########################################################################
# #
# COMPONENT_NAME: WebSphere MQ Integrator #
# #
# FILE: MEMO.PTF #
# #
# Program Number 5724-A82 #
# (C) COPYRIGHT International Business Machines Corp. 1999, 2001 #
# All Rights Reserved #
# Licensed Material - Property of IBM #
# #
# US Government Users Restricted Rights - Use, duplication or #
# disclosure restricted by GSA ADP Schedule Contract with IBM Corp. #
# #
########################################################################
# #
# PLEASE REVIEW THE README.TXT FILE WHICH HAS BEEN UPDATED IN THIS #
# MAINTENANCE RELEASE. #
# #
########################################################################
# #
# The WebSphere MQ Integrator V2.1 documentation has been updated #
# to the CSD02 level. The March 2002 levels are available from #
# the following Web page on the MQSeries Web site: #
# #
# http://www.ibm.com/software/ts/mqseries/library/manualsa/ #
# #
########################################################################
# #
# The licence information for this product has been updated to add #
# Microsoft (R) Data Access Components and Hewlett Packard Java RTE #
# for HP-UX licence information. A copy of these terms and conditions #
# can be found in the product directory after installation. #
# #
# (R) Microsoft is a registered trademark of Microsoft Corporation #
# in the United States, other countries, or both. #
# #
########################################################################
PTF Memo for WebSphere MQ Integrator Version 2.1 for Sun Solaris
----------------------------------------------------------------
The memo.ptf file contains information you need for updating WebSphere
MQ Integrator Version 2.1 for Sun Solaris. This memo.ptf file
corresponds to CSD05 which is defined as PTF U487641 and is divided
into the following sections:
o Installation and maintenance information
o Documentation changes
o Special information
o New Era of Networks information
o Service and technical support
o CSD history
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
INSTALLATION AND MAINTENANCE INFORMATION
----------------------------------------
In order to restore a previous level you must backup replaced files.
CSD05 requires approximately 110MB of disk space in the
/var/sadm/pkg directory to save the previous level.
The old files are backed off and stored at location
/var/sadm/pkg/mqsi-updn.
APPLYING MAINTENANCE
____________________
Maintenance updates are supplied on CD in the form of a Program
Temporary Fix (PTF), referred to as a Corrective Service Diskette
(CSD). You can find the latest information about available CSDs
on the Internet, at this address:
http://www.ibm.com/software/ts/mqseries/
You can also download CSDs from this web site.
APPLYING THE MAINTENANCE INFORMATION
____________________________________
The instructions for applying and removing maintenance can be found
in Appendix D (Applying Maintenance) of the Websphere MQ Integrator
for Sun Solaris Installation Guide (GC34-5842-04) but must be amended
by any changes documented below.
APPLYING NEW ERA OF NETWORKS MAINTENANCE INFORMATION
____________________________________________________
Ensure that all brokers are stopped before installing maintenance
and only restart them once installation is complete.
From the NNSY directory of the CSD image enter the command:
pkgadd -d nnsy
REMOVING NEW ERA OF NETWORKS MAINTENANCE INFORMATION
____________________________________________________
To uninstall New Era of Networks Rules and Formatter Support for
WebSphere MQ Integrator on Sun Solaris, use the pkgrm command for each
New Era component. For example:
pkgrm n0infr21
Repeat this for each NNSY package. List all the packages using :-
pkginfo | grep NNSY.
Alternatively mount the WMQIinstallation CD, change directory
to /cdrom/wmqi_solaris/NNSY/nnsy and run the pkgrm command as follows:
pkgrm *
To uninstall a CSD level, use the command:
pkgrm n0infr21.n
where n is the update level and repeat for each component.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
DOCUMENTATION CHANGES
---------------------
IC36010 - DOCUMENT USE OF CLEARMESSAGE FOR JAVA PLUGIN NODES
Use of the Java plug-in method clearMessage is not documented. Not
using this method can lead to growth in memory usage by the broker.
The WebSphere MQ Integrator Programming Guide, SC34-5603-03, Chapter
6; Implementing a plug-in node or parser, section; Implementing a
node in Java, subsection; Building an output message, will have a
new subsection entitled Propagating the message as follows:
Propagating the message
Before you propagate a message, you have to decide what message flow
data you want to propagate, and which of the node's terminals is to
receive the data. You should finalize the message before you
propagate it. After propagating a message, you must delete the output
message. For example:
1. To propagate the message to the output terminal "out":
MbOutputTerminal out = getOutputTerminal("out");
out.propagate(outAssembly);
2. To delete the output message:
outMessage.clearMessage();
Any other messages created during the processing of the node should
also be deleted in a similar manner.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
PQ67615 - CNIELEMENTCHARACTERVALUE/CPIELEMENTCHARACTERVALUE CAUSING 0C4
When using either cpiElementCharacterValue or cniElementCharacterValue
using an incorrect value in the length value can result in 0C4 abends
or in overwrites of user storage. The Programming Guide specifies
that the length parameter of these calls should have units of bytes.
This is incorrect; the length should be provided in units of CciChar.
The Programming Guide (SC34-5603-03) will be changed in Chapters 8
and 9 so that the following functions are now documented as taking
CciChar as the unit for the length parameter:
cniElementCharacterValue
cniElementDecimalValue
cniElementName
cpiElementCharacterValue
cpiElementDecimalValue
cpiElementName
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
PQ72256 - BUILDING A C++ USER EXIT LIBRARY, CUSTOMER MUST USE THE
CBCDRVR COMPILER (VERSIONS 2.8 - 2.10) ONLY
The New Era of Networks Rules and Formatter Support for WebSphere MQ
Integrator System Management Guide Version 5.6, SC34-6083-00,
Chapter 5, p.134, section entitled Building a C++ User Exit Library
should be added to as follows:-
The CBCDRVR compiler, versions 2.8 - 2.10, must be used.
When using compilers later than version 2.10 (aka V1R1), up to and
including V1R4 the V2R10 CBCDRVR shipped with their compiler for
backward compatibility can be used. (See C/C++ Compiler and Run-Time
Migration Guide for details relating to compilers post V2R10.)
Generally setting the _CXX_CNAME environment variable and including
the SCBCCMP library your STEPLIB, will pick up the CBCDRVR compiler.
For example, set the following in your USS users profile :-
export _CXX_CNAME="CBCDRVR"
STEPLIB=$STEPLIB:PP.CBC.ZOS120.SCBCCMP
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC36384 - IMPLEMENTATION OF THE NNSY REPEATING ATTRIBUTE TO FLAG
REPEATING FORMATS IN ESQL.
Implementation of the NNSY repeating attribute to flag repeating
formats in ESQL is available but was not documented in WebSphere
MQ Integrator Working with Messages Version 2.1.
The following section, which includes the description of the NNSY
repeating attribute, should have appeared in the WebSphere MQ
Integrator Working with Messages Version 2.1, Chapter 12.
The New Era of Networks domains. The section entitled:
'Transforming from XML and MRM to NEONMSG' should be entirely
replaced and re-titled:
Processing messages in the NEONMSG domain
-----------------------------------------
You can use the facilities provided by the message processing nodes
and the brokers to transform an input MRM or XML message to an output
NEONMSG message or to manipulate a message in the NEONMSG domain.
However, you must be aware that the NEONMSG parser does not recognize
repeating fields in these messages, unless you explicitly flag their
containing structure (format) as repeating: if the input message
includes repeating fields, and you do not flag them as such, the
output message generated will not accurately reflect the input.
The only node that can determine your input message repeat structure
by direct reference to the NNSY Formatter database is the MQInput
node on the initial parse of a message in the NEONMSG domain. This
is because the functionality provided in a message flow means you can
potentially deviate from message structure encoded in the Formatter
database.
In circumstances where you need to manipulate a message in the NEONMSG
domain or transform your message to this domain, in WMQI 2.1.0, you
can manually edit the message to insert the attribute
NNSY-XML-repeating="true" to each repeating collection.
(In MQSI 2.0.2 you must use NEON-XML-repeating="true".)
For example, an input message might contain:
IBM
1234-L
Large Widget
1234-M
Medium Widget
1234-S
Small Widget
This must be manually edited to contain:
IBM
1234-L
Large Widget
1234-M
Medium Widget
1234-S
Small Widget
An example of the ESQL needed might be:
SET "OutputRoot"."Properties"."MessageSet" = 'NEONMESSAGESET';
SET "OutputRoot"."Properties"."MessageType" = 'PurchaseOrder_MT';
SET "OutputRoot"."MQMD" = "InputRoot"."MQMD";
SET "OutputRoot"."MQMD"."Format" = 'MQSTR';
DECLARE K INT;
DECLARE N INT;
SET "OutputRoot"."NEONMSG"."PurchaseOrder_OC"."CustomerName_OF".
"CustomerName"="InputRoot"."XML"."PurchaseOrder"."CustomerName";
SET K = CARDINALITY,("InputRoot"."XML"."PurchaseOrder"."OrderDetail"[]);
SET N = 0;
WHILE N < K DO
SET N=N+1;
SET "OutputRoot"."NEONMSG"."PurchaseOrder_OC". "OrderDetail_OC"[N].
"NNSY-XML-repeating" = 'true';
SET "OutputRoot"."NEONMSG"."PurchaseOrder_OC"."OrderDetail_OC"[N].
"OrderDetail_OF"."PartNum"="InputRoot"."XML"."PurchaseOrder".
"OrderDetail"[N]."PartNum";
SET "OutputRoot"."NEONMSG"."PurchaseOrder_OC"."OrderDetail_OC"[N].
"OrderDetail_OF"."PartDesc"="InputRoot"."XML"."PurchaseOrder".
"OrderDetail"[N]."PartDesc";
END WHILE;
An example of corresponding NNSY Output Formats:
Format/PurchaseOrder_MT/Ordinal
|
Format/PurchaseOrder_OC/Ordinal
|
Format/CustomerName_OF
| |
| Field/CustomerName/1/Normal Access/CustomerName
| |
| - Output Control/CustomerName/Data Field(Name Search)
| |
| - Data/String
|
Format/OrderDetail_OC/Ordinal/Repeating/3
|
Format/OrderDetail_OF
|
Field/PartNum/1/Controlling Field/PartNum
| |
| - Output Control/PartNum/Data Field(Name Search)
| |
| - Data/String
|
Field/PartDesc/2/Access Sibling Instance/PartDesc
|
- Output Control/PartDesc/Data Field(Name Search)
|
- Data/String
Note: In the last example the tree structure of the formats must be
adjusted, regardless of business logic, by comparison to the input
message, because formats that contain formats themselves (Compound
formats) cannot contain both field references and format references
in the NNSY data model.
Note: The use of the repeating flag is pertinent, not only to cases
where you are transforming from an XML or MRM domain but also if you
intend to manipulate the NEONMSG tree (even though you may not have
explicitly left the NEONMSG domain). For example, if you process
your message through a Compute Node, the tree will again have to have
the repeating attribute flagged for all instances of repeats.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC35011 - IMPROVE DOCUMENTATION REGARDING NULL HANDLING
The documentation regarding how WebSphere MQ Integrator handles NULL
processing will be be updated to more clearly outline and explain the
process as follows :-
1. Websphere MQ Integrator Working With Messages, SC34-6039-01,
Chapter 4, The MRM logical message model; Section, The broker's view
of the logical message model; Sub-Section, Using ESQL to manipulate
the message tree; Null Handling will have the following information
added at the start of the section :-
Parser overview with Null Handing
---------------------------------
In a WMQI message flow, there are the concepts of a message bitstream
and a logical message tree. The logical message tree shields the user
from the physical bitstream by providing fields that can be
manipulated without having to consider the exact sequence of bytes.
In WMQI parsers are used to convert a message bitstream into a message
tree and vice versa. Each parser uses its own 'algorithms' to dissect
the bitstream and then create the appropriately named fields in the
message tree giving them the relevant field types and field values.
For the XML parser these algorithms involve identifying XML markup
language and then using these to separate this into the correct XML
objects. The MRM parser has different ways of parsing a bitstream
and these are represented by the physical layers in a messageSet.
For CWF, this involves reading a set sequence of bytes and translating
these into the fields value in the message tree. For TDS, the parser
uses the Data Element Separation method to parse the bitstream.
Depending on what has been set this could involve identifying
delimiters, tags, fixed length elements or patterns etc.
The parsers are also responsible for writing a new message bitstream
based on the values in the message tree. This will occur when a
bitstream representation is required, such as in the MQOutput node or
when calling the ESQL ASBITSTREAM function.
Parsing and writing can be mutually independent operations such that
they both do not have to be performed in a message flow. On this
basis, it is possible for users to manually create new message trees
in a compute node and then the parsers will attempt to create a
bitstream from this message tree. It is also very common that the
parsers will receive bitstreams that they did not create and have to
attempt to create a message tree from them.
The ESQL SET and CREATE statements can be used to populate a message
tree field with a Name, Type and Value. It is therefore possible that
the Value assigned to the message tree can be the result of an
expression that evaluates to a NULL. This can be caused by the use of
an uninitialised variable, a message tree field that does not exist,
the result of a database query or use of the NULL keyword directly.
When dealing with a message tree, there are two types of NULL :
- Implicit : A message tree field does not exist and so implicitly
has a NULL value.
- Explicit : A message tree field actually contains a NULL value
through the use of the SET with VALUE keyword or
similar ESQL.
When a message tree field has an explicit NULL value then this will be
interpreted in different ways depending on the owning parser of the
message tree field, the operations performed by the parser and the
ESQL operations performed on the message tree field.
Not all of the WMQI parsers have 'NULL handling' capabilities, but for
those that do the following is possible :
- A sequence of bytes can be identified in the message bitstream that
will cause a NULL value to be created in the related message tree
field.
- An explicit NULL value can be accepted in the message tree and this
will result in a set sequence of bytes to be written into the
message bitstream.
When explicit NULL handling is considered as part of these parser
operations then :
- NULL is not a consistent concept.
- In WMQI the concept of NULL is present in the logical model and is
created in the variables within the message flow. There is no
notion of a NULL value in a bitstream. The NULL value in the
message tree is created when a specific sequence of bytes are
encountered within the bitstream. This sequence of bytes will be
different for each parser.
- Each parser has its own representation of a NULL in a bitstream.
In the case of the MRM, this representation can be different for
each element if the user requires such behaviour.
- Each parser will interpret a NULL value in a message tree field
differently.
In summary, some WMQI parsers can lead to a NULL value to be created
in the message tree, and can handle a message tree field containing an
explicit NULL value when writing a new output message. However, other
operations can be performed on message tree fields outside the scope
of parsing and writing. It is possible for a user to create or
manipulate a parser created message tree through the use of ESQL or
nodes that cause message tree fields to be copied from one domain to
another. For more information regarding the use of explicit NULLs in
the XML domain, please see the ESQL reference (as point 2 below).
2. Websphere MQ Integrator ESQL Reference, SC34-5923-02, Chapter 2;
Messages and ESQL; Section, Referencing information in XML messages;
will have a new subsection added after 'Constructing XML in the
Compute Node' as follows :
The XML message domain and NULL values
--------------------------------------
The XML domain does not have a concept of a NULL value in the message
bitstream. This means that there is no sequence of bytes in an XML
message that will lead to an explicit NULL value to be created in the
message tree. Although the XML parser will not create a message tree
field with an explicit NULL value, ESQL within a message flow can
lead to the creation of such a field. When the XML parser needs to
construct a new message bitstream from the message tree, then it
could encounter explicit NULL values and therefore needs to be able
to handle them. In this case, an explicit NULL value is seen as 'no
characters to write' and therefore the field is created as the empty
tags. In the XML domain, there are several values that will lead to
the writing of XML empty tags in the resultant XML message :
- A message tree field that contains an explicit NULL value
- A message tree field that contains the zero length string ('')
- A message tree field that does not have a value
Once these have been written as the empty tags in an XML message then,
if the XML message is subsequently parsed, it is not possible to
distinguish between these three cases. When empty tags are parsed in
XML, then an XML message tree field is created that does not have a
value. This is different from a message tree field that contains an
explicit NULL value, and also different from a message tree field that
contains a zero length string.
Although the XML parser resolves these three different message tree
field values to the same result in the message bitstream, other
message tree operations can distinguish between them. Although the
XML parser will not create fields that contain explicit NULLS and zero
length strings, the user can perform operations in message flows that
can. Therefore, if a message flow is processing messages in the XML
domain, then the ESQL will need to distinguish between a explicit NULL
value, a zero length string, and a message tree field that does not
have a value. Not even a field that contains the explicit NULL value
is considered to be NULL when using ESQL to query the value of the
element.
The XML writing process can handle this condition however other
operations within ESQL will not treat this as a NULL value. As
previously stated, the XML domain does not have a concept of NULL and
the parser will never create a field that contains an explicit NULL
value. Because the XML domain is a character based domain, when the
value of a field is queried, it returns the result as a character
field. If the user creates an XML message tree field with an explicit
NULL, then this will not be seen as a NULL value for those ESQL
routines that get the value of the message tree field. What will
actually be returned is the character representation of the contents
of the field, which in this case will be the character string 'NULL'.
For example, if the user performs such comparisons as with IS NULL,
then this will never evaluate to a TRUE value. If the user copies
this message tree field to another message tree, then the target
message tree field will be populated with the character string 'NULL'.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC35575 - RENAME MSGFLOW FOR DEVELOPMENT & PRESERVE ORIGINAL COPY
A new section; Preserving message flows, will be added to the Using
the Control Center manual, SC34560205, in Chapter 5; Working with
message flows, before the Debugging message flows section to improve
the documentation regarding preservation of message flows as follows:-
If you wish to preserve a message flow to enable reuse or retain a
stable level, whilst further developing a copy it is important to
understand that renaming a message flow merely changes a label.
Message flows are identified in the configuration repository by a
generated unique identifier (CUUID). Therefore when a message flow
is exported from configuration repository to another it will retain
its CUUID. This means that if an exported message flow is imported
and modified on another system then subsequently re-imported to the
original system, the original message flow, even if it has been
renamed, will be overwritten with the imported modified message flow
unless a copy has been made of the original.
Preserving a copy of a message flow:
1. Click the Message Flows tab.
2. Select the message flow you wish to preserve, and click Copy.
3. Paste the message flow to the Message Flows root.
4. A copy of the original message flow is created as a new message
flow (this will have a different CUUID). Select the copy and
click Rename to change the label of the message flow.
5. You can now modify the renamed copy of the message flow independent
of the original.
6. Save your new message flow in the configuration repository using
either Check In or one of the File -> Check In menu options.
A reference with recommendation to read this will be made in Chapter
3, Using the Control Center - configuration data and the workspace,
in the 'Importing resources' section.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
PQ71204 - UPDATE TO THE PROGRAMMING GUIDE ON HANDLING ERRORS RETURNED
FROM THE JAVA PROPAGATE METHOD.
The following following paragraph will be added to the WebSphere MQ
Integrator Version 2.1 Programming Guide (SC34-5603-04) in Chapter 6,
the section 'Implementing a node in Java', sub-section 'Error
handling and exception' to document the exception handling for the
Java propagate method.
When an error occurs downstream of the plugin node the exception will
get flowed back up the flow and returned to the evaluate method of the
plugin. As the broker is handling this execption it is better not to
handle this exception within the plugin. To do this do not code the
propagate method within a try code block. If you wish to carry out
some error handling then once complete the exception should be thrown
back to the broker to allow it to complete its error handling.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IY36091 - DOCUMENTATION PROBLEM WITH HOW MESSAGES ARE PROCESSED IN BOOK
The WebSphere MQ Integrator Introduction and Planning Guide V2.1,
GC34-5599-04, Chapter 5: Messages, Section: How messages are processed
in a message flow the bulletted list will be changed to describe the
correct way in which messages are processed in a message flow as
follows:-
The characteristics of MQSeries messages are identified by the input
node of a message flow in the following way:
- The input node will allocate a parser to each header in the message
in turn, and the message body is simply treated as another header
in the chain.
- Being an MQSeries message, the first header is assumed to be the
MQMD and is parsed accordingly.
- Each header should indicate the type of header or data following it.
With MQSeries messages, this is normally done using the 'Format'
field. The MQRFH and MQRFH2 headers are slightly different in that
the Name/Value data in the header may also contain information
regarding the format of the following data. If the value in the
Format field is a known parser (recognised by WQMI) then this will
always take precedent over any Name/Value data.
See the WebSphere MQ Integrator Programming Guide for more details
about the content and use of these headers.
- If the value in the "Format" field (or in the Name/Value data) is
recognised by the input node, and it has an associated parser, this
parser will be used to parse the following header (which may be the
message body).
- If no specific parser has been identified by the 'Format' field, the
input node uses the default message template, defined as a property
of the input node, to determine how the remaining message body must
be parsed.
- If a default message domain hasn't been specified on the input node,
the BLOB parser is used.
It is normal practice to use either the folder in the MQRFH2
header or the MQInput node to indicate to the message flow the
characteristics of the message data.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IY37844 - NNSYMESSAGELOG MUST BE ACCESSIBLE BY THE BROKER AND IN THE
CURRENT DIRECTORY.
Chapter 12: Migrating from MQSeries Integrator of the WebSphere MQ
Integrator Administration Guide V2.1, GC34579206, Section: Migrating
New Era of Networks Rules and Formats, Subsection: Logs and log
records will be added to to specify that WebSphere MQ Integrator
must have access to the New Era Messagelog file as follows :-
Log records generated by all New Era of Networks Rules and Formatter
Support nodes and parsers are written to the log file defined for the
message flow in which the nodes appear. For details of how to specify
a log file for a message flow, see WebSphere MQ Integrator Using the
Control Center.
The NEONMap, NEONRulesEvaluation, and NEONTransform nodes also write
log information to the NNSYMessageLog.nml file. This file is created
in the directory from which you issue the mqsistart command
> and hence MQSeries Integrator must have write access to that
> current directory.
> The NEONMap, NEONRulesEvaluation, and NEONTransform nodes also write
> log information to the NNSYMessageLog.nml file. This file is
> created in the directory from which you issue the
> mqsistart command.
Note that the New Era log file is named NEONMessageLog.nml in MQSeries
Integrator V2.0.2.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IY37824 - WMQI ADMINISTRATION GUIDE IS MISSING DOCUMENTATION FOR
MQSI_SETUPDATABASE SUPPORTING ORACLE9
The Websphere MQ Integrator Administration Guide V2.1, SC34-5792-05,
Chapter 3: The broker, Topic: Creating a broker database, Section:
Using Oracle or Sybase databases on UNIX platforms, will be updated to
include instructions for creating an Oracle 9i database as follows :-
In the SubSection: For Oracle 8.1.6 the command to issue to perform
the required setup for Oracle 8.1.6 will be changed from:
mqsi_setupdatabase oracle
to:
mqsi_setupdatabase oracle8
where is the directory in which Oracle
8.1.6 is installed.
The SubSection: For Oracle 8.1.7 and Sybase will be changed to read
as follows:-
For Oracle 8.1.7, Oracle 9 and Sybase
You are recommended to complete this setup after you have installed
your database, but you can complete it before database installation
if you specify the intended database installation directory correctly
during this setup.
