bobbee



_
Chat with friends online, try MSN Messenger: http://messenger.msn.com

Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive


Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive



Checking in/out in WMQI crawls

2002-10-01 Thread Martin Rydman

Hi all!
 
After re-installing CSD03 on NT (rolled back to CSD02 once) checking in/out in the 
Control center is slow. Thing is, nothing happens in the computer. CPU usage 
spikes for a bit, then nothing. CPU is idle, no disk-swapping, nothing. Seems like 
some time-out occurs, because after a long wait, the operation complets without errors.
 
Any one run into this and solved it? I could always reinstall the whole package, but 
I'd rather not.
 
Any hints appreciated!
 
TIA
 
/Martin Rydman
 
 

Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive



Re: ESQL-deleteing dynamic ref

2002-07-30 Thread Martin Rydman

Peter,
 
indeed it does. I think I managed to set root=NULL after detaching it (don't have the 
code any more) and then somehow it didn't work (or maybe something else, it's pretty 
mindboiling hot here in Sweden today...). Upon receiving your mail, I retried Detach, 
and it works just fine
 
Thanks!
 
/Martin

-Original Message- 
From: Peter Cradwick [mailto:[EMAIL PROTECTED]] 
Sent: Tue 2002-07-30 11:56 
To: [EMAIL PROTECTED] 
Cc: 
Subject: Re: ESQL-deleteing dynamic ref



Martin,
I would have thought Detach would do it...
Mind you, I have only done this using the C PlugIn API when creating an
output message from an input message. Same procedure basically: navigate to
the element to remove ( start of the 'input' data) and feed it's pointer to
cniDetach() - works fine!
Peter




    "Martin Rydman"

  To: MQSERIES
Sent by:
"MQSeries List"   cc:



30-Jul-2002
09:56
Please respond
to "MQSeries
List"






Hi all!

I've got some code that identifies subtrees in a XML-message that contain
only empty elements. It's a recursive procedure using dynamic REFERENCE
variables. Here's a snippet of my code:


CREATE PROCEDURE RemoveEmptySegments (INOUT root REFERENCE)
BEGIN
 DECLARE DCntCursor REFERENCE TO root;
 DECLARE DataCount INTEGER;
 DECLARE ClimbThisNode BOOLEAN;
 SET ClimbThisNode = TRUE;
 IF LENGTH(TRIM(FIELDNAME(root))) = 3 THEN -- This must be a segment name
  SET DataCount = 0;
  CALL FindSegmentData(DCntCursor, DataCount);
  IF DataCount = 0 THEN
   SET ClimbThisNode = FALSE;
   SET root = NULL; -- <<<<<<<<<<<<<<<<<<< This won't work!
  END IF;
 END IF;

 IF ClimbThisNode THEN
  DECLARE cursor REFERENCE TO root;
  MOVE cursor FIRSTCHILD;
  WHILE LASTMOVE(cursor) DO
   CALL RemoveEmptySegments(cursor);
   MOVE cursor NEXTSIBLING;
  END WHILE;
 END IF;
END;


Now, since the root variable only references a subtree in the message-tree,
setting it to NULL only dereferences root (I'd presume, haven't verified
it). The message-tree, however, is not affected. I have verified that the
logic finds empty subtrees, because replacing SET root = NULL with SET root
= 'nod data' will show up in a trace node as expected.

I've tried DETACH root, but to no avail. So, how do I remove a subtree from
my message-tree, given a REFERENCE-variable pointing to the subtree?

TIA!

/Martin

Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive

Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive


Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive



ESQL-deleteing dynamic ref

2002-07-30 Thread Martin Rydman

Hi all!
 
I've got some code that identifies subtrees in a XML-message that contain only empty 
elements. It's a recursive procedure using dynamic REFERENCE variables. Here's a 
snippet of my code:
 

CREATE PROCEDURE RemoveEmptySegments (INOUT root REFERENCE)
BEGIN
 DECLARE DCntCursor REFERENCE TO root;
 DECLARE DataCount INTEGER;
 DECLARE ClimbThisNode BOOLEAN;
 SET ClimbThisNode = TRUE;
 IF LENGTH(TRIM(FIELDNAME(root))) = 3 THEN -- This must be a segment name
  SET DataCount = 0;
  CALL FindSegmentData(DCntCursor, DataCount);
  IF DataCount = 0 THEN
   SET ClimbThisNode = FALSE;
   SET root = NULL; -- <<< This won't work!
  END IF;
 END IF;

 IF ClimbThisNode THEN
  DECLARE cursor REFERENCE TO root;
  MOVE cursor FIRSTCHILD;
  WHILE LASTMOVE(cursor) DO
   CALL RemoveEmptySegments(cursor);
   MOVE cursor NEXTSIBLING;
  END WHILE;
 END IF;
END;

 
Now, since the root variable only references a subtree in the message-tree, setting it 
to NULL only dereferences root (I'd presume, haven't verified it). The message-tree, 
however, is not affected. I have verified that the logic finds empty subtrees, because 
replacing SET root = NULL with SET root = 'nod data' will show up in a trace node as 
expected. 
 
I've tried DETACH root, but to no avail. So, how do I remove a subtree from my 
message-tree, given a REFERENCE-variable pointing to the subtree?
 
TIA!
 
/Martin

Instructions for managing your mailing list subscription are provided in
the Listserv General Users Guide available at http://www.lsoft.com
Archive: http://vm.akh-wien.ac.at/MQSeries.archive