You must issue the following command:
mqsi_setupdatabase
where is
oracle8,
oracle9,
sybase 11,
sybase12, or
sybase12_5
depending on your database version and
is the directory in which the database is (or will be) installed.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC35769 - DISCEPANICES AROUNDTHE COMPOUND STATEMENT AND CASE FUNCTION.
The ESQL CASE statement function is incorrectly documented in the
WebSphere MQ Integrator ESQL Reference, SC34592302, the WebSphere MQ
Integrator CSD Release Guide, SC34619700, and the product README.TXT
file. There is also an implication that there is an ESQL keyword/verb
called "COMPOUND", which there is NOT. Future versions of the
documentation will correct this. The README.TXT file which
accompanies this maintenance has been updated. See point 95. ESQL
CASE Statements, in The All Operating Systems section.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
PQ66774 - ASBITSTREAM FUNCTION IS NOT DOCUMENTED IN THE ESQL GUIDE FOR
WMQI 2.1.
The ASBITSTREAM function was not documented in the ESQL Guide or the
README.TXT file which accompanied CSD03. The information will be
included in any future release of the Guide and is included in the
README.TXT file which accompanies CSD04.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IY35159 - PROBLEM DETERMINATION GUIDE ON PAGE 39 STATES THAT HARD LIMIT
FOR PROCESS ON AIX IS 512MB, HOWEVER IT IS USER DEFINED.
The AIX hard limit size described in the WebSphere MQ Integrator
Problem Determination Guide V2.1, GC34592000, Chapter 5, Dealing with
problems during administration, Section: Configuration problems with
a large number of components will be corrected and added to as
follows :-
AIX 4.3.3 and 5.1
For 32 bit system : 2 GB
On AIX , WMQI limits DataFlowEngines to two 256MB memory segments i.e.
512MB. Normally DFE processes do not require more than 512MB of the
AIX address space therefore the DataFlowEngine executable is linked so
as to have a default address space of 2 segments. It is possible to
extend the size of the AIX address space available to the
DataFlowEngine process by increasing the number of 256MB memory
segments available to it. This can be done by patching the
executable. Using the following shell commands you can patch
DataFlowEngine to use more memory segments.
/usr/bin/echo '\0100\0\0\0'|dd of=executable_file_name bs=4
count=1 seek=19 conv=notrunc
This creates a version of the DataFlowEngine which uses 4 segments,
i.e. 1GB.
NOTE: The above technique only overrides the soft limit, not the hard
limit. If your broker processes regularly reach these sizes, consider
spreading your messageflows across more execution groups to reduce the
size of each one below these limits.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC34391 - DB2 INSTANCE PARAMETERS MAXAGENTS AND MAXAPPLS NEED TO BE SET
The WMQI V2.1 documentation does not specify that the maxagents
parameter needs to be set at the same time as the maxappls parameter.
The following books will be amended to specify that the maxagents
parameter needs to be set at the same time as the maxappls parameter.
WebSphere MQ Integrator Administration Guide V2.1 (SC34579206),
Chapter 3; The broker, Defining and authorizing a broker database,
Creating a broker database, Broker database connections.
WebSphere MQ Integrator Using the Control Center V2.1 (SC34560205),
Chapter 5; Working with message flows, Creating a message flow,
Managing database resources.
WebSphere MQ Integrator Problem Determination Guide V2.1
(GC345920-01), Chapter 5; Dealing with problems during
administration, Problems with databases, Message SQL1040N.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC30627 - EXPLANATION OF CARDINALITY FUNCTION IS INCOMPLETE AND
INCONSISTENT IN USING THE CONTROL CENTER
The examples supplied in the Using the Control Center manual are
lacking and/or incorrect in clarifying how things actually work
with regards to the Cardinality function and XML messages that
contain pcdata, white space characters, attributes etc.
The following information on the Cardinality function will be
added to the WebSphere MQ Integrator Using the Control Center manual,
SC34-5602-02, Chapter 16 - Concepts of XML messages :-
WMQI V2 treats XML pcdata, attributes and sub-elements as
children of the element.
The ESQL Cardinality function returns the number of children
of a parent in the message tree. This means it counts pcdata,
attributes and sub-elements together, therefore when using the
Cardinality function with XML messages that contain pcdata,
white space characters, attributes etc. you need to distinguish
what you are determining the Cardinality of.
For example, to get the Cardinality of the number of elements
(tags), you would replace .* with .(XML.tag)
The MOVE statement will also navigate to such fields in an XML tree
when dealing with FIRSTCHILD, LASTCHILD, PREVIOUSSIBLING and
NEXTSIBLING. In CSD02 the TYPE and NAME clauses were added to allow
target fields of a specific NAME or TYPE meaning that 'non-data'
fields in the tree can be skipped over.
The WebSphere MQ Integrator ESQL Reference V2.1 (SC34592302),
Chapter 4; ESQL Statements, MOVE will be update to emphasise this
point.
The fourth paragraph currently states :-
If a TYPE clause, NAME clause or both are present, the target is
again moved in the direction specified (PREVIOUSSIBLING or
NEXTSIBLING or FIRSTCHILD or LASTCHILD) but to a field with the
given type, name or both.
If the specified move cannot be made .......
This will be added to as follows :-
If a TYPE clause, NAME clause or both are present, the target is
again moved in the direction specified (PREVIOUSSIBLING or
NEXTSIBLING or FIRSTCHILD or LASTCHILD) but to a field with the
given type, name or both.
This is particularly useful when the name and/or type of the
target field is known because this will reduce the number of MOVE
statements required to navigate to a field. This is because fields
that do not match the criteria will be 'skipped' over, and this
could also include unexpected message tree fields such as those
representing whitespace.
If the specified move cannot be made ........
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
PQ62125 - NEED DOCUMENTED GUIDELINES ON LOGGING 1GB LOB COLUMNS.
The following information will be added to the WebSphere MQ Integrator
for z/OS Customization and Administration Guide Version 2.1
(SC34-5919-01), Chapter 2; Customization, Customizing a broker,
Step 7. Create the broker database p39
Considerations with WMQI usage of large objects in DB2 tables on z/OS.
1) DB2 needs to allocate auxiliary tables to hold large objects (LOBs).
By default WMQI tables are created with CURRENT RULES = 'STD'. With
this option set, DB2 implicitly defines a LOB table space, an
auxiliary table and unique index for each LOB column (using default
sizes). The default sizes of these tables may be too small for your
particular requirements. This can mean that the maximum number of
extents for the table space is quickly used up.
If these tables have large and frequent updates the BIPDBTB member
in the CNTL library for the broker should be edited as follows :-
1.1) The line SET CURRENT RULES = 'STD'; should be commented out
1.2) For all tables using LOB columns, including the following
tables :- BSUBSCRIPTIONS, BRETAINEDPUBS, BACLENTRIES,
BMQPSTOPOLOGY, BROKERAAEG, BROKERRESOURCES, BRMRTDINFO,
BRMPHYSICALRES and BAGGREGATE
a) Add a ROWID column to each of the create table definitions.
e.g. ROW_ID ROWID NOT NULL GENERATED ALWAYS;
b) Create a table space and table to hold the LOB data. The table
space and table are called a LOB table space and an auxiliary table.
One LOB table space and one auxiliary table must be created for each
LOB column. Create these objects using the CREATE LOB TABLESPACE
and CREATE AUXILIARY TABLE statements. For the create table space
statement the following should be taken into consideration.
PRIQTY n
Specifies the minimum primary space allocation for a DB2-managed
data set. The primary space allocation is at least the integer n in
kilobytes, the actual value will vary depending on the page size for
the buffer pool used.
If PRIQTY is omitted, n is 200, 400, 800, or 1600 for 4KB, 8KB, 16KB
and 32KB page sizes, respectively.
SECQTY n
Specifies the minimum secondary space allocation for a DB2-managed
data set. The secondary space allocation is at least the integer n
in kilobytes. This will again vary with respect to the page size
for the buffer pool used.
If SECQTY is omitted, n is either 10% of PRIQTY or 50 times
the page size of the table space, whichever is larger.
c) Create an index on the auxiliary table.
The DSNTEJ7 sample shipped with DB2 demonstrates the above.
For existing brokers a DB2 utility tools such as RUNSTATS and
REORG can be used on the default named table spaces to
increase the primary and secondary quantities respectively.
2) LOBs defined with a maximum size of > 1GB are not logged by DB2.
This means that if you restore a backup copy of a database with
LOBs > 1 GB, DB2 will not recover changes to that table at restart.
To log changes to LOB columns you will need to reduce the size of
the LOB column to less than or equal to 1 GB and specify LOG YES on
the CREATE LOB TABLESPACE statement. Be aware that this may affect
the performance of the updates to these tables. If you do not wish
a LOB column to be logged, that is not greater than 1 GB, you will
need to specify LOG NO for the table space concerned.
Migration from an existing broker to one that supports logging of
LOB columns will require the recreation of the broker.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC29837 - COMMIT BY GROUPID DOCUMENTATION NOT CLEAR ABOUT ROLLBACK
A third bullet point will be added to the WebSphere MQ Integrator
Introduction and Planning Guide, Version 2.1 (GC345599-04), Part 2.
Business process planning, Chapter 4. Message flows, What is a
message processing node?, Error Handling, to clarify Commit By
GroupId, specifically when messages are being processed as part of
a message group, as follows:-
o If the message was being processed as part of a message group:
When handling a message flow error for a message that is part of
a message group, then the previous two points apply for the current
message being processed. However, consideration must be given to
the messages that have already been processed in the message group,
and those that have not yet been processed in the group.
The first point to note is that the use of "Commit by GroupId"
on the MQInput node is not synonymous with a Unit of Work.
When this option is selected this will lead to a commit being
issued when the last message in the message group has been
processed. If an error is encountered part way through the
group then the message flow error handling paths will be invoked
for this particular message as documented above. This error
processing could lead to a Backout being issued for the current
Unit of Work. Even when using "Commit By GroupId", it would
have been necessary to specify whether each preceding message
was participating in the Unit of Work using the standard
transactional control. Those group messages processed as part of
the Unit of Work prior to the backout, will be backed out and be
sent to either the backout queue, dead letter queue, or Input
queue dependent on the possible configurations documented above.
Irrespective of which of these is done, there may be remaining
group messages on the MQInput queue that can still be processed
by the message flow. If no extra group error handling has been
implemented, then these remaining messages will be retrieved by
the message flow, and processed in a new Unit of Work. When the
last message in the group is encountered and the "Commit by
GroupId" has been selected, then a commit will be issued. This
presents a situation where part of the group has been backed out
and part has been committed, these being divided by failing
messages. Hence, "Commit by GroupId" is not synonymous with
Units of Work when an error occurs for one of the group messages
which leads to a rollback.
To cope with this scenario, extra error handling routines should
be added to the message flow so that failing message groups can
be logged. This could be done in a database table, and when a
group message is retrieved, the table can be examined to see if
this group had failed at some point, allowing appropriate actions
to be taken. This combined with standard backout processing
would allow a message group to be processed in a similar fashion
to a Unit of Work.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IY29461 - ADD TO DOCUMENTATION THAT IT IS NECESSARY TO HAVE A MATCHING
CALL CNIDELETEMESSAGE IN A CUSTOM NODE IF THERE IS A
CNICREATEMESSAGE
A matching call to cniDeleteMessage is required for every call
to cniCreateMessage, otherwise a memory leak will occur.
The following information will be added to the MQSeries Integrator
Programming Guide, SC34-5603-01, regarding the use of
cniCreateMessage().
In Chapter 6, Implementing a plug-in node or parser; Implementing a
message processing node; Build an output message, a note will be
added to the end of the second paragraph (The input message is
read-only ......... syntax element navigation and access.)
as follows:-
Note that the copy of the input message created by the
cniCreateMessage function must be deleted by a matching call to the
cniDeleteMessage function after the message has been propagated, or
during cleanup processing after an exception has been thrown. This
will ensure resources allocated by cniCreateMessage are returned.
In Chapter 8, Node implementation and utility functions; Node utility
function interface; the description for the cniCreateMessage function
will include a reference to the cniDeleteMessage function,
as follows:-
Note that for every call to this function, there should be a matching
call to cniDeleteMessage to return allocated resources once the
processing on the output message has been completed.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
PQ60310 - 0C1 ABEND WHEN STARTING WMQI DATAFLOW ENGINE
The Websphere MQ Integrator, Problem Determination Guide V2.1,
GC34592000, Chapter 5, Dealing with problems during administration,
Problems with starting resources, will have the following
information added :-
Problems with starting resources
Abend code 0C1 when trying to start data flow engine on z/OS
Scenario: The first two WebSphere MQ Integrator address spaces start
successfully, but the third address space (the data flow engine)
fails to start. The result is an 0C1 abend.
The dataflow engine address space is spawned from the admin agent.
If the region size is too small, either because an insufficient
region size was specified in the procedure, or the region size was
overridden by the OS/390 IEFUSI exit, the data flow engine address
space might fail to start and fail with an 0C1 abend.
Check this by using IPCS on the dump (moving it from the HFS to a
traditional MVS dataset if required) by issuing the IPCS command:
verbx vsmdata,'noglobal,jobname(vcp0brk2)'
where vcp0brk2 is the name of the failing job. Find 'VSM LOCAL DATA
AREA AT ADDRESS '. The field ELIM gives the available region size,
and should be greater than 0C800000.
If the field SMFEL is not ffffffff, the IEFUSI exit has changed the
allowable region size. This value should also be greater than 0C80000.
If you have an IEFUSI exit, you need to check that the exit does not
limit the broker address spaces. For example, a commonly-used field is
OUCBSUBN. This can be STC or OMVS for the broker and indicates how
the address space was started.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC33120 - IMPROVE DOCUMENTATION ON THE NEW BEHAVIOUR OF PARSERS
The WebSphere MQ Integrator Working with Messages V2.1 book,
SC34-6039-01, Chapter 2. Parsers, Section: How a Message is
interpreted, SubSection: Using the supplied input nodes and
parsers will be updated with the following new information
before the paragraph that starts 'If you set up the message
headers or the input properties of a supplied .... '
At this point the selected parser is assigned to the message
body, however, for performance reasons, the bitstream is not
parsed. The message body will actually be parsed when a
reference is first made to it during the message flow. For
example, this will occur during the use of Root.XML.Field or
Root.MRM.Field etc (or InputRoot.XML.Field when in the compute
node) and therefore depending on paths taken in the flow, this
parse could take place at different points. This "parse when
first needed" approach is different to the V2.0.x versions of
the product, but in normal processing the logic of a message
flow will be unchanged. However when considering error
handling for invalid messages the following should be observed:
- If the input message is not referenced in a message flow, then
it will not be parsed. This means that an invalid message can
be processed and written out untouched by the message flow,
with no parser errors occurring.
- A parser error will be thrown at the point the message is
first parsed and therefore error handling paths should be
wired appropriately. For example, a ResetContentDescriptor
node that changes the owning parser will not cause an
immediate parse of the message by the new parser. Therefore
connecting the failure terminal of this node will not handle
any parser errors that may be generated.
- A trace node that specifies ${Root} and ${Body} will cause a
full parse of the message. Therefore the insertion of a trace
node for debugging purposes will cause parser errors that may
not be generated otherwise.
- A check node cannot be used to parse and validate input
messages. This node is used to check the current owning
parser, not message validity.
The WebSphere MQ Integrator Using the Control Center V2.1 book,
SC34-5602-05, Appendix A: The IBM Primitive nodes, Section:
ResetContentDescriptor node will be changed as follows:-
The first paragraph currently reads :
The ResetContentDescriptor node takes the bit stream of the
input message and reparses it using a different message template
and either the same or a different parser. The node can reset
any combination of message domain, type and format.
This will be changed to the following :
The ResetContentDescriptor node takes the bit stream of the
input message and will request that it be reparsed using a
different message template and either the same or a different
parser. The node can reset any combination of message domain,
type and format. The bitstream will not be parsed by the new
combination until a reference is made to the message body later
in the flow.
The third paragraph currently reads :
The format of an incoming message might be unknown when it
enters a message flow, so the BLOB parser is invoked. Later
on in the message flow, you might decided that the message
is defined in the MRM format and you can use the
ResetContentDescriptor node to
reparse the message using the MRM parser.
This will be changed to the following :
The format of an incoming message might be unknown when it
enters a message flow, so the BLOB parser is invoked. Later
on in the message flow, you might decided that the message
is defined in the MRM format and you can use the
ResetContentDescriptor node to
cause the reparse of the message using the MRM parser.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC33040 - INSUFFICIENT DETAIL ON IMPLEMENTING THE PASSTHRU FUNCTION.
Description of the PASSTHRU statement, previously described in the
MQSeries Integrator, Using the Control Centre manual, SC34560200,
and now described in the WebSphere MQ Integrator ESQL Reference
guide, SC34-5923-02, will be expanded in Chapter 4; ESQL statements,
as follows:
If two or more other parameters are passed, each parameter is bound
to the corresponding question mark in the statement string: that is,
the first parameter is bound to the first question mark, the second
parameter is bound to the second question mark, and so on.
The question marks are processed as parameter markers in a prepared
SQL statement by the DBMS. This means that if you are passing values
to functions in the DBMS you will have to be sure to have
appropriately typed your parameter markers. For example, in DB2 a
parameter marker would be cast as follows:
SET OutputRoot.XML.Result.Data =
PASSTHRU('SELECT * FROM user1.stocktable
where time=TIMESTAMP(CAST(? as TIMESTAMP)),
'1999-12-31-23.59.59.999999');
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
SPECIAL INFORMATION
-------------------
PLEASE REVIEW THE README.TXT FILE WHICH HAS BEEN UPDATED IN THIS
MAINTENANCE RELEASE.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IY38894 - BIP2322E AND BIP8040E ERRORS WHEN MIGRATING A V201 BROKER
USING AN ORACLE BROKER DATABASE TO V202/V210.
There is a known problem on AIX and HP/UX when migrating V201 brokers
that use Oracle for the broker database to V2.0.2 or V2.1. The
following errors may be reported when running the mqsicreatebroker
command to create a temporary broker to upgrade the broker database:-
BIP2322E: Database Error: SQL State IM003 Native Error Code 80
Error Text 'Specified driver could not be loaded'.
BIP8040E: Unable to connect to the database.
This occurs because the Driver parameter in the
/var/mqsi/odbc/.odbc.ini file for the Oracle ODBC Data Source Name
specified in the mqsicreatebroker command references the Merant Oracle
Driver that shipped with V2.0.1. To rectify this problem it is
necessary to amend the parameter for each Oracle Data Source used by
MQSI brokers to reference the relevant Merant Oracle Driver that ships
with V2.0.2 or V2.1. This involves replacing the reference to UKor815
with UKor816.
For example, on AIX change
Driver=/usr/opt/mqsi/merant/lib/UKor815.so
to
Driver=/usr/opt/mqsi/merant/lib/UKor816.so
on HP/UX change
Driver=/opt/mqsi/merant/lib/UKor815.sl
to
Driver=/opt/mqsi/merant/lib/UKor816.sl
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
29862 - Merant driver updates
CSD05 includes a new version of the DataDirect Technology (formerly
Merant) ODBC Driver Manager. Changes MUST be made to the ODBCINI
file entries and to any user compiled switch load files that use the
number '16' files. All entries containing the number '16' MUST be
changed to the number '18'. See below for details. Switch load
files supplied by WebSphere MQ Integrator will be updated
automatically by application of CSD05.
Note that the value for LANG should be set to C or en_US
While the number '16' files will remain there will be no further
updates to them. ODBC fixes will only be supplied on the number
'18' files.
AIX, Solaris and HP-UX.
-----------------------
The ODBCINI file.
-----------------
DB2 (on AIX only).
-----------------
The name of the DB2 driver used has also changed from db2.o to
libdb2.a . The file pointed to by your $ODBCINI environment variable
and any other user built files using db2.o will need to be changed .
For example change
Driver=/db2data/db2inst1/sqllib/lib/db2.o
to
Driver=/db2data/db2inst1/sqllib/lib/libdb2.a
where /db2data/db2inst1 is the DB2 instance home on your machine
ORACLE and SYBASE.
------------------
WMQI V2.1 CSD05 includes a new version of the DataDirect Technology
(formerly Merant) ODBC drivers and Driver Manager. The driver files
supplied by DataDirect have changed name from drivers containing the
number '16' to the number '18'. Changes MUST be made to the file
refrenced by your $ODBCINI environment variable . All entries
containing the number '16' MUST be changed to the number '18'.
On AIX change
Driver=/usr/opt/mqsi/merant/lib/UKor816.so
to
Driver=/usr/opt/mqsi/merant/lib/UKor818.so
and
Driver=/usr/opt/mqsi/merant/lib/UKase16.so
to
Driver=/usr/opt/mqsi/merant/lib/UKase18.so
On Solaris change
Driver=/opt/mqsi/merant/lib/UKor816.so
to
Driver=/opt/mqsi/merant/lib/UKor818.so
and
Driver=/opt/mqsi/merant/lib/UKase16.so
to
Driver=/opt/mqsi/merant/lib/UKase18.so
On HP/UX change
Driver=/opt/mqsi/merant/lib/UKor816.sl
to
Driver=/opt/mqsi/merant/lib/UKor818.sl
and
Driver=/opt/mqsi/merant/lib/UKase16.sl
to
Driver=/opt/mqsi/merant/lib/UKase18.sl
The qm.ini file.
----------------
Brokers using MQSeries queue managers that are performing XA
coordination with Sybase or Oracle on AIX or Solaris will need
their XA stanzas in the queue managers qm.ini file altered to use
the '18' files instead of the '16' files.
On AIX change
SwitchFile=/usr/opt/wmqi/merant/lib/UKor8dtc16.so
to
SwitchFile=/usr/opt/wmqi/merant/lib/UKor8dtc18.so
and
SwitchFile=/usr/opt/wmqi/merant/lib/UKase16.so
to
SwitchFile=/usr/opt/wmqi/merant/lib/UKase18.so
On Solaris change
SwitchFile=/opt/wmqi/merant/lib/UKor8dtc16.so
to
SwitchFile=/opt/wmqi/merant/lib/UKor8dtc18.so
and
SwitchFile=/opt/wmqi/merant/lib/UKase16.so
to
SwitchFile=/opt/wmqi/merant/lib/UKase18.so
There are no changes needed on HPUX for qm.ini.
WINDOWS
-------
Oracle and Sybase.
------------------
ODBC definitions.
-----------------
WMQI V2.1 CSD05 includes a new version of the DataDirect Technology
(formerly Merant) ODBC drivers and Driver Manager. The driver files
supplied by DataDirect have changed name from drivers containing the
number '16' to the number '18'. You MUST need recreate the ODBC
definitions for your databases so as they use the following drivers.
For Oracle
use the "MQSeries DataDirect Technologies 4.10 32-BIT Oracle" driver
For Sybase
use the "MQSeries DataDirect Technologies 4.10 32-BIT Sybase" driver
XA Resource definitions
-----------------------
Brokers using MQSeries queue managers that are performing XA
coordination with Sybase or Oracle will need to have their
resource properties modified to use the new DataDirect dlls.
Change
$MQSI\bin\UKor8dtc16.dll
to
$MQSI\bin\UKor8dtc18.dll
and
$MQSI\bin\UKase16.dll
to
$MQSI\bin\UKase18.dll
where $MQSI is the place where WMQI V2.1 is installed.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC35512 - UPDATE TO DOMAIN SECURITY AND AWARENESS DOCUMENTATION.
Problems have been experienced concerning the use of extended domain
security options, specifically domain awareness and the -l and -d
options, when creating the Configuration Manager.
The changes to the domain security function have been explained in
the readme.txt file (accompanying this maintenance release) in the
Documentation section, subsection: Updates, point 1, Using Fully
Qualified User Names for Control Center Users.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
SupportPac IA0F: WebSphere MQ Integrator - IDoc parser for SAP
The IDoc SupportPac, IA0F, was included in the product from CSD03
onwards. This SupportPac contains a set of parsers which add support
for parsing of input messages and creation of output messages in the
SAP IDoc Version 4 format.
The IDoc parser represents incoming data to WebSphere MQ Integrator
in a format that the Integrator compute node can manipulate. It also
allows data of a different format to be mapped to an IDoc stream.
For more information, see the support pack IA0F documentation at
www.ibm.com/software/integration/support/supportpacs/individual/ia0f.html
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
WebSphere MQ Everyplace 1.2.7 nodes
The WebSphere MQ Everyplace nodes (1.2.7) supported in WebSphere MQ
Integrator V2.1 can be obtained via the Level 2 Service organisation
at http://l3.hursley.ibm.com/ftp/fixes/MQe/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Migration from previous levels to CSD04
The redundant validate checkbox has been removed from the Advanced tab
of the input nodes in CSD04. Therefore, when migrating flows developed
pre-CSD04, ensure the validate check box on the Advanced tab of all
input nodes is unchecked. Failure to do this will result in BIP1758
error messages during a deploy of message flows containing such nodes.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
28704 - Merant Driver Manager update
Support for DB2 UDB Version 7.1 FixPak 2 or later on UNIX.
The DB2 flash 1042663 - "Merant Driver Manager and DB2 UDB V7 ODBC
driver on Unix" described incompatibilities that had been encountered
with Unicode support when the Merant Driver Manager accessed DB2's
ODBC driver on UNIX. The affected versions of DB2 UDB included
Version 7.1 with FixPak 2 or later, and Version 7.2.
The workaround was to use an alternative DB2 ODBC driver library
without Unicode support enabled.
CSD04 includes a new version of the DataDirect (formerly Merant)
Driver Manager which has removed the need for the workaround
documented in the above DB2 flash.
If your DB2 database is running on HP-UX or Solaris you will need to
remove the above workaround to enable this support.
If your DB2 database is running on AIX you will need to remove the
above workaround and also change the DB2 Driver entry in your
.odbc.ini file
from : Driver=/db2data/db2inst1/sqllib/lib/db2.o
to : Driver=/db2data/db2inst1/sqllib/lib/libdb2.a
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC34394 - MQOUTPUT NODE ABENDS WHEN WRITING AGGREGATED REPLY MESSAGE
This is resolved in CSD04 however if the problem has already
been encountered the BAGGREGATE table of the broker database will
contain records which MUST BE DELETED after applying CSD04 and
BEFORE STARTING THE BROKER. Failure to do this will prevent the
fix taking effect and errors will be reported in the system log
indicating 'failure due to invalid MQMD header at MQOUTPUT node'.
If this problem has not been encountered these entries will not be
present in the BAGGREGATE table and no further action is required.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC33207 - SILENT INSTALL FOR CSD FAILS
The problem with the backup folder has now been resolved.
For information on how to perform a silent install of a CSD
see Chapter 3; Installing Websphere MQ Integrator; Silent
Installation, of the Websphere MQ Integrator for Windows NT
and Windows 2000 Installation Guide Version 2.1, and use the
maintenance CD instead of the product CD.
Note that to perform a silent install of a CSD downloaded from
the web perform the following actions.
To create the response file change to the directory where the
CSD .exe file has been stored and run the following command :-
U200nnn.exe -s -a setup.exe -noinst -r -f1x:\mypath\my.iss
Where U200nnn is the PTF number of the CSD being loaded
and x:\mypath\my.iss is the target drive for the response
file being recorded
To perform the silent install run the following command:-
U200nnn.exe -s -a setup.exe -nomsgbox -s -f1x:\mypath\my.iss
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Migration from CSD02.
In order to activate new options within the Control Center, the
Config Manager must be deleted and re-created. This is done as follows
1) Stop the Config Manager
mqsistop ConfigMgr
2) Delete the Config Manager preserving the configuration and message repository.
mqsideleteconfigmgr
Use no options on this command.
3) Re-create the Config Manager
mqsicreateconfigmgr
Use all the options that were used when the Config Manager was last created.
4) Start the Config Manager
mqsistart ConfigMgr
5) Restart the Control Center and the new options will be available.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC30970 - GLOBALLY CO-ORDINATED FLOW INVOKING NEON PARSER CAUSE XA
PROBLEM LEAVING DB INVOLVED IN A TRANSACTION LOCKOUT.
This problem has been resolved by introducing a new environment
variable, MQSI_NNSY_XA. If it is not set there is no change in
operation i.e. one connection per process and New Era nodes connect
to the database themselves. If it is set to anything AND we are
using DB2, then xa support is enabled, i.e. we have one database
connection per thread and we connect to the New Era database on
their behalf.
This environment variable changes the connection to the New Era
database from a per process to a per thread basis. As each
connection has it's own cache of rules and formats it is expected
that memory usage will increase. For further information see the
entry in the NEW ERA OF NETWORKS INFORMATION section of this file.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
To use WebSphere MQ Integrator on a Solaris machine in Simplified
Chinese in zh locale you should modify the sample profile as shown
below. If you are using zh.GBK locale, there is no need for you to
modify the sample profile.
NLSPATH=/usr/lib/locale/%L/LC_MESSAGES/%N:$NLSPATH;
/opt/wmqi/messages/Zh_CN/%N:/opt/wmqi/messages/%N
LANG=zh
LC_ALL=zh
export NLSPATH LANG LC_ALL
Alternatively you may set up symbolic links to the message catalogue
files in /opt/wmqi/messages/Zh_CN from /usr/lib/locale/zh/LC_MESSAGES
if you wish to avoid changing the NLSPATH environment variable.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
CSD02 introduces the concept of prefixed identifiers for MRM elements.
Refer to the Working with Messages book, Chapter 4 section "Names and
Identifiers" for full details. If this concept is exploited, you must
ensure that all components of your WMQI domain are upgraded to CSD2,
that is, your Configuration Manager, all Control Centers, and all
brokers to which a message set containing such elements is to deployed.
Failure to upgrade in this manner will result in parse errors in the
broker. Please note also that the DTD Importer functionality of the
Control Center exploits prefixed identifiers, so if this is used to
populate message sets then the upgrade stated above must be performed.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
IC32427 - V201 AND V202 TO V2.1 MRM MIGRATION AND COEXISTENCE
WMQI V2.1 contains a major upgrade to MRM capability. As a
result of this upgrade, the format of the message dictionary
that flows from the Configuration Manager to brokers during
a deploy operation has changed compared with MQSeries Integrator
V2.0.x.
Migration details are in the Administration Guide, chapter 12.
Please note the following additional information:
Importing 2.0.x message sets into a 2.1 Configuration Manager.
When importing 2.0.x message sets into the 2.1 Configuration
Manager you may choose to create a default XML wire format
layer. This additional layer allows a 2.1 broker to process
application messages that specify a message domain of 'MRM'
and a message format of 'XML'. You create the layer by adding
the '-x XML' option to your mqsiimpexpmsgset command line.
You may specify any valid name for the new layer, but for
compatibility with existing messages and message flows you must
use 'XML'.
If you do not migrate your Control Centers, Configuration
Manager and all brokers in the domain to WMQI 2.1 at the same
time, then you must be aware of the following extra information
that applies to co-existence scenarios:
A 2.1 Configuration Manager may co-exist with 2.0.1 and 2.0.2
brokers, as long as message sets assigned to the 2.0.x brokers
do NOT exploit the new messaging functionality available in 2.1.
This is the recommended co-existence scenario.
In order to implement this scenario, you must perform the
following steps:
1) Apply the fix for APAR IC32427 to all the 2.0.x brokers in
the domain.
2) Migrate the existing 2.0.x Configuration Manager to V2.1 and
apply cumulative maintenance zero to the 2.1 Configuration
Manager.
Once the first step is complete you will still be able to
administer your brokers using the 2.0.x Configuration Manager.
After you complete the second step you will be able to add 2.1
brokers to the domain.
If the message set does exploit new MRM function, the deploy to
a 2.0.x broker will fail with a BIP5293 message. Function that
can cause this error includes:
- Types with composition other than 'ordered set' and
'unordered set'.
- Types with context other than 'open'.
- Types with Types as members.
- Types with Messages as members.
- Simple Types other than 'binary', 'boolean', 'float',
'integer', and 'string'.
- Type Members of type 'string' with no maximum length
specified. You may specify the maximum length by adding an
Element Value Constraint with the role of Max Length.
- Types with a value. If your Compound Type has a value, there
will be a 'Type' field shown on the main properties page for
the type, below the 'Suspended from Use' field.
A 2.0.1 or 2.0.2 Configuration Manager may co-exist with 2.1
brokers. Note that this co-existence scenario is NOT supported
for application messages that specify a message domain of 'MRM'
and a message format of 'XML'. In order to implement this
scenario you must perform the following steps, IN THIS ORDER:
1) Apply the fix for APAR IC32427 to all the 2.0.x brokers in
the domain.
2) Apply the fix for APAR IC32427 to the 2.0.x Configuration
Manager.
3) Apply cumulative maintenance zero to 2.1 brokers, and add
them to the domain. Once the first step is complete you will
still be able to administer your existing brokers using the
2.0.x Configuration Manager.
If you apply the fix for APAR IC32427 to the 2.0.x Configuration
Manager before applying the fix to the 2.0.x brokers then
message sets may fail to deploy. If the failure occurs, BIP5145
and BIP5280 messages will be reported by the broker.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
NEW ERA OF NETWORKS INFORMATION
-------------------------------
The installation of the New Era of Networks (NEON) component may have
altered the permissions of the /usr and /usr/lpp directories to 777.
The new /usr/lpp/neonsoft directory that is used for the NEON component
and some of its subdirectories may also have been given the same 777
permission. If security is a concern review the permissions on these
directories and change as required. To restore the default permissions
on /usr and /usr/lpp use:
chmod 755 /usr
chmod 755 /usr/lpp
CSD05
=====
Contents:
Part 1: General Issues
Part 2: New Era of Networks Formatter Issues
Part 3: New Era of Networks Rules Issues
Part 4: Visual Tester
Part 5: Documentation Issues
====================================================================
Part 1: General Issues
====================================================================
CR # 253551, 287258
APAR # IY37844 - NNSYMESSAGELOG MUST BE ACCESSIBLE BY THE BROKER
AND IN THE CURRENT DIRECTORY.
Problem:
The NNSYMessageLog.nml file is not being created or written causing
DFE to terminate.
Resolution:
Create and Write permissions must be granted to the UserId in the
directory where the broker is started.
--------------------------------------------------------------------
CR # 254087
APAR # IY24156 - INST_DB NOT RETURNING THE UNDERLYING ORACLE ERROR CODE
Problem:
This is a known inst_db problem. Because of how inst_db uses
the DBTIL component and how DBTIL stores errors internally, the
application does not report native database vendor errors during
failed connections.
Work-around:
Do the following:
1. Upon receipt of a generic error from inst_db, assume some sort of
login failure.
2. Then use vendor-specific interactive SQL tools to troubleshoot
the source of the login failure.
--------------------------------------------------------------------
CR # 275130, 296735
APAR # IY35755 - REFORMAT FAILS TO FIND INPUT DATA - NEONTRANSFORM
ENGINE FAILS TO TRANSFORM A MESSAGE
Problem:
Repeating Alternative Output formats sometimes are unable to map
all parsed input data.
Resolution:
The code has been adjusted to prevent this from happening. Repeating
Alternative Output formats now work properly.
--------------------------------------------------------------------
CR # 299402
APAR # IC34947 - APAR IC34947 NNSY NODES/PARSER NOT PICKING UP THE
CCSID OF THEMESSAGE FROM THE FLOW - MQMD.CCSID
Problem:
There is a problem with an WMQI 2.1 message flow. It does the
following:
1) An input Node that sets the Message Domain to NEON and sets a
Message type and format which then parsers the message.
2) The header gets removed.
3) The NEON fields are taken and an XML output is created.
4) The data is then output on a queue.
Resolution:
A correction was made to the wrapper code for the New Era API. The
MQSI 2.1 message flows now operate properly.
---------------------------------------------------------------------
CR # 296422
APAR # PQ66950 - MEMORY LEAK WHEN USING THE NEON FEATURE OF WMQI.
Problem:
On OS/390, the WMQI Broker runs out of memory after processing about
10,000 messages.
Resolution:
INFR was adjusted to correct the problem.
---------------------------------------------------------------------
CR # 303392
APAR # IC35410 - MQSI THROWS NEWERAON FORMAT EXCEPTION AS SETERROR
ALWAYS IGNORES NN_ERSTATUS_OK (0), CHANGING IT TO
NN_ERSTATUS_ERROR (-1)
Problem:
The intended usage for NNExitRet::SetError was when user exit
routines are in an error state. SetError allowed an informative error
message to be logged before the user exit erred out. Passing a
message in with NN_ERSTATUS_OK is assumed to be a mistake, and
SetError changed NN_ERSTATUS_OK to NN_ERSTATUS_ERROR silently.
Resolution:
The new behavior is:
* SetError allows logging of true error messages if passed with
NN_ERSTATUS_ERROR
* SetError allows logging of messages in non-error situations if
passed with NN_ERSTATUS_OK
* SetByteArrayValue will clear all NN_ERSTATUS_ERROR messages, but
not NN_ERSTATUS_OK (debug) messages
====================================================================
Part 2: New Era of Networks Formatter Issues
====================================================================
CR # 277090
PMR # 69192
Problem:
NNFie does not properly export some output controls, causing a
failure to occur.
Resolution:
The SQL was fixed to properly export Substitute output controls.
--------------------------------------------------------------------
CR # 290523
APAR # IY38886 - ZERO LENGTH MSGS ON OUTPUT Q WHEN LOGLVL=3
Problem:
On AIX, the Log Level setting #3 corrupts the broker behavior.
Resolution:
The Message Logging code was fixed to properly handle all the levels.
--------------------------------------------------------------------
CR # 290843
APAR # IC34564 - 'USED IN' DOES NOT SHOW CORRECTLY FOR THE LITERAL
THAT HAS % SIGN IN IT
Problem:
Formatter GUI does not display Output Controls in the Used In list.
Resolution:
The code was modified and now Output Controls appear on the list.
--------------------------------------------------------------------
CR # 296377
APAR # IC35605 - INCORRECT OUTPUT DATA LENGTH WHEN THERE IS NO
VALID DATA CAUSES A MESSAGE TO FAIL - NNSY.
PMR # 42404
Problem:
Output fields with effective length zero were being handled
incorrectly. They appear in output messages when running under the
WMQI 2.1 Broker.
Resolution:
The code was changed to properly handle length zero output fields.
--------------------------------------------------------------------
CR # 296736
APAR # IY38935 - IN RULES EXPORT FILES AN EXTRA SPACE AFTER AN EXIST
OPERATOR IN RULES EXPRESSION,CAUSES A CONFLICT IN NNRIE
PMR # 43909
Problem:
NNFie has trouble importing STRING literals. On a second import of
the same NNFie file, conflicts are reported incorrectly.
Resolution:
The code was changed to fix the reading of literals type
(STRING/BINARY) from the database on non-Windows platforms.
--------------------------------------------------------------------
CR # 298811
APAR # IY39051 - ALTERNATIVE FORMAT DOES NOT FIND IT'S INPUT DATA
Problem:
When migrating from versions 4.11 to 5.6, data loss occurs and
missing sections appear within repeating alternative components.
Resolution:
A change was made to allow processing of controlling fields that
appear within a repeating alternative.
--------------------------------------------------------------------
CR # 302048
PMR # 07912
Problem:
NNFie fails to import some files from MQSI 1.1.
Resolution:
A message handling error caused the incorrect failures and was fixed.
--------------------------------------------------------------------
CR # 304880
APAR # IC35605 - INCORRECT OUTPUT DATA LENGTH WHEN THERE IS NO
VALID DATA CAUSES A MESSAGE TO FAIL - NNSY.
PMR # 42652
Problem:
Trim and codepage problems occur with output messages on the
NEONEval node.
Resolution:
The code was changed to properly handle length zero output fields.
====================================================================
Part 3: New Era of Networks Rules Issues
====================================================================
CR # 275544
APAR # IC33620 - RULES EXPRESSION CHANGES WHEN IT IS APPLIED
PMR # 90880
Problem:
When clicking on the 'Apply' button in the Rules GUI, the expression
changes, resulting in a extra set of parentheses and the | (pipe)
changing to an & (ampersand).
Resolution:
The code was fixed to correctly parse expressions when you use
additional parentheses in your expression.
--------------------------------------------------------------------
CR # 290252
APAR # IC35428 - RE-EVALUATED RULE DOES NOT GET HIT
Problem:
Certain functions that are valid under OldNodes processing such as
Re-evaluate and TagSearch do not work under the NewNodes version.
Work-around:
Use the OldNodes processing for functions that do not work under
NewNodes.
--------------------------------------------------------------------
CR # 294240
APAR # IC35013 - ERROR VIEWING THE PROPERTIES OF A RULE CONTAINING
FORMATS WITH FIELDS THAT DO NOT START WITH AN
ALPHANUMERIC CHARACTER
Problem:
Field names beginning with a character other than alphanumeric
causes the Formatter GUI to error.
Resolution:
The code was adjusted to allow for special characters.
--------------------------------------------------------------------
CR # 296736
APAR # IY38935 - IN RULES EXPORT FILES AN EXTRA SPACE AFTER AN EXIST
OPERATOR IN RULES EXPRESSION,CAUSES A CONFLICT IN NNRIE
PMR # 43845
Problem:
Some Rules export files have an extra space after an EXIST operator
in the Rules expression. This space is trimmed before the data is
imported, but it causes a conflict in NNRie.
Resolution:
The program now trims trailing spaces from the Rules expression
when the data is imported.
---------------------------------------------------------------------
CR # 297111
APAR # IY38799 - RULESET AND ADDED AGAIN AS A 'FORMATTER' MESSAGE TYPE.
Problem:
When you create a Rules message type that has the same name as a
Format even when the message type is deleted from a Ruleset and added
again as a Formatter message type, it continues to show as a Rules
message type.
Resolution:
The product now checks to see if the message type is in a Ruleset:
- If it does exist in a Ruleset, an error message is generated and
you will need to delete it and add one with a different evaluation
type.
- If it does not exist in a Ruleset, the Evaluation Type changes to
the type of the newly added message type.
--------------------------------------------------------------------
CR # 299233
APAR # IY38713 - THE MQMD FORMAT IS BLANK, PROCESSING A MESSAGE
USING A NEONRULESEVALUATION NODE.
Problem:
When a message is processed using a NEONRulesEvaluation node, the
output message does not contain the expected value set in the MQMD
format of its Subscription option. The MQMD Format is blank.
Resolution:
The MQMD format field is now being populated.
====================================================================
Part 4: Visual Tester
====================================================================
CR # 296737
APAR # IC35084 - INPUT CONTROL SET AS REGULAR EXPRESSION + DELIMITER
CAUSES APITEST/VISUAL TESTER TO TERMINATE
Problem:
When an input control is incorrectly set as Regular Expression
+ Delimiter instead of just Delimiter, apitest and Visual Tester
terminate abnormally.
Resolution:
The code was adjusted and now checks for a NULL Regular Expression
before trying to use it in an error message.
====================================================================
Part 5: Documentation Issues
====================================================================
CR # 260215 - System Management Guide
CMVC # 20626, NN406I - Missing Insert
Problem:
The location of the NNSYMessagelog.nml file is not indicated in
the current documentation.
Resolution:
On Unix
The NNSYMessageLog.nml file is written to the current working
directory at application startup. For broker logging, this file is
written to the current directory when mqsistart was issued.
For migration and test utility logging, such as msgtest, NNFie,
and NNRie, the NNSYMessageLog.nml file is written to the current
working directory.
On Windows NT
For broker and Control Center logging, the NNSYMessageLog.nml file
is written to \bin where wmqi_inst_loc is the location
of the WebSphere MQ Integrator installation; for example,
"C:\Program Files\IBM\WebSphere MQ Integrator 2.1".
For migration and test utility logging, such as msgtest, NNFie, and
NNRie, the NNSYMessageLog.nml file is written to the current working
directory.
--------------------------------------------------------------------
CR # 273641 - System Management Guide
CMVC # 25353 - Decimal Values Are Allowed in a Numeric Field
Problem:
If you define an input control field with a data type as Numeric,
New Era of Networks Formatter parses with decimal values that are
inconsistent with the Numeric data type information described on
pages 171, 180, 186, 187 of the System Management Guide.
The SM Guide states:
"NUMERIC data types only support characters (0-9) and do not
support decimals. While an error might not occur with decimals,
the decimals may be ignored and the number will be truncated to a
whole number.
Resolution:
When using the Numeric data type in your input or output control
definitions, observe the following additional guidelines:
Input Control Definition Guidelines
- If you define an input control field value as a decimal, New Era
of Networks Formatter will not fail to parse.
Output Control Definition Guidelines
- If you want the decimal portion of the input date to appear
in an output message, select N/A or String as the data type in
your output control definition. New Era of Networks Rules will
ignore the decimal portion of the input data.
- If you do not want the decimal portion of the input date to appear
in an output message, select Numeric as the data type in your
output control definition.
----------------------------------------------------------------------
CR # 275197 - System Management Guide
PMR # 89584
Problem:
The following section of information is missing from the WMQI V2.1
System Management Guide.
Resolution:
Include the following information in the WMQI V2.1 System Management
Guide.
Assigning Permissions
To assign permissions, you must create more than one user in your
database.
In Oracle
The following procedure describes how to create database users in
Oracle. To access WMQI databases, users must be created
and associated with the user role that was created during the WMQI
installation.
The process for creating users in Oracle includes:
- Creating users
- Granting roles to users
- Creating synonyms for database objects
Creating Users
You must create users to identify individual user names in the
database.
To create a user, type the following command and press ENTER.
create user identified by ;
where USERNAME and PASSWORD are required parameters.
Granting Roles to Users
To give users permissions to access the database data, you can grant
permissions to an individual user or create roles with certain
permissions and associate users with specific roles.
A role was created for user access to the WMQI database
during the installation process. This role uses the following format:
_ROLE
where is your site-specific role name provided during
installation.
To grant the role to each created user:
1. Log in to the database as SYS.
2. Establish a connection to the Oracle instance using SQL Plus and
the SYS username and password.
3. Type the following command to grant the role to the new user:
grant _ROLE to ;
The _ROLE role is granted to the user identified by
USERNAME.
NOTE: For rules permissions, each user can have only one role granted
and this role must be the same for all users.
Creating Synonyms for Database Objects
To create synonyms for the WMQI database objects, users
must run a database script.
To run the WMQI synonyms SQL script:
1. Locate the nn_synonyms.sql script in the install.sql_rulfmt56\inst_db\or806\tmp
sub-directory of the installation directory. Log in to the directory
where the nn_synonyms.sql resides.
2. Run the nn_synonyms.sql script for each user using the following
command:
sqlplus @@ nn_synonyms.sql> synonyms.log
3. Review the contents of the synonyms.log file for any errors. Ignore
the errors related to dropping synonyms.
In Sybase or SQL Server
The following procedures describe how to create users in Sybase or
Microsoft SQL server.
The commands are run within the command line program isql, and all
references to SQL Server include both Sybase and SQL Server.
Except for changing passwords, these procedures are limited to either
the system administrator or database owner.
Users must have login accounts and a user name in each database they
want to access. Adding login accounts, database users, and object
permissions can be done by the system administrator, security officer,
or database owner. A single person can occupy one or more of these
roles. Check with your database administrator for information about your
specific environment.
The process includes:
- Creating login accounts
- Assigning users to a database
- Defining user groups
Creating Login Accounts
Login accounts give users access to the SQL Server. To access a database,
a login must be assigned to a user name to the databases the user wants
to access. Only a system administrator or security officer has the
authority to create login accounts.
To create a login account, enter the following:
sp_addlogin loginName, password [, defdb [, deflang [,sid [, encryptopt]]]]
where loginName is the sysname with no default,
where password is the login; login is sysname with no default,
where defdb is the default database for the login (database is sysname
with a default of master),
where deflang is the name of the default language assigned when a user
logs in to SQL Server (language is sysname with a default of NULL; if
language is not specified, language is set to the servers current default
language defined by the sp_configure variable default language).
NOTE: Changing the server default language does not change the default
language for existing logins. The language remains the same as the
default language used when the login was added.
where sid is the security identification number (SID) [sid is varbinary
(16) with a default of NULL. If sid is NULL, the system generates a SID
for the new login, specifying a value other than NULL, but the number must
not already exist. SID is useful when you are scripting or moving SQL
Server logins from one server to another and you want the logins to have
the same SID between servers.],
and where encryptopt specifies whether the password is encrypted when
stored in the system tables (encryption_option is varchr(20) and can be
one of these values:
Value Description
NULL (default) The password is encrypted.
skip_encryption The password is not encrypted. The supplied
password was encrypted by an earlier version of
SQL Server and is used for upgrade purposes only.
Assigning Users to a Database
To use a database, the server login must be associated with a user name
in the database. Only the database owner (DBO) can assign users to a
database.
From the database in which the user is to be added, enter the following:
sp_adduser loginName [, nameInDB] [, group]
where the loginName is the user's server login account,
where the nameInDB parameter is the name for the user in the database,
where the nameInDB defaults to the loginName if it is not specified,
and where the group enables the DBO to add the user to an existing group
in the database. If a group is not specified, the user is placed in the
default group, PUBLIC.
NOTES:
- To add permissions both in New Era of Networks Rules and in New Era of
Networks Formatter, you must add users as users not aliases. They must
be members of the same user group.
- SQL Server: Permission errors occur if the user is the same as the
database owner (DBO). the user must be part of NNSYGRP, but cannot be
the DBO.
Defining User Groups
During installation process, NNSYGRP is created for WMQI
users. Users must be linked to the NNSYGRP group to access the WMQI
databases. In addition, each database user, with the exception
of the database owner, must be granted permissions to access objects
within that database.
To add users to NNSYGRP, do one of the following:
- To use the sp_adduser procedure, enter the following:
sp_adduser loginName [, nameInDB] [, group]
where the loginName is the user's server login account,
where the nameInDB parameter is the name for the user in the database,
where the nameInDB defaults to the loginName if it is not specified,
and where the group enables the DBO to add the user to an existing
group in the database. If a group is not specified, the user is placed
in the default group, PUBLIC.
- To use the sp_changegroup system procedure, enter the following:
sp_changegroup groupName, userName
where groupName is the name of the group to which the user is added,
and where userName is the database user name.
In DB2 UDB
The following procedures describe how to create database users in DB2.
It includes the following:
- Creating users
- Creating aliases for database objects
To allow other users to have access to the objects created during
installation, a new user must be created with aliases to link the user’s
objects to the objects in the installer’s schema. New users are added to
the NNSYGRP group created during installation. You must be a member of
NNSYGRP in order to access any of the objects owned by another NNSYGRP
member.
Creating Users
To grant user access to the database, you must create additional user
accounts at the local system level.
Create additional users in the NNSYGRP group.
Creating Aliases for Database Objects
Each new user must run a database script to create aliases for the WMQI
database objects created during the WMQI database installation.
NOTE: To have assignable permissions, you must be a member of the
system group NNSYGRP.
Ask your database administrator to create aliases for OS/390.
To create aliases using the nn_aliases.sql script:
1. Connect to the WMQI database using the user ID of the user
you want to add.
2. Establish a DB2 CLP session.
Example:
db2 connect to user using
3. Locate the nn_aliases.sql script in the
\install.sql_rulfmt56\inst_db\tmp directory of the installation
directory.
4. Run the nn_aliases.sql script using the following command:
db2 -f nn_aliases.sql -l aliases.log -t
5. Review the contents of the aliases.log file for any errors. Ignore
the errors related to dropping aliases.
----------------------------------------------------------------------
CR # 290003 - User's Guide
APAR # IY33142 - IMPLICIT AND EXPLICIT DATA CONVERSIONS FOR MATH
EXPRESSIONS
Problem:
The New Era of Networks Formatter chapter unclearly states:
NOTE: A math expression cannot be preceded by data type conversion
output operations unless the output data type is ASCII. For example,
the numeric format International Decimal cannot be preceded a
math expression.
Resolution:
More information is required in the note.
NOTE:
In an Output Operation Collection when the output data type is
different from the input data type, a data conversion needs to
occur. However, you can control the timing of the conversion.
If an Output Operation Collection specified in the Output Control
contains a Data/Convert operation, then the conversion is delayed
until the explicit conversion operation is reached. In this case,
all operations preceding the Data/Convert act on the parsed data,
and all operations following the Data/Convert act on the converted
data.
If there is no Data/Convert operation in the Output Operation
Collection, then an implicit conversion operation immediately
acts on the input data. Subsequently, all the other defined
operations in the collection act on the converted data.
Compare the following Output Operation Collection examples:
Collection_A
Math_Mult_by_2
Trim_neg_sign
Collection_B
Math_Mult_by_2
Trim_neg_sign
Convert
If the Output Control specifies IBM Zoned Decimal as the Data Type
and the input data is "12" (ASCII x'3132'), then Collection_A will
fail, but Collection_B will work. The math operations work on ASCII
numeric data only. Since Collection_A does not contain a convert
operation the data will be converted to Zoned Decimal before any other
operation occurs. The data reaching the Math operation will be Zoned
(EBCDIC x'f1f2') and the Math operation parser will not recognize it.
Collection_B contains a Convert operation, so the data flowing through
it will remain ASCII until the Convert operation is reached.
This enhancement enables Output Operation Collections to behave in a
consistent manner. In 4.11 certain operations, such as Math
Expressions, pulled their input from the formatter parsed tree
discarding output from any previous operations. This meant that these
operations had to be listed before other operations. Confusion
existed around the order in which operations could occur and numbers
could not be manipulated before a math expression. This was solved
by serializing the output from each operation, allowing the data to
flow through the collection with the input to each operation being
the serialized output of the previous operation. Implicit
conversion had to fit in this flow somewhere, so putting the implicit
conversion first in the flow was a conscious decision.
--------------------------------------------------------------------
CR # 291079 - System Management Guide
PMR # 60004
Problem:
In the Configuration chapter of the System Management Guide, the
following Logging Levels are stated in error:
0 - Logs informational messages, warnings, errors, and fatal
errors.
1 - Logs warnings, errors, and fatal errors.
2 - Logs errors and fatal errors.
3 - Logs fatal errors only.
Resolution:
The Logging Levels should read:
0 - Logs messages of any severity.
1 - Logs error and warning messages.
2 - Logs only error messages.
3 - Disables all logging.
----------------------------------------------------------------------
CR # 296736 - System Management Guide and CSD #3 Readme
APAR # IY38935 - IN RULES EXPORT FILES AN EXTRA SPACE AFTER AN EXIST
OPERATOR IN RULES EXPRESSION,CAUSES A CONFLICT IN NNRIE
Problem:
Both the System Management Guide and the CSD#3 readme incorrectly
document the use of the [-p] option.
Resolution:
The System Management Guide should read:
The [-p] option can only be used when migrating from MQSI 1.1 or
MQSI 2.01 on NT, AIX, Solaris, and HP-UX to WMQI 2.1 on z/OS.
All other migration scenarios do not support the use of the [-p]
option.
----------------------------------------------------------------------
CR # 300000 - System Management Guide
PMR # 65852
Problem:
The procedure for configuring the database set up on z/OS on pages 27,
29, 33 and 34 used incorrect terminology. The procedures talk about
"Connecting to the database server and the database."
Resolution:
The terminology should be changed from "database" to "table qualifier."
----------------------------------------------------------------------
APAR # IC34946 - DOCUMENTATION DOES NOT INCLUDE ALL SUPPORTED
CODEPAGES BY NNSY COMPONENT
Problem:
The documentation does not include all codepages that the NNSY
component supports. The documented list only identifies the code sets
that were tested; however, the component is designed to work with a
more comprehensive list. The NNSY component uses ICU 1.8 and is
designed to work with code sets that do not have the embedded NULL's
found in UCS-2, UCS-4, and UTF-16.
Resolution:
The Supported Code set table should include the following:
Code Set Code Set Alias Description
Identifier Name
--------------------------------------------------------------------
1208 ibm-1208 UTF8 Unicode transformation
utf-8 8 bit
ibm-1208
cp1208
======================================================================
CSD04
=====
Contents:
Part 1: New Era of Networks Formatter Issues
Part 2: New Era of Networks Rules Issues
Part 3: Documentation Issues
====================================================================
Part 1: New Era of Networks Formatter Issues
====================================================================
CR # 279277
APAR # IY31377 - IMPORT NEW ERA FORMATS FROM OLDER VERSION INTO V2.1
Problem:
On AIX/Oracle, when exporting rules and formats from MQSI 2.0.2
using NNRie and NNFie and importing them into WMQI 2.1, two
problems are encountered:
#1 Trailing spaces are generated in New Era of Network Formatter
components.
#2 New Era of Networks Rules Subscription names that contain the %
(percent) character fail to import.
Work-arounds:
#1 To correct the NNFie problem, delete the trailing spaces in the
NNFie export file from the component lines that failed to import.
#2 To correct the NNRie problem, remove the % character from the
Subscription names, do one of the following:
- Use the MQSI V2.02 GUI to remove the % characters and then
reexport the file.
- Open the NNRie export file and manually delete the %
characters from the Subscription names.
====================================================================
Part 2: New Era of Networks Rules Issues
====================================================================
CR # 275544
APAR # IC33620 - RULES EXPRESSION CHANGES WHEN IT IS APPLIED
Problem:
When clicking on the Apply button in the New Era of Networks
Formatter GUI, the expression changes resulting in an extra set of
parentheses, and pipes (|) changing to ampersands (&).
Resolution:
The code was fixed to correctly parse expressions.
--------------------------------------------------------------------
CR # 286132
APAR # IY32488 - EXCEPTION LIST OMITS NEON DETAILS
Problem:
When a message comes out of the failure terminal of a
NEONRulesEvaluation node, the Exception List contains an error
message (NN311E) that is written to the NNSYMessageLog.nml and the
(Unix) mqsisyslog or (NT) Event Log. However, it does not contain
the details of the error (#1001) which caused the problem.
Resolution:
The code was adjusted to provide error message details.
====================================================================
Part 3: Documentation Issues
====================================================================
CR # 260215 - System Management Guide
CMVC # 20626 - NN406I: Missing insert etc
Problem:
The location of the NNSYMessagelog.nml file is not indicated in
the current documentation.
Resolution:
On Unix
The NNSYMessageLog.nml file is written to the current working
directory at application startup. For broker logging, this file is
written to the current directory when mqsistart was issued.
For migration and test utility logging, such as msgtest, NNFie,
and NNRie, the NNSYMessageLog.nml file is written to the current
working directory.
On Windows NT
For broker and Control Center logging, the NNSYMessageLog.nml file
is written to \bin where wmqi_inst_loc is the location
of the WebSphere MQ Integrator installation; for example,
"C:\Program Files\IBM\WebSphere MQ Integrator 2.1".
For migration and test utility logging, such as msgtest, NNFie, and
NNRie, the NNSYMessageLog.nml file is written to the current working
directory.
--------------------------------------------------------------------
CR # 273641 - System Management Guide
CMVC # 25353 - Decimal values are allowed in a numeric field
Problem:
If you define an input control field with a data type as Numeric,
New Era of Networks Formatter parses with decimal values that are
inconsistent with the Numeric data type information described on
pages 171, 180, 186, 187 of the System Management Guide.
The SM Guide states:
"NUMERIC data types only support characters (0-9) and do not
support decimals. While an error might not occur with decimals,
the decimals may be ignored and the number will be truncated to a
whole number.
Resolution:
When using the Numeric data type in your input or output control
definitions, observe the following additional guidelines:
Input Control Definition Guidelines
- If you define an input control field value as a decimal, New Era
of Networks Formatter will not fail to parse.
Output Control Definition Guidelines
- If you want the decimal portion of the input date to appear
in an output message, select N/A or String as the data type in
your output control definition. New Era of Networks Rules will
ignore the decimal portion of the input data.
- If you do not want the decimal portion of the input date to appear
in an output message, select Numeric as the data type in your
output control definition.
----------------------------------------------------------------------
CR # 275197 - System Management Guide
PMR # 89584
Problem:
The following section of information is missing from the WMQI V2.1
System Management Guide.
Resolution:
Include the following information in the WMQI V2.1 System Management
Guide.
Assigning Permissions
To assign permissions, you must create more than one user in your
database.
In Oracle
The following procedure describes how to create database users in
Oracle. To access WMQI databases, users must be created
and associated with the user role that was created during the WMQI
installation.
The process for creating users in Oracle includes:
- Creating users
- Granting roles to users
- Creating synonyms for database objects
Creating Users
You must create users to identify individual user names in the
database.
To create a user, type the following command and press ENTER.
create user identified by ;
where USERNAME and PASSWORD are required parameters.
Granting Roles to Users
To give users permissions to access the database data, you can grant
permissions to an individual user or create roles with certain
permissions and associate users with specific roles.
A role was created for user access to the WMQI database
during the installation process. This role uses the following format:
_ROLE
where is your site-specific role name provided during
installation.
To grant the role to each created user:
1. Log in to the database as SYS.
2. Establish a connection to the Oracle instance using SQL Plus and
the SYS username and password.
3. Type the following command to grant the role to the new user:
grant _ROLE to ;
The _ROLE role is granted to the user identified by
USERNAME.
NOTE: For rules permissions, each user can have only one role granted
and this role must be the same for all users.
Creating Synonyms for Database Objects
To create synonyms for the WMQI database objects, users
must run a database script.
To run the WMQI synonyms SQL script:
1. Locate the nn_synonyms.sql script in the install.sql_rulfmt56\inst_db\or806\tmp
sub-directory of the installation directory. Log in to the directory
where the nn_synonyms.sql resides.
2. Run the nn_synonyms.sql script for each user using the following
command:
sqlplus @@ nn_synonyms.sql> synonyms.log
3. Review the contents of the synonyms.log file for any errors. Ignore
the errors related to dropping synonyms.
In Sybase or SQL Server
The following procedures describe how to create users in Sybase or
Microsoft SQL server.
The commands are run within the command line program isql, and all
references to SQL Server include both Sybase and SQL Server.
Except for changing passwords, these procedures are limited to either
the system administrator or database owner.
Users must have login accounts and a user name in each database they
want to access. Adding login accounts, database users, and object
permissions can be done by the system administrator, security officer,
or database owner. A single person can occupy one or more of these
roles. Check with your database administrator for information about your
specific environment.
The process includes:
- Creating login accounts
- Assigning users to a database
- Defining user groups
Creating Login Accounts
Login accounts give users access to the SQL Server. To access a database,
a login must be assigned to a user name to the databases the user wants
to access. Only a system administrator or security officer has the
authority to create login accounts.
To create a login account, enter the following:
sp_addlogin loginName, password [, defdb [, deflang [,sid [, encryptopt]]]]
where loginName is the sysname with no default,
where password is the login; login is sysname with no default,
where defdb is the default database for the login (database is sysname
with a default of master),
where deflang is the name of the default language assigned when a user
logs in to SQL Server (language is sysname with a default of NULL; if
language is not specified, language is set to the servers current default
language defined by the sp_configure variable default language).
NOTE: Changing the server default language does not change the default
language for existing logins. The language remains the same as the
default language used when the login was added.
where sid is the security identification number (SID) [sid is varbinary
(16) with a default of NULL. If sid is NULL, the system generates a SID
for the new login, specifying a value other than NULL, but the number must
not already exist. SID is useful when you are scripting or moving SQL
Server logins from one server to another and you want the logins to have
the same SID between servers.],
and where encryptopt specifies whether the password is encrypted when
stored in the system tables (encryption_option is varchr(20) and can be
one of these values:
Value Description
NULL (default) The password is encrypted.
skip_encryption The password is not encrypted. The supplied
password was encrypted by an earlier version of
SQL Server and is used for upgrade purposes only.
Assigning Users to a Database
To use a database, the server login must be associated with a user name
in the database. Only the database owner (DBO) can assign users to a
database.
From the database in which the user is to be added, enter the following:
sp_adduser loginName [, nameInDB] [, group]
where the loginName is the user's server login account,
where the nameInDB parameter is the name for the user in the database,
where the nameInDB defaults to the loginName if it is not specified,
and where the group enables the DBO to add the user to an existing group
in the database. If a group is not specified, the user is placed in the
default group, PUBLIC.
NOTES:
- To add permissions both in New Era of Networks Rules and in New Era of
Networks Formatter, you must add users as users not aliases. They must
be members of the same user group.
- SQL Server: Permission errors occur if the user is the same as the
database owner (DBO). the user must be part of NNSYGRP, but cannot be
the DBO.
Defining User Groups
During installation process, NNSYGRP is created for WMQI
users. Users must be linked to the NNSYGRP group to access the WMQI
databases. In addition, each database user, with the exception
of the database owner, must be granted permissions to access objects
within that database.
To add users to NNSYGRP, do one of the following:
- To use the sp_adduser procedure, enter the following:
sp_adduser loginName [, nameInDB] [, group]
where the loginName is the user's server login account,
where the nameInDB parameter is the name for the user in the database,
where the nameInDB defaults to the loginName if it is not specified,
and where the group enables the DBO to add the user to an existing
group in the database. If a group is not specified, the user is placed
in the default group, PUBLIC.
- To use the sp_changegroup system procedure, enter the following:
sp_changegroup groupName, userName
where groupName is the name of the group to which the user is added,
and where userName is the database user name.
In DB2 UDB
The following procedures describe how to create database users in DB2.
It includes the following:
- Creating users
- Creating aliases for database objects
To allow other users to have access to the objects created during
installation, a new user must be created with aliases to link the user's
objects to the objects in the installer's schema. New users are added to
the NNSYGRP group created during installation. You must be a member of
NNSYGRP in order to access any of the objects owned by another NNSYGRP
member.
Creating Users
To grant user access to the database, you must create additional user
accounts at the local system level.
Create additional users in the NNSYGRP group.
Creating Aliases for Database Objects
Each new user must run a database script to create aliases for the WMQI
database objects created during the WMQI database installation.
NOTE: To have assignable permissions, you must be a member of the
system group NNSYGRP.
Ask your database administrator to create aliases for OS/390.
To create aliases using the nn_aliases.sql script:
1. Connect to the WMQI database using the user ID of the user
you want to add.
2. Establish a DB2 CLP session.
Example:
db2 connect to user using
3. Locate the nn_aliases.sql script in the
\install.sql_rulfmt56\inst_db\tmp directory of the installation
directory.
4. Run the nn_aliases.sql script using the following command:
db2 -f nn_aliases.sql -l aliases.log -t
5. Review the contents of the aliases.log file for any errors. Ignore
the errors related to dropping aliases.
----------------------------------------------------------------------
CR # 290003 - User's Guide
APAR # IY33142 - IMPLICIT AND EXPLICIT DATA CONVERSIONS FOR MATH
EXPRESSIONS
Problem:
The New Era of Networks Formatter chapter unclearly states:
NOTE: A math expression cannot be preceded by data type conversion
output operations unless the output data type is ASCII. For example,
the numeric format International Decimal cannot be preceded a
math expression.
Resolution:
More information is required in the note.
NOTE:
In an Output Operation Collection when the output data type is
different from the input data type, a data conversion needs to
occur. However, you can control the timing of the conversion.
If an Output Operation Collection specified in the Output Control
contains a Data/Convert operation, then the conversion is delayed
until the explicit conversion operation is reached. In this case,
all operations preceding the Data/Convert act on the parsed data,
and all operations following the Data/Convert act on the converted
data.
If there is no Data/Convert operation in the Output Operation
Collection, then an implicit conversion operation immediately
acts on the input data. Subsequently, all the other defined
operations in the collection act on the converted data.
Compare the following Output Operation Collection examples:
Collection_A
Math_Mult_by_2
Trim_neg_sign
Collection_B
Math_Mult_by_2
Trim_neg_sign
Convert
If the Output Control specifies IBM Zoned Decimal as the Data Type
and the input data is "12" (ASCII x'3132'), then Collection_A will
fail, but Collection_B will work. The math operations work on ASCII
numeric data only. Since Collection_A does not contain a convert
operation the data will be converted to Zoned Decimal before any other
operation occurs. The data reaching the Math operation will be Zoned
(EBCDIC x'f1f2') and the Math operation parser will not recognize it.
Collection_B contains a Convert operation, so the data flowing through
it will remain ASCII until the Convert operation is reached.
This enhancement enables Output Operation Collections to behave in a
consistent manner. In 4.11 certain operations, such as Math
Expressions, pulled their input from the formatter parsed tree
discarding output from any previous operations. This meant that these
operations had to be listed before other operations. Confusion
existed around the order in which operations could occur and numbers
could not be manipulated before a math expression. This was solved
by serializing the output from each operation, allowing the data to
flow through the collection with the input to each operation being
the serialized output of the previous operation. Implicit
conversion had to fit in this flow somewhere, so putting the implicit
conversion first in the flow was a conscious decision.
--------------------------------------------------------------------
CR # 291079 - System Management Guide
PMR # 60004
Problem:
In the Configuration chapter of the System Management Guide, the
following Logging Levels are stated in error:
0 - Logs informational messages, warnings, errors, and fatal
errors.
1 - Logs warnings, errors, and fatal errors.
2 - Logs errors and fatal errors.
3 - Logs fatal errors only.
Resolution:
The Logging Levels should read:
0 - Logs messages of any severity.
1 - Logs error and warning messages.
2 - Logs only error messages.
3 - Disables all logging.
----------------------------------------------------------------------
CR # 296736 - System Management Guide
PMR # 43909
Problem:
Both the System Management Guide and the CSD#3 readme incorrectly
document the use of the [-p] option.
Resolution:
The System Management Guide should read:
The [-p] option can only be used when migrating from MQSI 1.1 or
MQSI 2.01 on NT, AIX, Solaris, and HP-UX to WMQI 2.1 on z/OS.
All other migration scenarios do not support the use of the [-p]
option.
----------------------------------------------------------------------
CR # 300000 - System Management Guide
PMR # 65852
Problem:
The procedure for configuring the database set up on z/OS on pages 27,
29, 33 and 34 used incorrect terminology. The procedures talk about
"Connecting to the database server and the database."
Resolution:
The terminology should be changed from "database" to "table qualifier."
CSD03
=====
Contents:
Part 1: General Issues
Part 2: New Era of Networks Formatter
Part 3: New Era of Networks Rules
Part 4: Documentation Issues
Part 1: General Issues
====================================================================
CMVC # 17303
CR # 253330
Problem:
When the ICU_DATA variable is not specified in a UNIX environment,
the DataFlowEngine (DFE) does not start successfully, repeatedly
re-starts, and writes a series of messages to the system log, such
as:
thehulk WMQIv210[25819]: [ID 702911 user.error]
(BROKER.Parallel)[1]BIP2203E: An execution group has encountered
a problem whilst starting. : BROKER.dc07ea0b-e600-0000-0080-d56401eb87f5:
/build/S210_P/src/DataFlowEngine/ImbMain.cpp: 491: main: :
thehulk WMQIv210[25834]: [ID 702911 user.error]
(BROKER.Parallel)[1]BIP2203E: An execution group has encountered
a problem whilst starting. : BROKER.dc07ea0b-e600-0000-0080-d56401eb87f5:
/build/S210_P/src/DataFlowEngine/ImbMain.cpp: 491: main: :
thehulk WMQIv210[25861]: [ID 702911 user.error]
(BROKER.Parallel)[1]BIP2203E: An execution group has encountered
a problem whilst starting. : BROKER.dc07ea0b-e600-0000-0080-d56401eb87f5:
/build/S210_P/src/DataFlowEngine/ImbMain.cpp: 491: main: :
Work-around:
To correct the problem, do one of the following:
- Set the ICU_DATA variable.
On AIX:
ICU_DATA=$NNSY_ROOT/share/icu/data
export ICU_DATA
On HP-UX :
export ICU_DATA=$NNSY_ROOT/share/icu/data
On Solaris:
ICU_DATA=$NNSY_ROOT/share/icu/data
export ICU_DATA
-Or add the following to the WMQI User Profile:
On AIX /usr/opt/wmqi/sample/profiles/profile.aix
On HP-UX /opt/wmqi/sample/profiles/profile.hpux
On Solaris /opt/wmqi/sample/profiles/profile.sol
--------------------------------------------------------------------
CMVC # 24380
CR # 265227
Description:
On AIX 4.3.3, the WMQI DataFlowEngine can fail when New Era of
Networks technology is used in the flow, and many threads are used.
The DataFlowEngine will fail with an abend file.
Work-around:
Run with fewer message processing threads (message flows plus
additional instances). You can also create more execution groups and
spread the message flows among the execution groups.
--------------------------------------------------------------------
CMVC # 24600
CR # 269179
Description:
On AIX, when DEBUG MALLOC is turned on and when NEON Nodes are
loaded during start-up of the Broker, there is corruption of the
heap.
Work-around:
Use MQSI_PRELOAD to verify that the ICU libraries are loaded when
the process starts. Set the following environment variable:
export MQSI_PRELOAD=/usr/lpp/nnsy/bin/
libicuin18.a:/usr/lpp/nnsy/bin/
libicuio18.a:/usr/lpp/nnsy/bin/
libicuio18.a:/usr/lpp/nnsy/share/icu/data/libicudt18b.a
--------------------------------------------------------------------
CMVC # 24629
CR # 267223
Description:
While uninstalling both the GA and CSD component versions or just
the CSD component version on HP-UX 11.00, the following error
message appears:
There is unmarked software which depends on software
that is marked for remove. The remaining software cannot
operate properly without the software you plan to remove.
Work-arounds:
On Windows NT
Ignore the dependency warning messages and continue with
uninstalling.
On HP-UX 11.00
To remove the CSD patch and leave the GA component:
1. Run swremove, click Options, and then Change Options to list
the remove options.
2. Deselect Enforce dependency analysis errors in agent and
click OK.
3. Scroll to the component in the Software Selection menu and do
one of the following:
- If the component was added by the CSD, then mark the component.
The component will be marked Yes.
- If the component was patched by the CSD, then press ENTER to
open the Software Selection sub-menu and mark the specific
patch level of the component to be removed. Select ..(go up).
The component will be marked Partial.
NOTE: If a Dependency Message Dialog appears when marking a
full component or a component patch level for removal, ignore
it and click OK.
4. Select Actions, Remove (analysis)..., OK, Yes, and Done.
5. Repeat steps 3 and 4 for each New Era of Networks component,
starting with the highest numbered component and working down
the list of components ending with the lowest numbered component
(for example, n7nrfdk52, n6nrfx52, ..., n0infr20).
To remove both the CSD and the GA components
1. Run swremove, click Options, and then Change Options to list
the remove options.
2. Deselect Enforce dependency analysis errors in agent and
click OK.
3. Scroll to the component in the Software Selection menu and mark
the component. The component will be marked Yes.
NOTE: If a Dependency Message Dialog appears when marking a full
component for removal, ignore it and click OK.
4. Select Actions, Remove (analysis)..., OK, Yes, and Done.
5. Repeat steps 3 and 4 until all New Era of Networks components are
removed from the "Software Selection" menu.
--------------------------------------------------------------------
CR # 253460
Description:
inst_db leaves files containing plain text passwords in its
directory structure.
Resolution:
Scripts were created to delete files containing plain text passwords
after an inst_db run. To use this feature of inst_db, run the
following:
o Windows NT inst_db.cmd
o UNIX inst_db.sh
These files run the inst_db program plus remove the text-encoded
password files from your inst_db directories.
---------------------------------------------------------------------
CR # 268274
Description:
On z/OS, database commits cannot be issued from a New Era of Networks
Formatter user exit because this impacts the WMQI broker.
Resolution:
On z/OS, the New Era of Networks Formatter user exits are now run in
a child thread, allowing them to issue database commits without
impacting the WMQI Broker.
NOTE: COBOL user exits compiled with version 3.0 and higher can run
in a multi-threaded environment. To run COBOL in a multi-threaded
environment, the keyword RECURSIVE must be specified in the COBOL
program on the PROGRAM ID statement, and the compiler option THREAD
must be specified in the compile step.
--------------------------------------------------------------------
PMR # 89584 - System Management Guide
CR # 275197
Description:
The following section of information is missing from the WMQI V2.1
System Management guide.
Resolution:
Include the following information in the WMQI V2.1 System Management
Guide.
Assigning Permissions
To assign permissions, you must create more than one user in your
database.
In Oracle
The following procedure describes how to create database users in
Oracle. To access New Era of Networks Rules and Formatter databases,
users must be created and associated with the user role that was
created during the New Era of Networks Rules and Formatter
installation.
The process for creating users in Oracle includes:
- Creating users
- Granting roles to users
- Creating synonyms for database objects
Creating Users
You must create users to identify individual user names in the
database.
To create a user, type the following command and press ENTER.
create user identified by ;
where USERNAME and PASSWORD are required parameters.
Granting Roles to Users
To give users permissions to access the database data, you can grant
permissions to an individual user or create roles with certain
permissions and associate users with specific roles.
A role was created for user access to the New Era of Networks Rules
and Formatter database during the installation process. This role
uses the following format:
_ROLE
where is your site-specific role name provided during
installation.
To grant the role to each created user:
1. Log in to the database as SYS.
2. Establish a connection to the Oracle instance using SQL Plus and
the SYS username and password.
3. Type the following command to grant the role to the new user:
grant _ROLE to ;
The _ROLE role is granted to the user identified by
USERNAME.
NOTE: For rules permissions, each user can have only one role granted
and this role must be the same for all users.
Creating Synonyms for Database Objects
To create synonyms for the New Era of Networks Rules and Formatter
database objects, users must run a database script.
To run the New Era of Networks synonyms SQL script:
1. Locate the nn_synonyms.sql script in the install.sql_rulfmt56\inst_db\or806\tmp
sub-directory of the installation directory. Log in to the directory
where the nn_synonyms.sql resides.
2. Run the nn_synonyms.sql script for each user using the following
command:
sqlplus @@ nn_synonyms.sql> synonyms.log
3. Review the contents of the synonyms.log file for any errors. Ignore
the errors related to dropping synonyms.
In Sybase or SQL Server
The following procedures describe how to create users in Sybase or
Microsoft SQL server.
The commands are run within the command line program isql, and all
references to SQL Server include both Sybase and SQL Server.
Except for changing passwords, these procedures are limited to either
the system administrator or database owner.
Users must have login accounts and a user name in each database they
want to access. Adding login accounts, database users, and object
permissions can be done by the system administrator, security officer,
or database owner. A single person can occupy one or more of these
roles. Check with your database administrator for information about your
specific environment.
The process includes:
- Creating login accounts
- Assigning users to a database
- Defining user groups
Creating Login Accounts
Login accounts give users access to the SQL Server. To access a database,
a login must be assigned to a user name to the databases the user wants
to access. Only a system administrator or security officer has the
authority to create login accounts.
To create a login account, enter the following:
sp_addlogin loginName, password [, defdb [, deflang [,sid [, encryptopt]]]]
where loginName is the sysname with no default,
where password is the login; login is sysname with no default,
where defdb is the default database for the login (database is sysname
with a default of master),
where deflang is the name of the default language assigned when a user
logs in to SQL Server (language is sysname with a default of NULL; if
language is not specified, language is set to the servers current default
language defined by the sp_configure variable default language).
NOTE: Changing the server default language does not change the default
language for existing logins. The language remains the same as the
default language used when the login was added.
where sid is the security identification number (SID) [sid is varbinary
(16) with a default of NULL. If sid is NULL, the system generates a SID
for the new login, specifying a value other than NULL, but the number must
not already exist. SID is useful when you are scripting or moving SQL
Server logins from one server to another and you want the logins to have
the same SID between servers.],
and where encryptopt specifies whether the password is encrypted when
stored in the system tables (encryption_option is varchr(20) and can be
one of these values:
Value Description
NULL (default) The password is encrypted.
skip_encryption The password is not encrypted. The supplied
password was encrypted by an earlier version of
SQL Server and is used for upgrade purposes only.
Assigning Users to a Database
To use a database, the server login must be associated with a user name
in the database. Only the database owner (DBO) can assign users to a
database.
From the database in which the user is to be added, enter the following:
sp_adduser loginName [, nameInDB] [, group]
where the loginName is the user's server login account,
where the nameInDB parameter is the name for the user in the database,
where the nameInDB defaults to the loginName if it is not specified,
and where the group enables the DBO to add the user to an existing group
in the database. If a group is not specified, the user is placed in the
default group, PUBLIC.
NOTES:
- To add permissions both in New Era of Networks Rules and in New Era of
Networks Formatter, you must add users as users not aliases. They must
be members of the same user group.
- SQL Server: Permission errors occur if the user is the same as the
database owner (DBO). the user must be part of NNSYGRP, but cannot be
the DBO.
Defining User Groups
During installation process, NNSYGRP is created for New Era of Networks
Rules and Formatter users. Users must be linked to the NNSYGRP group to
access the New Era of Networks Rules and Formatter databases. In
addition, each database user, with the exception of the database owner,
must be granted permissions to access objects within that database.
To add users to NNSYGRP, do one of the following:
- To use the sp_adduser procedure, enter the following:
sp_adduser loginName [, nameInDB] [, group]
where the loginName is the user's server login account,
where the nameInDB parameter is the name for the user in the database,
where the nameInDB defaults to the loginName if it is not specified,
and where the group enables the DBO to add the user to an existing
group in the database. If a group is not specified, the user is placed
in the default group, PUBLIC.
- To use the sp_changegroup system procedure, enter the following:
sp_changegroup groupName, userName
where groupName is the name of the group to which the user is added,
and where userName is the database user name.
In DB2 UDB
The following procedures describe how to create database users in DB2.
It includes the following:
- Creating users
- Creating aliases for database objects
To allow other users to have access to the objects created during
installation, a new user must be created with aliases to link the user's
objects to the objects in the installer's schema. New users are added to
the NNSYGRP group created during installation. You must be a member of
NNSYGRP in order to access any of the objects owned by another NNSYGRP
member.
Creating Users
To grant user access to the database, you must create additional user
accounts at the local system level.
Create additional users in the NNSYGRP group.
Creating Aliases for Database Objects
Each new user must run a database script to create aliases for the New Era
of Networks Rules and Formatter database objects created during the New
Era of Networks Rules and Formatter database installation.
NOTE: To have assignable permissions, you must be a member of the
system group NNSYGRP.
Ask your database administrator to create aliases for OS/390.
To create aliases using the nn_aliases.sql script:
1. Connect to the New Era of Networks Rules and Formatter database using
the user ID of the user you want to add.
2. Establish a DB2 CLP session.
Example:
db2 connect to user using
3. Locate the nn_aliases.sql script in the
\install.sql_rulfmt56\inst_db\tmp directory of the installation
directory.
4. Run the nn_aliases.sql script using the following command:
db2 -f nn_aliases.sql -l aliases.log -t
5. Review the contents of the aliases.log file for any errors. Ignore
the errors related to dropping aliases.
--------------------------------------------------------------------
CMVC # 17674
CR # 253717
Problem:
An invalid neonreg.dat file specifying the machine name parameter
instead of the DB2 database name parameter caused NNFie and NNRie
to be terminated by Windows NT.
Resolution:
Correct the nnsyreg.dat file.
____________________________________________________________________
CMVC # 23428
CR # 263272
Problem:
The -C optional parameter is not working properly in NNFie. This
option specifies the file which contains the program arguments.
Resolution:
NNFie was fixed to handle the -C option properly.
NOTE: If your Text Editor does not automatically insert a new line
at the end of a command file by default, you must manually enter the
a new line character (x0A). If the new line character is not
present, NNFie fails when using the -C option.
________________________________________________________________________
CMVC # 26497
CR # 270160
Problem:
There has been internal security updates in oracle 9i. This has
caused inst_db with 9i to malfunction.
Resolution:
Fixed problems from running the inst_db program as system. This is
with a small documented step that the dba will have to run.
To run inst_db in an Oracle 9i database to install the rulfmt56
schema, do the following:
1. Run sqlplus as sys user.
2. Issue the following command inside sqlplus:
GRANT SELECT on dba_role_privs to system with grant option
3. To create a new rulfmt database schema, run inst_db as the system
rather than as the sys account.
______________________________________________________________________________
CMVC # IC30970
Using NEON domain within a globally coordinated transactional message
flow.
There are known problems when using DB2 database and working with the
NEON domain (NEON parser, NEONRules and NEONFormatter nodes) within a
message flow which is globally coordinated.
To overcome these problems, take the following action:
1. Install CSD3
2. Edit your nnsyreg.dat file
On Unix:
Edit Session.MQSI_PLUGIN section
replace NNOT_SHARED_LIBRARY = dbt26db250
with NNOT_SHARED_LIBRARY = dbt26mqi
and NNOT_FACTORY_FUNCTION = NNSesDB2Factory
with NNOT_FACTORY_FUNCTION = NNSesODBCFactory
On Win32:
replace NNOT_SHARED_LIBRARY = dbt26db250
with NNOT_SHARED_LIBRARY = dbt26msodbc
and NNOT_FACTORY_FUNCTION = NNSesDB2Factory
with NNOT_FACTORY_FUNCTION = NNSesODBCFactory
3. Ensure that your Rules and Formats database is defined as
an ODBC datasource either in .odbc.ini file (unix) or in the
Windows GUI (control panel->ODBC dataSources)
4. Set the following environment variable
MQSI_NNSY_XA=1
NOTES:
1. When using NNFie or NNRie, you are recommended to not use the
MQSI_PLUGIN session but to have a separate session for these tools,
containing the following values:
NNOT_SHARED_LIBRARY = dbt26db250
NNOT_FACTORY_FUNCTION = NNSesDB2Factory
2. If you wish to have some message flows coordinated and some not
coordinated, you are recommended to have them in separate execution
groups. i.e. do not mix coordinated and non coordinated message flows
in the same execution group.
3. NEONMSG domain ( i.e. NEONMSG parser, NEONRulesEvaluation, NEONMap
NEONTransform nodes) cannot, as yet, be used in a globally coordinated
message flow.
______________________________________________________________________________
CMVC # 27810
Stress problems with NNSY & Oracle9i
When using New Era of Networks nodes/parsers you may observe
problems running large numbers of message flows in an execution group.
Symptoms of this problem include a BIP2228 message written to the
syslog log, e.g.
WMQIv210[21988]: (BROKER.default)[3085]BIP2228E: Severe error:
build/S210_D/src/CommonServices/Unix/ImbAbend.cpp 417 signal
received Abend file: var/mqsi/errors/BROKER.default.21988.3085.abend
action: abort
If you experience any such problems, reduce the number of message flows
in each execution group.
______________________________________________________________________________
Part 2: New Era of Networks Formatter
====================================================================
CMVC # 23080, 23098
CR # 262539, 262603
Description:
On z/OS, an export file cannot be FTP or imported from a different
operating system from which it was initially exported.
Resolution:
To migrate formats and rules between platforms, observe the
following guidelines:
- Use the NNFie and NNRie utilities provided with New Era of
Networks Support for WMQI to migrate rules and formats between
platforms.
- Do not transfer export files that use different code pages
between platforms by ftp.
NOTE: Characters that are in the syntactic character set, that
is, characters whose code points are invariant between most code
pages may not be converted correctly during an ftp transfer.
For example, the exclamation mark (!) is at code point 0x5A in
some EBCDIC code pages and can be converted as the pipe symbol
(|). NNFie and NNRie use the ICU function calls to convert
character data. These calls are more sensitive to variations
in code points between code pages.
For more information, see the WMQI 5.6 System Management Guide,
Process Overview section.
__________________________________________________________________
CMVC # 26039
CR # 275396
Problem:
A pre-CSD #3 version of NNFie cannot be used to import a file from
the latest version of NNFie in CSD #3.
Work-around:
1. Export your formats using a pre-CSD #3 WMQI V2.1 NNFie version.
or 2. Import the export file using WMQI V2.1 CSD #3 NNFie version.
Part 3: New Era of Networks Rules
====================================================================
CMVC # 25354 - New Era of Networks Formatter Online Help
CR # 273640
Problem:
The New Era of Networks Formatter online help does not describe
Numeric Output Data Types.
Resolution:
The following information describes the Numeric Output Data Types:
A numeric string is a sequence of characters encoded in the native
encoding for the machine on which New Era of Networks Formatter
executes. A numeric string contains only the characters 0 to 9
(no + or -).
_________________________________________________________________________
CMVC # 20626 - System Management Guide
CR # 260215
Problem:
The location of the NNSYMessagelog.nml file is not indicated in
the current documentation.
Resolution:
On Unix,
The NNSYMessageLog.nml file is written to the current working
directory at application startup. For broker logging, this file
is written to the current directory when mqsistart was issued.
For migration and test utility logging, e.g. msgtest, NNFie &
NNRie- The NNSYMessageLog.nml file is written to the current
working directory.
On NT,
For broker & Control Center logging, the NNSYMessageLog.nml file
is written to \bin where wmqi_inst_loc is the
location of the Webspher MQ Integrator installation ,
e.g. "C:\Program Files\IBM\WebSphere MQ Integrator 2.1".
For migration and test utility logging, e.g. msgtest, NNFie &
NNRie- The NNSYMessageLog.nml file is written to the current working
directory.
Part 4: Documentation Issues
====================================================================
CR # 264846, 273815 System Management Guide
Description:
The following changes are required in Chapter 4, Formats and Rules
Database Migration under the section Migrating Between Versions and
Platforms:
1) On page 110, instead of the paragraph 'To export data....'
substitute:
If your export files are from a Version 1.1 database and
you want to represent your literal values as ASCII strings, you
must include the -p option when you run NNFie. You cannot use the
-p option when importing from a Version 2.0.2 or 2.1 database.
2) On page 111, the sentence beginning, 'For more information...'
needs to be removed.
3) This information must be added to the following z/OS migration
procedures:
- from 1.1/2.0.1 Windows NT to 2.1 z/OS (bottom of p.114)
- from 2.0.2 Windows NT to 2.1 z/OS (middle of p.115)
- from 1.1/2.0.1 AIX/HP/Solaris to 2.1 z/OS (bottom of p.115)
- from 2.0.2 AIX/HP/Solaris to 2.1 z/OS (bottom of p.116)
For Windows NT Version 2.1, you must use an intermediate
machine with Windows NT and WebSphere MQ Integrator to complete
this migration.
a. Create the nnfie.exp and nnrie.exp export files by running
the NNFie and NNRie utilities on the source machine.
b. Transfer these files to the intermediate machine.
c. To import to WebSphere MQ Integrator for z/OS v2.1, use the
version 2.1 NNFie and NNRie utilities on the intermediate
machine while using an ODBC connection to import to the DB2
database on the target machine.
----------------------------------------------------------------------
PMR # 07191 - System Management Guide
Description:
The WMQI 5.6 System Management Guide contains some errors.
Resolution:
Page 27, 29, 32, and 34: The NNOT_SHARED_LIBRARY items are incorrect.
The tables should read:
--------------------------------------------------------------
Database Type NNOT_SHARED_LIBRARY NNOT_FACTORY_FUNCTION
--------------------------------------------------------------
Microsoft SQL
Server 6.5, 7 dbt26sq165 NNSesMS6Factory
Oracle 8 dbt26or806 NNSesOra8Factory
Sybase CTLIB dbt26syb11 NNSesSybCTFactory
DB2 dbt26db250 NNSesDB2Factory
Page 134: The path for the example Makefile is incorrect.
o Verify that the following drives are mapped. These drives are
supplied in the $(INSTALL)/examples/Makefile.
K:\ - The location of the appropriate database installation files.
L:\ - The Microsoft Visual Studio 6.0 directory including vc98/bin
and vc98/lib.
The information should read:
Verify that the following drives are mapped. These drives are
defined in the example Makefile. To locate the example go to:
nnsy\rulfmt56\examples\NNSYrf directory
CSD02
=====
Contents:
Part 1: General Issues
Part 2: New Era of Networks Formatter
Part 3: New Era of Networks Rules
Part 4: Documentation Issues
====================================================================
Part 1: General Issues
====================================================================
CMVC # 17303
CR # 253330
Description:
If the ICU_DATA variable is not specified in a UNIX environment,
DataFlowEngine (DFE) crashes.
Work-around:
You must specify the ICU_DATA environment variable in a UNIX
environment. This defect will be resolved in a future patch release.
--------------------------------------------------------------------
CMVC # 22900
CR # 261619
Description:
You can create export files containing string literals x'0A' and
x'1A', but these characters can cause problems when reimporting.
Value x'0A' causes the record to split onto another line in the
export file, and value x'1A' causes the following error to be
written to the nnfie.log when importing on an ASCII platform:
ERROR: Failed to separate and fetch a piece of the input data.
Resolution:
NNFie was modified to only write HEX-encoded literal values to the
NNFie export file. This modification stops the export file from
being corrupted.
--------------------------------------------------------------------
CMVC # 24145
CR # 266308
Description:
On AIX 4.3.3, the WMQI Data Flow Engine can fail when using Sybase
as a New Era of Networks repository. The Data Flow Engine can abend.
Work-around:
Run with fewer message processing threads (message flows plus
additional instances). You can also create more execution groups and
spread the message flows among the execution groups.
--------------------------------------------------------------------
CMVC # 24600
CR # 269179
Description:
On AIX, when DEBUG MALLOC is turned on and when NEON Nodes are
loaded during start-up of the Broker, there is corruption of the
heap.
Work-around:
Use MQSI_PRELOAD to verify that the ICU libraries are loaded when
the process starts. Set the following environment variable:
export MQSI_PRELOAD=/usr/lpp/nnsy/bin/
libicuin18.a:/usr/lpp/nnsy/bin/
libicuio18.a:/usr/lpp/nnsy/bin/
libicuio18.a:/usr/lpp/nnsy/share/icu/data/libicudt18b.a
--------------------------------------------------------------------
CMVC # 24629
CR # 267223
Description:
While uninstalling both the GA and CSD component versions or just
the CSD component version on HP-UX 11.00, the following error
message appears:
There is unmarked software which depends on software
that is marked for remove. The remaining software cannot
operate properly without the software you plan to remove.
Work-arounds:
To remove the CSD patch and leave the GA component:
1. Run swremove, click Options, and then Change Options to list
the remove options.
2. Deselect Enforce dependency analysis errors in agent and
click OK.
3. Scroll to the component in the Software Selection menu and do
one of the following:
- If the component was added by the CSD, then mark the component.
The component will be marked Yes.
- If the component was patched by the CSD, then press ENTER to
open the Software Selection sub-menu and mark the specific
patch level of the component to be removed. Select ..(go up).
The component will be marked Partial.
NOTE: If a Dependency Message Dialog appears when marking a
full component or a component patch level for removal, ignore
it and click OK.
4. Select Actions, Remove (analysis)..., OK, Yes, and Done.
5. Repeat steps 3 and 4 for each New Era of Networks component,
starting with the highest numbered component and working down
the list of components ending with the lowest numbered component
(for example, n7nrfdk52, n6nrfx52, ..., n0infr20).
To remove both the CSD and the GA components
1. Run swremove, click Options, and then Change Options to list
the remove options.
2. Deselect Enforce dependency analysis errors in agent and
click OK.
3. Scroll to the component in the Software Selection menu and mark
the component. The component will be marked Yes.
NOTE: If a Dependency Message Dialog appears when marking a full
component for removal, ignore it and click OK.
4. Select Actions, Remove (analysis)..., OK, Yes, and Done.
5. Repeat steps 3 and 4 until all New Era of Networks components are
removed from the "Software Selection" menu.
This problem will be resolved in a future release.
--------------------------------------------------------------------
CMVC # 24635
CR # 269174
Description:
When moving files across operating systems, string literals are
subject to code page changes. This can cause an import failure even
when they have been changed to Binary.
Resolution:
All literals are now stored as HEX in the export files.
--------------------------------------------------------------------
CR # 253460
Description:
inst_db leaves files containing plain text passwords in its
directory structure.
Resolution:
Scripts were created to delete files containing plain text passwords
after an inst_db run. To use this feature of inst_db, run the
following:
o Windows NT inst_db.cmd
o UNIX inst_db.sh
These files run the inst_db program plus remove the text-encoded
password files from your inst_db directories.
---------------------------------------------------------------------
CR # 268274
Description:
On z/OS, database commits cannot be issued from a New Era of Networks
Formatter user exit because this impacts the WMQI broker.
Resolution:
On z/OS, the New Era of Networks Formatter user exits are now run in
a child thread, allowing them to issue database commits without
impacting the WMQI Broker.
NOTE: COBOL user exits compiled with version 3.0 and higher can run
in a multi-threaded environment. To run COBOL in a multi-threaded
environment, the keyword RECURSIVE must be specified in the COBOL
program on the PROGRAM ID statement, and the compiler option THREAD
must be specified in the compile step.
====================================================================
Part 2: New Era of Networks Formatter
====================================================================
CMVC # 22158
CR # 257695
Description:
If repeat count field is not within the immediate sibling format,
parsing of repeating formats fails.
Resolution:
The algorithm has been modified so that if the repeat count field is
not found in any immediate sibling, it walks up the format
hierarchy in search of the field.
--------------------------------------------------------------------
CMVC # 22898
CR # 261617
Description:
On z/OS, the New Era of Networks Formatter GUI is unable to show
conflicts when they exist.
Resolution:
NNFie failed to parse the command sent by the New Era of Networks
Formatter GUI to report conflicts. NNFie was fixed to handle the
command properly.
--------------------------------------------------------------------
CMVC # 23080, 23098
CR # 262539, 262603
Description:
On z/OS, an export file cannot be FTP or imported from a different
operating system from which it was initially exported.
Resolution:
To migrate formats and rules between platforms, observe the
following guidelines:
- Use the NNFie and NNRie utilities provided with New Era of
Networks Support for WMQI to migrate rules and formats between
platforms.
- Do not transfer export files that use different code pages
between platforms by ftp.
NOTE: Characters that are in the syntactic character set, that
is, characters whose code points are invariant between most code
pages may not be converted correctly during an ftp transfer.
For example, the exclamation mark (!) is at code point 0x5A in
some EBCDIC code pages and can be converted as the pipe symbol
(|). NNFie and NNRie use the ICU function calls to convert
character data. These calls are more sensitive to variations
in code points between code pages.
For more information, see the WMQI 5.6 System Management Guide,
Process Overview section.
--------------------------------------------------------------------
CMVC # 23881
CR # 266591
Problem 1:
The New Era of Networks Formatter GUI fails to import or export if
an nnfie.err file exists anywhere in the system path.
Resolution 1:
The GUI now deals with the nnfie.err file correctly and ignores
files that are not related to the current import/export session.
Problem 2:
Using NNFie to import from the command line can cause the following
error:
Assertion failed: CNTL_NAME && CNTL_NAME[0].
Resolution 2:
A fix was made in NNFie to eliminate the error.
Problem 3:
Using an Oracle 8.1.7 client, the New Era of Networks Formatter GUI
can cause an assert on exit:
'DDE Server Window: nnfmt.exe - Application Error.
Resolution 3:
A change was made in the database session handling code to fix this.
---------------------------------------------------------------------
CR # 260376
Description:
It may appear that the Import/Export function of the New Era of
Networks Formatter GUI does not work. In actuality, the problem lies
in the path with the nnfie.err file.
Resolution:
The New Era of Networks Formatter GUI properly handles the files
that NNFie creates and ignores files not associated with the session
that the user is currently in.
--------------------------------------------------------------------
CR # 268922
Description:
When editing maps, errors display when no errors have actually
occurred. The New Era of Networks Formatter GUI is attempting to
retrieve rows from the database by passing the incorrect retrieval
argument.
Resolution:
Now the correct argument is passed to the database.
====================================================================
Part 3: New Era of Networks Rules
====================================================================
CMVC # 23774
CR # 264519
Description:
When using the Import/Export feature in the New Era of Networks
Rules GUI, a db password is left in nnrie.dat file.
Resolution:
Remove the temporary nnrie.dat file when the import/export dialog
terminates.
====================================================================
Part 4: Documentation Issues
====================================================================
CMVC # 25618
CR # 264846, 273815 System Management Guide
Description:
The following changes are required in Chapter 4, Formats and Rules
Database Migration under the section Migrating Between Versions and
Platforms:
1) On page 110, instead of the paragraph 'To export data....'
substitute:
If your export files are from a Version 1.1 database and
you want to represent your literal values as ASCII strings, you
must include the -p option when you run NNFie. You cannot use the
-p option when exporting from a Version 2.0.1 or 2.1 database.
2) On page 111, the sentence beginning, 'For more information...'
needs to be removed.
3) This information must be added to the following z/OS migration
procedures:
- from 1.1/2.0.1 Windows NT to 2.1 z/OS (bottom of p.114)
- from 2.0.2 Windows NT to 2.1 z/OS (middle of p.115)
- from 1.1/2.0.1 AIX/HP/Solaris to 2.1 z/OS (bottom of p.115)
- from 2.0.2 AIX/HP/Solaris to 2.1 z/OS (bottom of p.116)
For Windows NT Version 2.1, you must use an intermediate
machine with Windows NT and WebSphere MQ Integrator to complete
this migration.
a. Create the nnfie.exp and nnrie.exp export files by running
the NNFie and NNRie utilities on the source machine.
b. Transfer these files to the intermediate machine.
c. To import to WebSphere MQ Integrator for z/OS v2.1, use the
version 2.1 NNFie and NNRie utilities on the intermediate
machine while using an ODBC connection to import to the DB2
database on the target machine.
----------------------------------------------------------------------
CR # 275197 System Management Guide
APAR# IY29253 - V5.6 VERSION OF THE NNSY GUIDES MISSING STEPS ON HOW TO
GIVE MULTIPLE USERS DB ACCESS VIA TABLE ALIASES IN DB2.
Description:
The section of information, Assigning Permissions, is missing from the
WMQI V2.1 System Management guide.
Resolution:
The following information needs to be added to the Configuration
chapter:
Assigning Permissions
To assign permissions, you must create more than one user in your
database.
In Oracle
The following procedure describes how to create database users in
Oracle. To access e-Biz Integrator databases, users must be created
and associated with the user role that was created during the e-Biz
Integrator installation.
The process for creating users in Oracle includes:
- Creating users
- Granting roles to users
- Creating synonyms for database objects
Creating Users
You must create users to identify individual user names in the
database.
To create a user, type the following command and press ENTER.
create user identified by ;
where USERNAME and PASSWORD are required parameters.
Granting Roles to Users
To give users permissions to access the database data, you can grant
permissions to an individual user or create roles with certain
permissions and associate users with specific roles.
A role was created for user access to the e-Biz Integrator database
during the installation process. This role uses the following format:
_ROLE
where is your site-specific role name provided during
installation.
To grant the role to each created user:
1. Log in to the database as SYS.
2. Establish a connection to the Oracle instance using SQL Plus and
the SYS username and password.
3. Type the following command to grant the role to the new user:
grant _ROLE to ;
The _ROLE role is granted to the user identified by
USERNAME.
NOTE: For rules permissions, each user can have only one role granted
and this role must be the same for all users.
Creating Synonyms for Database Objects
To create synonyms for the e-Biz Integrator database objects, users
must run a database script.
To run the e-Biz Integrator synonyms SQL script:
1. Locate the nn_synonyms.sql script in the install.sql_rulfmt56\inst_db\tmp
sub-directory of the installation directory. Log in to the directory
where the nn_synonyms.sql resides.
2. Run the nn_synonyms.sql script for each user using the following
command:
sqlplus @@ nn_synonyms.sql> synonyms.log
3. Review the contents of the synonyms.log file for any errors. Ignore
the errors related to dropping synonyms.
In Sybase or SQL Server
The following procedures describe how to create users in Sybase or
Microsoft SQL server.
The commands are run within the command line program isql, and all
references to SQL Server include both Sybase and SQL Server.
Except for changing passwords, these procedures are limited to either
the system administrator or database owner.
Users must have login accounts and a user name in each database they
want to access. Adding login accounts, database users, and object
permissions can be done by the system administrator, security officer,
or database owner. A single person can occupy one or more of these
roles. Check with your database administrator for information about your
specific environment.
The process includes:
- Creating login accounts
- Assigning users to a database
- Defining user groups
Creating Login Accounts
Login accounts give users access to the SQL Server. To access a database,
a login must be assigned to a user name to the databases the user wants
to access. Only a system administrator or security officer has the
authority to create login accounts.
To create a login account, enter the following:
sp_addlogin loginName, password [, defdb [, deflang [,sid [, encryptopt]]]]
where loginName is the sysname with no default,
where password is the login; login is sysname with no default,
where defdb is the default database for the login (database is sysname
with a default of master),
where deflang is the name of the default language assigned when a user
logs in to SQL Server (language is sysname with a default of NULL; if
language is not specified, language is set to the servers current default
language defined by the sp_configure variable default language).
NOTE: Changing the server default language does not change the default
language for existing logins. The language remains the same as the
default language used when the login was added.
where sid is the security identification number (SID) [sid is varbinary
(16) with a default of NULL. If sid is NULL, the system generates a SID
for the new login, specifying a value other than NULL, but the number must
not already exist. SID is useful when you are scripting or moving SQL
Server logins from one server to another and you want the logins to have
the same SID between servers.],
and where encryptopt specifies whether the password is encrypted when
stored in the system tables (encryption_option is varchr(20) and can be
one of these values:
Value Description
NULL (default) The password is encrypted.
skip_encryption The password is not encrypted. The supplied
password was encrypted by an earlier version of
SQL Server and is used for upgrade purposes only.
Assigning Users to a Database
To use a database, the server login must be associated with a user name
in the database. Only the database owner (DBO) can assign users to a
database.
From the database in which the user is to be added, enter the following:
sp_adduser loginName [, nameInDB] [, group]
where the loginName is the user's server login account,
where the nameInDB parameter is the name for the user in the database,
where the nameInDB defaults to the loginName if it is not specified,
and where the group enables the DBO to add the user to an existing group
in the database. If a group is not specified, the user is placed in the
default group, PUBLIC.
NOTES:
- To add permissions both in New Era of Networks Rules and in New Era of
Networks Formatter, you must add users as users not aliases. They must
be members of the same user group.
- SQL Server: Permission errors occur if the user is the same as the
database owner (DBO). the user must be part of NNSYGRP, but cannot be
the DBO.
Defining User Groups
During installation process, NNSYGRP is created for e-Biz Integrator
users. Users must be linked to the NNSYGRP group to access the e-Biz
Integrator databases. In addition, each database user, with the exception
of the database owner, must be granted permissions to access objects
within that database.
To add users to NNSYGRP, do one of the following:
- To use the sp_adduser procedure, enter the following:
sp_adduser loginName [, nameInDB] [, group]
where the loginName is the user's server login account,
where the nameInDB parameter is the name for the user in the database,
where the nameInDB defaults to the loginName if it is not specified,
and where the group enables the DBO to add the user to an existing
group in the database. If a group is not specified, the user is placed
in the default group, PUBLIC.
- To use the sp_changegroup system procedure, enter the following:
sp_changegroup groupName, userName
where groupName is the name of the group to which the user is added,
and where userName is the database user name.
In DB2 UDB
The following procedures describe how to create database users in DB2.
It includes the following:
- Creating users
- Creating aliases for database objects
To allow other users to have access to the objects created during
installation, a new user must be created with aliases to link the user's
objects to the objects in the installer's schema. New users are added to
the NNSYGRP group created during installation. You must be a member of
NNSYGRP in order to access any of the objects owned by another NNSYGRP
member.
Creating Users
To grant user access to the database, you must create additional user
accounts at the local system level.
Create additional users in the NNSYGRP group.
Creating Aliases for Database Objects
Each new user must run a database script to create aliases for the e-Biz
Integrator database objects created during the e-Biz Integrator database
installation.
NOTE: To have assignable permissions, you must be a member of the system
group NNSYGRP.
Ask your database administrator to create aliases for OS/390.
To create aliases using the nn_aliases.sql script:
1. Connect to the e-Biz Integrator database using the user ID of the user
you want to add.
2. Establish a DB2 CLP session.
Example:
db2 connect to user using
3. Locate the nn_aliases.sql script in the \install.sql_rulfmt56\inst_db\tmp
directory of the installation directory.
4. Run the nn_aliases.sql script using the following command:
db2 -f nn_aliases.sql -l aliases.log -t
5. Review the contents of the aliases.log file for any errors. Ignore the
errors related to dropping aliases.
----------------------------------------------------------------------
PMR # 07191 - System Management Guide
Description:
The WMQI 5.6 System Management Guide contains some errors.
Resolution:
Page 27, 29, 32, and 34: The NNOT_SHARED_LIBRARY items are incorrect.
The tables should read:
--------------------------------------------------------------
Database Type NNOT_SHARED_LIBRARY NNOT_FACTORY_FUNCTION
--------------------------------------------------------------
Microsoft SQL
Server 6.5, 7 dbt26sq165 NNSesMS6Factory
Oracle 8 dbt26or806 NNSesOra8Factory
Sybase CTLIB dbt26syb11 NNSesSybCTFactory
DB2 dbt26db250 NNSesDB2Factory
Page 134: The path for the example Makefile is incorrect.
o Verify that the following drives are mapped. These drives are
supplied in the $(INSTALL)/examples/Makefile.
K:\ - The location of the appropriate database installation files.
L:\ - The Microsoft Visual Studio 6.0 directory including vc98/bin
and vc98/lib.
The information should read:
Verify that the following drives are mapped. These drives are
defined in the example Makefile. To locate the example go to:
nnsy\rulfmt56\examples\NNSYrf directory
======================================================================
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
SERVICE AND TECHNICAL SUPPORT
-----------------------------
This is the first release of WebSphere MQ Integrator Version 2.1 for
Sun Solaris, 5724A8202. It supersedes IBM MQSeries Integrator
Version 2 for Sun Solaris, 5765D6700. The following APARs have been
incorporated into it.
Name Abstract
IC26734 - SQL ERRORS OCCUR WHEN WORKING WITH DATABASEUPDATE NODE
IC26799 - BIP2293 WHEN DEPLOYING A MESSAGE USING PLUG-IN PARSER
IC26880 - FAILURE TO START/STOP ALL MESSAGE FLOWS FOR AN EXECUTION GRP
IC26952 - CREATION OF RESPONSE MSG FAILS DUE TO NULL NAMEVALUECCSID
IC27033 - TRACE DOES NOT SHOW CORRECT DATA - XML STATEMENT WITH A NULL
IC27038 - ADD MAPPING IN AN OUTPUT MESSAGE COMPUTE NODE IS NOT POSSIBLE
IC27068 - REGISTER A SUBSCRIPTION GIVES THE EVENT ERROR 7053 -UNEXPECTED
IC27301 - SETTING MQMD.EXPIRY FIELD TO AN ABSOLUTE VALUE IN COMPUTE NODE
IC27350 - RFH2 THAT IS ADDED AFTER THE BODY OF THE MESSAGE SHOULD NOT BE
IC27394 - INCOMPATIBILITY BETWEEN MQSI-TIMESTAMP-DATATYPE AND THE FORMAT
IC27806 - DATA FLOW ENGINE WILL NOT RECOVER FROM ERRORS DURING DELETION
IC27959 - DELETING A DATABASE REFERENCE FROM A COMPUTE NODE DOES NOT STO
IC28174 - COMPOUND NODE OF A FLOW CONTAINING LABELS CANNOT BE INCLUDED
IC28312 - FLOAT DATA TYPES OF LENGTH 8 ARE NOT HANDLED CORRECTLY WHEN SE
IC28324 - COBOL COPYBOOK IMPORT PROBLEM, MESSAGE BIP1812S RECEIVED.
IC28401 - BIP2211E ERROR OCCURRED BY USING BYUSERID IN ORDER MODE OF INP
IC28443 - BIP1812S RECEIVED USING THE CONTROL CENTER GENERATE FUNCTION I
IC28446 - INSTALL GUIDE DOES NOT DISCUSS ABOUT MQSI_PARAMETERS_FILE VARI
IC28447 - INSTALL GUIDE PAGE 129/130 HAVE INCORRECT NAMES FOR NT CONSIST
IC28452 - RULES NODE - SUBSCRIPTION REFORMAT OPERATION IS IGNORING THE IN
IC28518 - 1. CREATING A BITSTREAM THAT CONTAINS A REPEATING SEGMENT OF
IC28520 - LOOP ON COMPUTE NODE - REPEATING ELEMENT USES A REPEAT COUNT
IC28605 - MESSAGE FLOW LOOP WITH MSGBIP2608I MSGBIP2623I AND MSGBIP2611I
IC28607 - HEX FF VALUES NOT VALID FOR STRING TYPES
IC28724 - IMPORT OF COBOL COPY BOOK THAT DOES NOT HAVE A COMPLETELY ASCE
IC28810 - SUBSCRIPTIONS WON T SHOW UP FOR A SINGLE USERID
IC28926 - SIGNED PACKED DECIMALS WILL ACCEPT F AS WELL AS C AND D
IC29022 - BIP1812E CAUSED DURING A DEPLOY IN STRUCTURES WITH MANY EBMEDD
IC29033 - MQSI V201 USING DUPLICATE FLOW FEATURE TO CREATE MESSAGE FLOW
IC29180 - THE INTERVAL DATETIME TYPE DOESN T WORK AS DOCUMENTED
IC29184 - XML-PARSE ENVIRONMENT INCORRECT
IC29200 - XML MESSAGE GETTING TRUNCATED AFTER PROCESSING THROUGH COMPUTE
IC29210 - MQSIMRMCOPYMSGSET CHANGES TWO ATTRIBUTES OF THE COPY IT MAKES
IC29235 - SPURIOUS ESQL SYNTAX ERROR IN COMPUTE NODE WHEN SETTING A FIELD
IC29371 - MQSI DOES NOT PROPERLY CONVERT HEX VALUE FOR UNICODE CODE-POINT
IC29387 - MESSAGE TRANSLATION INTERFACE SEEMS TO HAVE A PROBLEM WITH THE
IC29416 - COMPUTE NODE ADD MAPPING TO MRM LAYOUT GENERATES BAD ESQL
IC29489 - DBCS MESSAGES WERE NOT CONVERT CORRECTLY FROM PC CODEPAGE TO H
IC29506 - MQSI SUPERUSER IBMMQSI2 CAN CHECKOUT RESOURCES, BUT GETS AN ERR
IC29632 - MQSI - INCORRECT STRUCLENGTH AND NAMEVALUELENGTH IN RFH2 LOOP
IC29775 - MQSI BROKER DO NOT PASS THE MQMD CONTEXT INFORMATION WHEN A
IC29789 - IT APPEARS THAT A CDATA SECTION IS MARKED AS "SPECIAL" IN THEXM
IC29801 - MQSI V2 FAILS TO RECONNECT AFTER ORACLE HAS BEEN BOUNCED
IC29992 - IF DEFAULTHEX IN THE DB TABLE HAS 1000 BYTES OR MORE THEN SPACE
IC30171 - REORDER OF ELEMENTS FOR A MESSAGE SET, WITHIN A COMPOUND TYPE
IC30173 - MQINPUT CATCH TERMINAL DOES NOT COMMIT DATABASE LOCKS.
IC30195 - EG ABEND WHEN USING TWO MESSAGES OF THE SAME STRUCTURE WITHIN
IC30424 - AFTER APPLIED CSD2, THE MESSAGE BROKER INTERNAL PROBLEM ERROR O
IC30497 - ERRORS OCCURRING DURING MQSI GUI CONNECTION TO ORACLE 8.1.6 NE
IC30653 - WHEN YOU UNSET A PROMOTED PROPERTY, DEFAULT IS LOST
IC30673 - SQL SYNTAX CHECKER REPORTS ERROR ON VALID SYNTAX WHICH DEPLOYS
IC31150 - GETTING AN ERROR BIP1835E. THE CREATE RUNTIME MESSAGE DIC.
IC31345 - COBOL IMPORT PROBLEM WHEN LEVEL NUMBERS FIELDS NOT ASCENDING
IC31362 - MESSAGE GROUPING FUNCTION ONLY PROCEED THE FIRST MESSAGE
IC31445 - DEFINING A STRING MRM-MESSAGE ELEMENT AS NULL-TERMINATED WITHIN
IC31787 - MQSICHANGEUSERNAMESERVER ACCEPTS INVALID USERID, AND CHANGES
IC31806 - PROBLEM IMPORTING COBOL COPYBOOK. IMPORTER DOESN T LIKE EXTRA
IC31825 - BOTHRESH AND BOQNAME NOT WORKING PROPERLY WHEN ERROR OCCURS ON
IY12651 - DATA FLOW ENGINE WILL NOT RECOVER FROM ERRORS DURING DELETION
IY13621 - MQSI MRM MESSAGE IS CAUSING THE BROKER TO SPIN
IY14540 - AIX BROKER CONFIGURATION LIBPATH MODIFICATION MISSING IN DOCUM
IY14660 - MEMORY CORRUPTION ON AIX USING DB2 VIA TCP/IP WITH NIS ENABLED
IY14720 - MSGBIP2301E WHEN A BROKER TRIES TO LOAD THE MQSI SAMPLE PLUGIN
IY14970 - DOCUMENTATION INCORRECT/NEEDS CHANGE ON COMPILERS SUPPORTED BY
IY15310 - MEMORY LEAK AND THROUGHPUT DEGREDATION
IY16379 - EXECUTION GROUP TERMINATES ON STARTUP WITH BIP2123
IY16620 - MQSI CORE DUMP CAUSED BY ADDRESS ALIGHNMENT ISSUE WHEN BINDING
IY16679 - INADEQUATE ERROR MESSAGE PRODUCED FOR EXECUTION GROUP FAILURE
IY16970 - DFE STOPS PROCESSING MESSAGES FOR A FLOW. INCOMING MESSAGES BA
IY17486 - COBOL STRUCTURES, CONTAINING SIGNED DECIMAL NUMBERS (PIC S9),
IY17561 - RULES NODE NOT PROCESSING A MQS EXPIRY NO PROPAGATE FIELD IN SU
IY17772 - AFTER UPGRADE MQSI VERSION F/V1.1 T/V2.0.1, MQSI BROKER HANG
IY17780 - DFE NOT STARTING PROCESSING OF MESSAGES ON THE INPUT QUEUES
IY17806 - WHEN BROKER HAS LONG CONNECTION TO DATABASE, SOMETIMES THE CONN
IY17808 - XML PARSER ADD TRUE TO THE END OF XML DECLARATION.
IY18104 - CAST FUNCTION INTERMITTENTLY FAILING
IY18146 - IF THE BROKERDB IS ORACLE, ADDITIONAL NLSPATH IS REQUIRED.
IY18245 - PERFORMANCE FIX TO ADDRESS INEFFICENCY IN BIT STREAM FOR
IY18458 - ERROR PUT TO DLQ BY THROW NODE IS UNREADABLE - 4048 ERROR AND
IY19242 - LOCAL_TIMEZONE ESQL IS RETURN 239/241 INSTEAD OF 240 ABOUT 50%
IY19382 - DATAFLOWENGINE LOOP
IY19787 - BIP1205E DATABASE EXCEPTION SQL0302N ON DEPLOY
IY20050 - ERROR WHEN OUTPUT TERMINAL THAT IS LABELLED WITH A NAME THAT
IY21179 - DFE COSUMING HIGH CPU AND HANGING WITH NO ERRORS
IY22261 - BROKER IN AIX SMP ENVIRONMENT MEMORY MANAGEMENT VULNERABLE TO F
IY23087 - EXCESSIVE CASHING PERFORMED BY MQSI CAN LEAD TO MEMORY HEAP
The WebSphere MQ support page is located at:
http://www.ibm.com/software/ts/mqseries/support
from which you can obtain the latest information about various
WebSphere MQ topics including :-
- Hints and Tips
- APARs and Fix Packages
- FAQs
- Support downloads
- Newsgroups
For the latest versions of WebSphere MQ Integrator Version 2.1 for
Windows NT and Windows 2000 documentation see the following Web page
on the MQSeries Web site:
http://www.ibm.com/software/ts/mqseries/library/manualsa/
The README file shipped with WebSphere MQ Integrator Version 2.1
contains information that was current at that time. The very latest
information is available at:-
http://www.ibm.com/software/ts/mqseries/support/readme/index.html
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
CSD HISTORY
===========
CSD05 - U487641
_______________
This is CSD05 for WebSphere MQ Integrator Version 2.1 for Sun
Solaris. It is delivered via PTF U487641 for APAR IY41518.
It contains fixes for ALL the problems listed below:
Name - Abstract
IY41518 - CUMUL MAINT #05 FOR WEBSPHEREMQ INT V2.1 FOR SUN
IC32907 - SELECT FROM A DATABASE IS NOT RETURNING THE MICROSECON
IC32971 - BIP2430 - ESQL CONCATENATION WITH NON EXISTANT FIELD
IC33620 - RULES EXPRESSION CHANGES WHEN IT IS APPLIED
IC34564 - 'USED IN' DOES NOT SHOW CORRECTLY FOR THE LITERAL THAT
IC34708 - RCD NODE CAUSES MRM MSGTYPE PREFIX DOUBLE PRE-PEND.
IC34946 - DOCUMENTATION DOES NOT INCLUDE ALL SUPPORTED CODEPAGES
IC34947 - NNSY NODES/PARSER NOT PICKING UP THE CCSID OF THE MESS
IC35006 - USE OF REPLYTO FIELDS IN THE DESTINATIONDATA FOLDER.
IC35011 - WMQI V2.1 - C/T REQUEST TO IMPROVE DOCUMENTATION REGAR
IC35013 - ERROR VIEWING THE PROPERTIES OF A RULE CONTAINING FORM
IC35052 - ADDRESS TREE COPYING FROM USER PARSER CREATED MESSAGE
IC35084 - INPUT CONTROL SET AS REGULAR EXPRESSION + DELIMITER CA
IC35097 - COBOL INPORT : DEFAULT VALUES NOT CREATED FOR DECIMAL
IC35139 - REPEAT KEYWORD IN MOVE INCORRECTLY REJECTED IN CONTROL
IC35298 - REPEATING RECORDS ARE SEEN AS EXTRANEOUS IF THE LAST F
IC35342 - CODE PROBLEM WITH MULTIBYTE CHARACTER SETS
IC35370 - DECIMAL POINT IS NOT SUBSTITUTED IN MRM COBOL LANGUAGE
IC35397 - ADD MSGFLOW VERY SLOW - IMPROVE PERFORMANCE
IC35410 - MQSI THROWS NEWERAON FORMAT EXCEPTION AS SETERROR ALWA
IC35479 - QUEUE MANAGER HANGS AFTER INITIATING USER LOGS OFF
IC35512 - UPDATE TO DOMAIN SECURITY AND AWARENESS DOCUMENTATION.
IC35549 - CONTROL CENTER HANGING - DUE TO THREADING PROBLEMS.
IC35575 - RENAME MSGFLOW FOR DEVELOPMENT & PRESERVE ORIGINAL COP
IC35605 - INCORRECT OUTPUT DATA LENGTH WHEN THERE IS NO VALID DA
IC35664 - ONLINE HELP FILES INTEGRATED INTO CSD03 AND CSD04 HAVE
IC35769 - DISCEPANICES AROUNDTHE COMPOUND STATEMENT AND CASE FUN
IC35806 - BIP9398S,9351S ON IMPORTING THE MESSAGES IN THE SAME M
IC35898 - 'TDS Group Indicator written for omitted compound elem
IC35939 - DEADLOCKS WHEN USING THE AGGREGATE REPLY NODE.
IC35942 - MESSAGES FAIL WHILE UPDATING THE TABLE BMQESTDMSGOUT U
IC35949 - XML SCHEMA IMPORT FAILS, WHEN 'SIMILAR' ENUMERATION AR
IC36010 - DOCUMENT USE OF CLEARMESSAGE FOR JAVA PLUGIN NODES
IC36384 - IMPLEMENTATION OF THE NNSY REPEATING ATTRIBUTE TO FLAG
IY24156 - INST_DB NOT RETURNING THE UNDERLYING ORACLE ERROR CODE
IY25775 - PROBLEM WITH BIGINT DATATYPE ON AIX
IY27218 - ERROR WHEN SELECTING DATETIME WITH MILLISECONDS ON MSS
IY34362 - ORACLE DB CONNECTION IS NOT RE-ESTABLISHED IF USER DB
IY35755 - REFORMAT FAILS TO FIND INPUT DATA - NEONTRANSFORM ENGI
IY36059 - ESQL SYNTAX ERROR CAUSING A EG TO ABEND
IY36091 - DOCUMENTATION PROBLEM WITH HOW MESSAGES ARE PROCESSED
IY37213 - FOLDERBITSTREAM CONSTANT INCORRECT ON AIX PLATFORM
IY37803 - BROKER FAILS TO CONNECT TO ITS DATABASE, AFTER A DATAB
IY37824 - WMQI ADMINISTRATION GUIDEIS MISSING DOCUMENTATION FOR
IY37844 - NNSYMESSAGELOG MUST BE ACCESSIBLE BY THE BROKER AND IN
IY37997 - 'MQCMITS ISSUED WHEN NO MSG AVAILABLE'.
IY38039 - MQSICOPYMSGSET DOES NOT PRESERVE UNSET VALUES
IY38589 - BIP2228E ON ORACLE DB IPL
IY38673 - BIP2135E DATA CONVERSION ERROR DURING CONVERT FROM SJI
IY38713 - THE MQMD FORMAT IS BLANK, PROCESSING A MESSAGE USING A
IY38799 - RULESET AND ADDED AGAIN AS A FORMATTER MESSAGE TYPE.
IY38886 - ZERO LENGTH MSGS ON OUTPUT Q WHEN LOGLVL=3
IY38891 - JAVAW.EXE EXCEPTION (CSD04) ESQL SYNTAX CHECKER
IY38894 - MIGRATION FROM V201 TO V202 USING ORACLE 815
IY38935 - IN RULES EXPORT FILES AN EXTRA SPACE AFTER AN EXIST OP
IY39051 - ALTERNATIVE FORMAT DOES NOT FIND IT'S INPUT DATA
IY39777 - CSD04 CAUSING DB2 STORED PROCEDURES TO FAIL.
IY39952 - ERROR BIP5341 OCCURS WHEN COPYING MRM INT FIELD TO STR
IY40097 - CASE FUNCTION CANNOT RETURN LIST/ROWS - BIP2556
IY40186 - SEMOP FAILURE IN IMBNAMEDMUTEX CAUSES DATAFLOWENGINE A
IY40730 - 'MQINPUT NODE: MESSAGE TYPE SHOULD SHOW IDENTIFIERS NO
IY40842 - MQSI NOT USING 'C' AS A CODE FOR SIGNED EBCDIC EXTENDE
IY41088 - 'END OF BITSTREAM' DOESN'T WORK FOR BINARY ELEMENTS.
IY41996 - ACCEPT ZERO LENGTH STRINGS FOR MRM NUMERIC FIELDS
IY42133 - TEXT FOR ERROR BIP5199 NEEDS TO BE CORRECTED.
PQ66950 - MEMORY LEAK WHEN USING THE NEON FEATURE OF WMQI.
PQ67615 - CNIELEMENTCHARACTERVALUE/CPIELEMENTCHARACTERVALUE CAUS
PQ70700 - ABEND0C4 OR DICTIONARY LOAD ERROR INTERMITTENTLY WHILE
PQ71204 - UPDATE TO THE PROGRAMMING GUIDE ON HANDLING ERRORS RET
PQ71205 - BIP5502E:MRM INT/FLOAT->EBCDIC INC. SIGNED EXTDEC WITH
PQ72256 - BUILDING A C++ USER EXIT LIBRARY, CUSTOMER MUST USE TH
24125 - MRM: Msg 5424 user response could be more helpful.
27730 - HOT:ACORD:VAL:BIP5374 output ACORD message
28447 - Resource Manager cannot handle WMQI 5.0 dictionaries
28806 - v2.1 csd 03 breaks IH05/IC02 SupportPacs
28835 - Dublin - Incorrect parsing
29230 - XML Schema Importer does not handle UTF-16 + See comment
29382 - logErrorData macro causes java core dump in CMI on AIX p
29648 - When trying to run the sample program scribble on window
29862 - Merant driver updates
30107 - Problem importing a DTD
30150 - Invalid ESQL syntax checker scope error when using proce
30263 - readme update for ACORD
30709 - Poor perf. expanding large MRM structures in Compute nod
CSD04 - U485556
_______________
This is CSD04 for WebSphere MQ Integrator Version 2.1 for Sun
Solaris. It is delivered via PTF U485556 for APAR IY36048.
It contains fixes for ALL the problems listed below:
Name - Abstract
IY36048 - CUMULATIVE MAINT #04 FOR WEBSPHEREMQ INTEGRATOR V2.1 FOR SUN
IC33207 - SILENT INSTALL FOR CSD FAILS
IC33620 - RULES EXPRESSION CHANGES WHEN IT IS APPLIE
IC33844 - JUMPTORULES FUNCTION NOT WORKING THROUGH FORMATTER GUI
IC34178 - USING DATE FORMAT FROM XML TO MRM, THERE IS A DECREASE OF ONE
IC34394 - MQOutput node abends when writing aggregated reply message
IC34449 - ZOOM FACTOR FAILS WHEN VIEWING A MESSAGE FLOW USING 'FIT TO
IC34463 - 2068 ERROR WHEN A FAILED MESSAGE IS BACKED OUT TO A CLUSTERE
IC34470 - BIP1813W OCCURS WHEN TRYING TO ADD A PHYSICAL LAYER ON A MESSAG
IC34489 - USE OF INVALID CODEPAGE IN MRM CAUSES BROKER TO ABEN
IC34526 - PROBLEMS IMPORTING XML SCHEMA FILE USING THE NEW PROVISION
IC34563 - BIP5288E ERROR AT CSD03 LEVEL
IC34594 - UNABLE TO RENAME A MESSAGE FLOW ON THE LEFT PANEL OF
IC34602 - A BROKER DATABASE CONNECTION *LEAK* UPON MQISDP CLIENT
IC34651 - PROPAGATED SUBSCRIPTION RECV'D ON SYSTEM.BROKER.CONTROL.QUEUE
IC34656 - AFTER CSD03, COPY AND PASTE ON A MESSAGE FLOW DOESN'T WORK
IC35035 - XML SCHEMA IMPORTER IS MISSING AT U200179 LEVEL OF WEBSPHERE
IY29380 - BIP8146E MESSAGE IS NOT PICKING UP THE CORRECT INSERT FOR REGIS
IY30276 - Problem selecting > 1998 bytes from Oracle CLOB column
IY31377 - IMPORT NEW ERA FORMATS FROM OLDER VERSION INTO V2.1
IY32488 - EXCEPTION LIST OMITS NEON DETAILS
IY33022 - XA TRANSACTION BETWEEN A SOLARIS BROKER AND DB2 ON OS/390NOT WO
IY33142 - IMPLICIT AND EXPLICIT DATA CONVERSIONS FOR MATH EXPRESSIONS.
IY33716 - AggregateReply node loops following catch terminal
IY33904 - STATUS OF REMOVED MESSAGE SETS IS NOT UPDATED CORRECTLY ON THE
IY34240 - BIP2080 MESSAGE IS LOGGED IN SYSLOG BECAUSE EXECUTION GROUP ( E
IY34332 - THE VDB CAUSES A BIP5009E FOR SOME FIELD NAMES IN NON-XML
IY34561 - DATABASE EXCEPTION WITH AN SQLCODE =-501
IY34804 - AGGREGATE NODES FAIL TO DEPLOY UNDER XA
IY34805 - WMQI 2.1 CSD 03 LATE BREAKING TDS UPDATES AFFECTING SWIFT USERS
IY34949 - COALESCE INCORRECTLY RETURNS A NULL VALUE.
IY34951 - NEGATIVE EXTDEC WITH TRAIL 0 INCORRECT ON SOME EBCDIC CODEPAGES
IY35037 - THE CWF TAB HAVING STRING JUSTIFICATION SET TO RIGHT JUSTIFY AN
IY35159 - PROBLEM DETERMINATION GUIDE ON PAGE 39 STATES THAT HARD LIMIT
IY35724 - TDS PARSER FAILS TO DETECT THE END OF A REPEAT.
IY36341 - NO INFORMATION/INSTRUCTIONS AVAILABLE FOR UNINSTALLING NEW
PQ66348 - THE BITSTREAM FUNCTION HAS BEEN REGRESSED IN CSD03 WHEN
PQ66584 - WMQI DOESN'T FREE UP CACHE FOR PREPARED STATEMENTS.
PQ66773 - ASBITSTREAM IS NOT AVAILABLE IN THE CONTROLCENTER FUNCTION
PQ66774 - ASBITSTREAM FUNCTION IS NOT DOCUMENTED IN THE ESQL GUIDE
PQ66775 - CONTROL CENTER DOES NOT ACCEPT ASBITSTREAM WITH ANY OPTIONS,
PQ66803 - SELECT STATEMENT WITHIN PLUGIN NODE CAUSES SQLSTATE HY010
f20575 - MRM-Select layers to be pasted
f24996 - IO02 message file need uppercase names
f25963 - EUI "RegExp" verifications on "Data Pattern" value member
f26073 - Improve messageset identifier usability
f26106 - Tidy up DataConv error messages
11098 - Only first added broker is shown in collective's properties
15870 - Smartguide name edit controls allow names > 254 characters
16174 - Message pane does not display message clearly
16953 - Message flow view reset while flow is being modifie
17059 - one item of each type displayed as property for a collection
17288 - Exception Delete MsgProcNode from flow use pulldown/Delete key
18981 - Usability suggestion for Create Element Value
19499 - Broker subscriptions cannot be selected using the mouse.
19780 - Promote Property dialog wrong display after window /pane resize
20518 - Plug-in Wizard attribute/group ordering
20585 - Copying and pasting message flows and sets in assignments tab.
20794 - Node with promoted properties should not be delete
20860 - MRM-Inconsistent error messages
21679 - On-line help can't be invoked from Check Message Flow dialog
21905 - Focus does not default to parent if tree object delete
22009 - checkin and unlock very close together.
22222 - Type field not highlighted by tabbing if no selection made
22271 - MQSI2.1:GUI Nodes:Context menu is not getting cleared after Es
22522 - For all TDS string properties, the length > 254 is accepte
22601 - Finish of import dialog should be disable after remove all file
23511 - MRM-can edit frozen message set tab but then lose changes.
23571 - Dbl-clicking brings up properties when properties are disable
24196 - STRESS: repeated SEGV in aggr timeout threa
25183 - Topics: Principal highlight moves when navigating ACLs dropdown
25320 - MRM-User can create messages with base types
25353 - Decimal values are allowed in a numeric fiel
26231 - corruption of ImbStringBase object
26461 - NEON nodes in the export list when exporting from configmgr
26570 - Unable to Ctrl+Tab through node properties
26724 - Subscriptions: Unable to browse subscriptions using U/D arrow k
27442 - MRM:TDWF Tag is not setting default value correctly
27726 - ACORD:VAL:5371 on Parse/validate ACORD embedded compound type
27727 - Check out list dialog displays new objects
27753.1 - Internal error raised by PWFData
27786 - Broker registry key remains on Win XP
27841 - SWIFT:VAL:min and max occurs not validated at compute node
27844 - SWIFT:VAL:Closed type inside opendef makes opendef act close
27846 - ACORD: MRM BIP5506 in ACORD Message if non-?
27853 - ACORD:VAL:MRM Anon field inheriting values from prev elements
27891 - ACORD:MRM BIP5421/BIP5505 parsing repeating ACORD message body
27911 - Select As Function is wrong flagge
27921 - SWIFT:VAL:Missing mand. not validated after selfdef elem on o/p
27929 - SWIFT:VAL: MT562, problem with repeat within repeat validation
27944 - SWIFT:VAL: MT210_0015 repeat seq. causes DFE loop
27955 - SWIFT:VAL: MT300 repeating sequence causes val failure.
27958 - SWIFT:VAL: MT600 failis validat'n seqB missing but seqC present
27971 - ERRMSG BIP5452 has minor typo
27978 - SWIFT:VAL: Embedded Message in MT792 causes validation error
27979 - EUI Check out list causes workspace updates to be lost
27986 - EUI Java exception adding value constraint
27988 - SWIFT:VALID: MT521I messages giving BIP5371
27991 - EUI MRM unexpected hierarchy collapse
28005 - Interface to RgXExp matcher to take end offset
28026 - SWIFT: VAL: MT450: 52B: wrong path in choice being validated.
28037 - SWIFT: 2002: MT501 has 16R with missing : on output.
28047 - SWIFT: VAL: MT562 Some still failing
28048 - SWIFT: VAL: DFE Loop with MT801 message
28056 - SWIFT: 2001: VAL: MT204 57A: failures with validation.
28057 - SWIFT: 2002: VAL: MT103 invalid data pattern in 53B
28059 - German CSD3 Refresh install contains one garbage char
28088 - Correct m_Parent issues
28090 - Sub-field naming for SWIFT field 71F is incorrect
28099 - Failure using EVAL() in filter node on OS/390
28123 - SWIFT: Mset Problem MT500 field MT500^A should be optional
28124 - SWIFT: MT330 Fails to parse
28165 - SWIFT - 2002 failures: MT210, MT306, MT307, MT308
28171 - SWIFT - Modelling problem with 2002 MT321
28172 - Update the preimporter
28180 - The NT/2K CD for CSD03 does not install.
28183 - SWIFT : Missing sequence B1 in MT571 definition
28185 - Allow for MQSeries differences in endmqlsr
28187 - SWIFT: MT700 41D: extra inserted.
28195 - SWIFT: 2002: MT500 missing optional seq B1 not parsed correctly
28198 - SWIFT: 2002: MT514 SeqE shoulkd allow 3 not 2 repeats of 22F
28199 - Embedded messages no longer separated by delimiter of parent
28202 - SWIFT: 2002: MT536 etc. 98A/B/C repeats incorrect
28203 - SWIFT: 2002: MT519 etc repeats of 94C incorrect
28210 - SWIFT: 2002: repeats wrong in MT529 for 70
28213 - SWIFT: 2002: MT584 17B repeats wrong.
28215 - SWIFT: 2002: MT574 data pattern inccorect causes parse fail
28216 - SWIFT: 2002: MT578 seqE 17B should be optional
28217 - SWIFT: 2002: MT308 some msgs give BIP5371 with tags 22H or 19B
28221 - SWIFT: 2002: MT380 test 0004 fails with BIP5377
28228 - During uninstall psapi.dll gets locked by MQ5.3
28229 - SWIFT: 2002: MT513 test 1661 gives BIP5375 on seq F
28238 - SWIFT: 2002: MT500 parsing error
28241 - SWIFT: 2002: MT519 95S/P/U/R repeats wrong
28246 - SWIFT: 2002: MT578 seqD 20C should have 2 repeats
28247 - SWIFT: 2002: MT308 tests fail due to 70C in B4B being mandatory
28276 - Abend of DFE following several thousand SWIFT tests
28332 - Datetime problems
28350 - The MQInput node 'validate' property should not be present
28362 - Mandatory binary elements
28375 - AS400 database: (GEAC) broker loops in fetch with Client Access
28514 - MRM- Empty tags in mcd cause broker aben
28532 - Tag comparisons in parser
28548 - Res. Mgr. caching of dictionaries failing
28612 - Persistent msgs written to Temporary Dynamic Queues
28668 - Plug-in Wizard attribute group drag and drop
28682 - Finish of import dialog should be disabled if no text entere
28704 - Merant Driver Manager update
28706 - Aggregation performance improvements
28755 - Apply track 27891 in CP000_dea
28946 - REGR:Drag and drop into an execution group fails with BIP00004E
29097 - ERRMSG BIP5452 has minor typo
29100 - Unable to drag & drop more than 1 msgflow into an exgrp 28946
29144 - Build error on 390 caused by defect 28514
9177 - Inconsistent adding of brokers between Domain and Topology pan
CSD03 - U483515
________________
This is CSD03 for WebSphere MQ Integrator Version 2.1 for Sun
Solaris. It is delivered via PTF U483515 for APAR IY31453.
It contains fixes for ALL the problems listed below:
Name - Abstract
IY31453 - CUMULATIVE MAINT #03 FOR WEBSPHEREMQ INTEGRATOR V2.1 FOR SOLARI
IC28837 - RENAME A DEPLOYED FLOW, DEPLOY EXEC GROUP SHOWS NEW FLOW NAME I
IC29837 - COMMIT BY GROUPID DOCUMENTATION NOT CLEAR ABOUT ROLLBACK
IC30627 - EXPLANATION OF CARDINALITY FUNCTION IS INCOMPLETE AN
IC30970 - GLOBALLY CO-ORDINATED FLOW INVOKING NEON PARSER CAUSE XA PROBLE
IC31219 - MQSI STACK OVERFLOW BY MESSAGE FLOW THAT LOOPS
IC31421 - IN MQSI 2.0.2 USERIDENTIFIER IN MSG CONTEXT GETS SET THE USERI
IC32774 - PROBLEM WITH COPY MSG HEADERS IN COMPUTE NODE
IC32883 - BIP5341 ERROR IS THROWN WHEN ASSIGNING DATE VALUES TO THE MRM
IC33120 - IMPROVE DOCUMENTATION ON THE NEW BEHAVIOUR OF PARSERS
IC33126 - INCORRECT COPYBOOK IMPORT CAUSES CONFIGMGR TO HANG
IC33163 - MQI V2.1 MRM CWF PARSER BEHAVIOUR NEEDS TO BE CHANGED TO OPERAT
IC33180 - THE MRM PARSER WILL NOT ACCEPT AN MRM FOLDER THAT HAS ANY FIEL
IC33333 - MQE SUBSCRIBER RECEIVING A MESSAGE CONTAINING CERTAIN BINARY CH
IC33382 - TDS PARSER THROWING ERRORS FOR REPEATING FIELDS.
IC33383 - INSTALLING THE CSD02 ON A CONTROL CENTER MACHINE ONLY GIVES
IC33412 - XML TO IDOC CONVERSION RESULTS IN TIMESTAMP PROBLEM
IC33445 - VISUAL DEBUGGER CAN'T HANDLE MESSAGE FLOWS IN WHICH THE MQINPUT
IC33480 - BIP5009 PARSER ERROR - VISUAL DEBUGGER ADDS SPURIOUS END TAG
IC33490 - COBOL IMPORTER INCORRECTLY SETS THE PADDING AND JUSTIFICATION
IC33549 - AIX BROKER CSD02 UNABLE TO PARSE TAD/DELIMITED MESSAGE - TREATI
IC33588 - EXTENSIONS TO MQSIIMPEXPMSGSET TO HANDLE ERRORNEOUS COBOL
IC33610 - BIP0068E CONTROL CENTER STARTUP PROBLEM
IC33627 - CCI_EXCEPTION PLUGIN SAMPLE TO BE UPDATED TO INCLUDED AN EXAMPL
IC33693 - BIP1812S WHEN IMPORTING COBOL COPYBOOK ON MQ V5.2
IC33699 - DEFAULT VALUE FOR THE DATE IS NOT BEING UNDERSTOOD BY ONE OF TH
IC33728 - MRXCOBOL.DLL OMITTED ON CC ONLY INSTALLATION.
IC33871 - WMQI TRIMMING FIXED LENGTH ELEMENTS IS CAUSING PROBLEMS.
IC33959 - MQSIREADLOG/MQSIFORMATLOG GIVES BIP8029E ERROR WHEN USERI
IC33962 - MRXYYYY.DLL'S BEING UNLOADED AS MSG SETS ARE BEING WORKED ON
IC34001 - NEW ERA DOES NOT WORK ON HP PA-RISC1.1 MACHINE
IC34041 - CONTINOUS PUBLICATION NODE RESULTING IN EG FAILURE WITH BOOTSTR
IC34060 - PROBLEM WITH ESQL EDITOR IN THE COMPUTE NODE WHEN ENTERING WHEN
IC34115 - COBOL SIGN KEYWORDS NOT USED ON DECIMALS DURING IMPORT
IC34191 - COBOL COPYBOOK IMPORTER SETS MINOCCURS INCORRECTLY
IC34206 - SCADA SUBSCRIPTION FAILS ON 2ND BROKER IN A COLLECTIVE
IC34246 - BIP2461E WHEN INTERNALLY CASTING CHAR TO DECIMAL
IC34289 - WHEN AN EXECUTION GROUP IS RENAMED - IT REVERTS TO THE OLD NAME
IC34391 - DB2 INSTANCE PARAMETERS MAXAGENTS AND MAXAPPLS NEED TO BE SET
IY21834 - case#1274377 CHAR CONVERSION NOT PERFORMED ON CLOBS PASSE
IY26056 - ADDING MQCIH HEADER TO THE MESSAGE AND ENCODING AND CCSID VALUE
IY27441 - MQSI V202 BROKER CAN T RESET BIP2135E STATUS ONCE THE MESSAGE
IY28005 - LOOP ON MSG BIP2609E BIP2608E BIP2610E IF THE MQINPUT NODE IS A
IY28995 - BIPSERVICE TERMINATING ON AIX IN AN ENVIRONMENT WITH LOT OF MQS
IY29128 - BROKER 'CONFIGURATION MESSAGES' ARE SENT TO THE MQSERIES DEA
IY29419 - BROKER EXITING EARLY BEFORE CONFIGURATION MESSAGE ARRIVE FROM E
IY29461 - ADD TO DOCUMENTATION THAT IT S NECESSARY TO HAVE A MATCHINGCALL
IY29896 - REPEATING FIELDS, VIA REPEATING ATTRIBUTE, DO NOT WORK IN THE
IY30451 - BROKER ABENDS WHEN PARSING AN XML MESSAGE AGAINST A MESSAGE
IY30965 - WMQI V2.1 PARSER LOOPS IF NO MESSAGE BODY TO AN XML DOCUMENT
IY32778 - MQSILIST COMMAND SHOWS NO RESULT OR ERROR
IY32888 - PROBLEMS WITH DEPLOYING THE EXECUTION GROUPS THAT ARE IN INVALI
IY33138 - INPUT MQRFH2 MCD FOLDER DETAILS INCORRECTLY USED IN RCD NODE
IY33268 - EXECUTION GROUP ABENDS AFTER SELECTING THE CONTINUE WHEN A FLOW
IY33912 - MESSAGES GO TO DEAD LETTER QUEUE WITH DLH REASON CODE 65536
PQ56571 - MQSICREATEBROKER ON Z/OS FAILS WITH INCORRECT INSTALL PATH
PQ57687 - SETTING FOR ICU_DATA IS ERRONEOUS
PQ59154 - RUNNING MQSICUSTOMISE CAUSES DATA TO BE APPENDED TO BOTTOM OF
PQ60310 - 0C1 ABEND WHEN STARTING WMQI DATAFLOW ENGINE
PQ60976 - LARGE MESSAGES IN MESSAGE FLOW CAUSE DEBUGGER TO LOSE
PQ61201 - 'LOCAL ENVIRONMENT' IS BEING CALLED 'DESTINATION LIST'BY DEBUGG
PQ62125 - NEED DOCUMENTED GUIDELINES ON LOGGING 1GB LOB COLUMNS.
PQ65755 - WMQI SHOULD EITHER PASS THE USERID AND THE PASSWORD ON THE
f13552 - Improved Flow of Control in SQL
f13638 - Improve ConfigMgr Role Management
f14155.1- MRM dictionary validation
f14313 - Relocatable runtime capability needed on UNIX
f18003 - Unable to insert BLOBs/CLOBs of >4k in oracle
f19256.4.1 - Implement indexes on non-390 platforms
f19256.4.3 - Make broker table lock switchable
f19604.2- Performance enhancements: DataFlowEngine
f20573 - Exit to flow NT domain info from Control Center
f23727.1- Reduce AdminAgents memory requirements on deploy
f23841 - Enhanced BITSTREAM function
f24881 - Improve scalability of large message processing
f24911 - Upgrade MQSeries Everyplace to latest level
f25127.2- Enhanced THROW statement
f25879 - Optionally disable comment in MRM-XML
f25959 - Allow amount fields to contain a trailing comma
f25963 - Parsing using regular expressions
f25967 - mqsiimpexpmsgset defaults reset
f26060 - ModelMessages.properties correction
f26066 - Datetime defaults inconsistent between CWF and TDS/XML
f26081 - GenXML: Enable quotes in Processing Instructions
f26087 - General feature for the SWIFT messageset work
f26106 - Error Message Architecture review for MRM
f26124.1- Generate runtime objects for SQL
f26172.1- Minimise UnicodeString use in dataconv
f26172.2- Remove UnicodeString from bitstreamworkerstatechecker
f26309 - A set of User/Group Plug-ins to improve Performance
f26336 - Optimized (small) jar for MQe clients
f26480 - CSD03: ship new WMQI license text
f26613 - Support for E format floats in TDS
f27040 - Switch on XA fix
f27332 - Customers want to access the database from inside SQL routines
f7631.3 - Allow zero or more input terminals in the plugin wizard
10775 - BIP2212: Error message does not say which node caused the error
11522 - mqsicreate... qmgr exists but no DLQ
11721 - C & P. Cross-msg-set warning is insufficiently selective
13040.4 - Allow format field of header to be set for IDO
14996 - Change default setting in export preferences
15614 - Errors written to event log when catch terminal is connected.
15624 - MQInput node catch terminal changes the exception list
15869 - Smartguides : Element no removed from list if added to Summary
16522 - Inconsistent behaviour if renaming broker & checked in Topology
17516.3 - CWF Padding char different on 2.1 from 202
17877 - MRM-Properties Pane History and Description Tab entries lost
18047 - MRM-BIP0074E 1. presentation inconsistent 2. Inappropriate wor
18229 - Database Node message now appearing on Failure Terminal
19266 - MRM msg BIP0707E text
19501 - Broker subscriptions cannot be deleted using the delete key.
19508 - Default domain list for MQInput, MQeInput etc needs JMSdomains
19557 - Help window doesn't close when Argo is shut down
19575 - Help "Tour" window close when Argo is shut down
20175 - Need to characterise behaviour of XA with 2 DSNs resolving to a
20256 - spaces in physical format name causes problems
20501 - cut and paste of elements not working for XML layer in CC.
20583 - String "PublicGroupRef" not translated in Topic View
20676 - DFE startup trace cannot be obtained.
20715 - Cannot access 'Refresh' using shortcut key
20999 - Testing TDS properties values unitary, MRM refuse EUI right val
21369 - MRM: When changing properties of a type within a sequence type,
21446 - Command-line usage is inconsistent with other products
21459 - loop back connections hidden if nodes are rotate
21565 - List option gives wrong error message
21640 - CC Syntax Checker rejects UUIDASxxxx functions on RHS of !! fn.
21715 - CAST decimal to decimal fails
21770 - Consistency between Syntax Check of BITSTREAM and FIELDNAME fns
21826 - tdsextractor: Incorrect text in warning message - PT_WARNING
21989.1 - Validation of CWF pad character
22077 - Improve ImbLanguage tracing
22367 - Tidy up ImbConsoleParser code
22506 - Debug node shown in Add Message Flow list only on Assignments
22514 - 'Undo' does not fully work on Assignments page
22594 - Remove Delete of element under elementQualifier should disable
22607 - Missing mnemonics Value Constraint/Language Binding/Physical Fo
22612 - Create an element with an existing name should be forbidden
22672 - Messages 9141 & 9142 should be informational
22854 - Wrong logo for warning message in Topics properties
22930 - ERRMSG: BIP5511: inserts wrong, raw conversion error code
22980 - ERRMSG: MRM BIP5180 message text
22983 - ERRMSG: MRM BIP5505 message content
23060.1 - Remove use of MQSI_DONT_XA_UNENLIST
23130 - TDS floats don't use non-standard negative signs on output
23436 - BIP9177I is misleading.
23456 - MRM-Check-out confusion follows Finalize Failure
23514 - Progresss bar in Visual Debugger title contains MQSI reference
23550 - FIELDVALUE function badly neede
23623 - ERRMSG: MRM BIP5199 message content
23729 - Length of element.Tag value should equal parent type.Tag Length
23761 - dump of BAGGRGEATE table does not correctly show NULL data
23764 - insert missing from message - so produced wrong value
23794 - The message BIP8137 is uninformative
23931 - Copy message flow after compute node properties
23988 - 2.1->2.02: Relax broker to allow bytes for Ext Dec Length Units
24007 - Message in rstclog written incorrectly.
24037 - Create Broker uses -k flag with diff meaning to 2.1
24090 - Semaphore can be left locked if exceptions are thrown
24116 - RFH2 Parser is very slow
24149 - Message BIP9172E should be BIP9172I.
24232 - MRM-4 MRMBxx22 testcases are failing
24287 - BIP2163 text missing apostrophe
24344 - MRM Generate messagebook xtrcterr.log content
24414 - Import MFs - Wrong displayed BIP0662W when there's MF category
24433 - Debugger - No mnemonic for the Settings in Actions menu
24434 - Debugger - Mnemonic "O" is used twice in Actions menu
24464 - Plugin Node - After drag and drop , attribute duplication
24499 - OS390: 2060 should be sent to the console by default.
24575 - Inconsistent behaviour when more than one choice in message.
24589 - Repeating compound type >max occurs not output correctly in CWF
24694 - Type-o in scripts/...template/dsnaoini SUBSYSTEME
24733 - Aggregated response contains data from another group
24805 - MtiImbSyntaxElement::copyElement(ImbSyntaxElement* source)
24861 - MRM Skip count does not work for compound elements
24997 - Messages BIP2206I, BIP2282W, and BIP2320W are obsolete.
24998 - Message BIP3902I is obsolete.
25045 - Check for correct ODBC driver level
25072 - Aggregation node gives DB2 error
25178 - TDS doesn't output default values for missing fixed elements
25369 - Multipart: Message embedded in choice failing for AED/VL/FL
25427 - TDS Fixed length output incorrectly for repeating choice.
25431 - Using DBCS name does not seemed to work
25456 - ERRMSG: Wording of BIP5429E needs amending
25527 - Bad error message when saving log
25559 - MRM BIP5168 message content
25573 - -l option iseen n help text for configmgr commands
25590 - Invalid field values in MQM
25600 - RecoverableException ?????????? 0 ??????????
25652 - MQeMbMsgObject's priority field set incorrectly
25672 - MQMD field values differ between MQe & MQSI
25673 - Exception on setting value 9996 into Feddback field for Mb msg
25735 - setUniqueId declared as public in MQeMbMsgObject
25746 - MRM- Null Handling CWF Integer Extend Dec Padded Literal Null
25748 - ERRMSG: CWF General Error BIP5167 insertions are muddled up
25754 - GUI allows TDS esc of more than one char
25758 - Data conversion to code page 932 (double byte) and back to 437
25824 - MRM TDS and multipart message keys
25830 - Select As Function is wrong flagge
25835 - Call to execute a Stored Procedures fails on z/os.
25842 - MRM- Null Handling TDS Fix Length Float NullLog / Pad confusion
25857.1 - XML Schema Extensions and Restrictions
25899 - Changes with 25636 coming up in English on NLV
25903 - Not seeing Control Center log if security exits set MCAUserI
25914 - MRM:No hourglass cursor while expanding large message subtree
25925 - EVAL cannot execute a CALL statement
25934 - CWF:Multiple levels of unresolved messages/choice not supporte
25939 - Add message popupmenu item only available on Compoun
25944 - Severe Abend when NEON parsing is used with MQeInputNode
25978 - TDS Validator to check for unbounded repeats
25986 - PrivateRegistry causes MQeInputNode failure
25993 - Deploy failures cause a cascade of failures
26002 - Unable to add panes for nested multipart msgs in compute node
26003 - In the EUI, Attributes cannot be specified on the XML Root Tag
26014 - BIP5355 messgae content
26016 - STRESS: BIP8303 produced by broker during cache update
26019 - Syslog does not report failure to load lils
26022 - Insert missing in BIP9170E when compcif not found.
26024 - Aggregation stress test abends after 20 mins
26042 - MRM- Broker Abend due to 'StrToDatetime' metho
26046 - MRM TDS Message Format with NULLs not Parsing Correctly on 390
26054 - MRM-Trace node stops tracing tree at an unresolved choice.
26075 - MRM-CWF IEEE floats not as accurate on 390 as they should be
26076 - Stack overflow in TDS parser
26084 - Certain MRM tests fail on AIX and Solaris.
26086.1 - Covering TDS Parser side of 26086
26091 - Incorrect length of ISA
26096 - Regression breaking plugin input nodes with MRM messages
26097 - Stopped flow not recreated on broker stop/restart
26109 - Broker abends when parsing an XML message
26127 - broker crashes if java node accesses GlobalEnvironment
26130 - MQe/SCADA Participant tables are not protected by a semaphore
26131 - In testing for repeating tags look for group in
26133 - MRM CWF setting to null unexpected short message
26134 - mqsimrmimpexp - mispelling in help instructions
26137 - MRM TDS Data lost on output
26138 - BIP2011 message content
26140 - NT handling of 390 floats has regresse
26155 - ERRMSG: BIP5004 full of meaningless numbers and terms
26160 - Multiple topics import incorrectly
26181.2 - CC not prevent user to repeat element in simple unordered type
26182 - Poor diagnosis of missing length count
26185 - Pubsub expiry tests failing since clock change to BST
26189 - DTD Importer-Imports mixed content model as ordered set
26190 - Abend of DFE when a message set is deploye
26193 - Use of 16S as group terminator failing in TDS
26198 - MRM- XML Schema Importer incorrect treatment of "fixed"
26203 - MRM- XML Schema Importer namespace errors
26225 - MRM- XML Schema Importer NOTATION datatype not recognise
26226 - MRM- XML Schema Importer Name datatype not recognise
26227 - looks like a syntax error in ImbPointer.hpp
26237.1 - CMI: (stddi) BIP5323, BIP5323 defined in wrong place.
26237.2 - CMI: (bsutils) New PWFCh & PWFString implementation breaks CMI
26237.3 - CMI: (stddi) StdDictionaryIterator uses ImbOwnedLogSource
26237.4 - CMI: (cwf) CWF worker specific CMI resync changes
26239 - MRM- XML Schema Importer "use" clause on attribute not importe
26247 - Sample DB2 switch load file must be rebuilt for MQ 5.3
26248 - MRM- XML Schema Importer wrong treatment of recursive import
26249 - GenericPropertyParser class name should bt upper case
26262 - Error in Neon
26272 - Parsing of empty 16R 16S sequence fails
26307 - DFE fails to start with 0c1, if region size too small
26317 - MRM- XML Schema Importer MaxExclusive for STRING wrong
26321 - Change abend code from 2C3 to 2C1
26322 - MRM- 5354E MTI Internal Error with TDS layer
26324 - CWF parameter 'First Week of Year' in message sets on C
26337 - The lil imbdfneo.lil can not always
26339 - MRM- XML Schema Importer "nil" namespace prefix
26343 - MRM- XML Schema Importer using Finalized Message Set
26380 - MRM- TDS Writer Internal Error BIP5432
26384 - Performance inline
26398 - e-fix2961:Broker abend with TDS containing embedded message
26425 - TDS with VDP set and explicit decimal gives incorrect output
26431 - Parsing of LX loops in X12 not correct
26439 - mqsi_setupdatabase creates link for wrong libraries
26447 - Glossary generates the German version.
26454 - HOT: MRM-Certain chars in doctype text cause serious CC errors
26460 - Import Topics enabled and processed when Topics not checked-out
26471 - Multipart messages with absolute paths cause abends
26475 - MRM- Broker Abend when writing to TDS with Base & Simple Types
26491 - MRM DTD Generator and Choice elements
26499 - mqsideploy -r doesn't work in Japanese environment (CCSID=932)
26502 - No retry on BSUBSCRIPTIONS table lock.
26503 - Hang in JVM during DFE termination.
26536.1 - Disable Member Rendering for Compound Elements
26572 - I enter a direct setting to a float in the ESQL of a compute no
26594 - Syntax checker flags CREATE ... AS X as error
26616 - Registry setting needed for trace file size
26628 - Exception for TDS input of multipart recursive message
26645 - ESQL statement deleted if comment was defined in another Locale
26646 - 'Reply Queue*' fields NOT disabled when 'Request' is unselecte
26664 - Unknown locale "null" causes core file
26672 - load failure of brokerprincipalsg110 when broker security set
26691 - Priroty being allocated to higher level repeats in X12
26700 - MRM- Datetime Element Value does not allow 24hr clock times
26717 - Oracle package names not recognised on CALL of external (store
26725 - Warning 2235 generated during broker deploy
26738 - Arguements not being passed properly
26744 - Need way to call JVM signal handler on aben
26747 - XML - Inconsistent xmi.label for Suppress Timestamp Comment
26779 - oracle9i changes for solaris and HPUX in mqsi_setupdatabase
26787 - Invalid CCSID causes aben
26796 - configmgr jvm heap size goes negative
26805 - MRM- DateTime template value constraints can't be deploye
26820 - MRM- Decimal Output incorrectly truncated - REG
26822 - MRM- Decimal TDS Precision not working correctly
26865 - Precision should not be applied to float or decimal on iput
26867 - S210M-L20529_D solaris MRM m/set deploys fail.
26869 - Incorrect conversion of MbXException to ImbRecoverableException
26871 - Missing text continuation characters in ModelMessages file
26881 - A call to getData() on a MtiImbSyntaxElement Usertrace error
26883 - Package new UserGroup plugins
26890 - When creating the execution group, it appears to pass although
26892 - bipconfigmgr.exe dies when connection from Control Center attem
26898 - Plugin Sample code causes exception violation. (STILL)
26904 - ImbParserManager.cpp", line 76.100
26916 - Never deleting iMutex in ImbConfigurationNode::Transaction
26924 - Japanese translation error on MRProject_ja.properties
26926 - MRM-XML Suppress Timestamp Comment should be Yes/No dropdown
26931 - Unable to use new version of MA88 in Java node.
26935 - MRM-MRMB3426 causes dfe to enter tight loop
26938 - MRM-Can no longer import old message set
26950 - Params_ko.properties needs fixing
26953 - MRMBVT failing
26958 - BIP2160 abend on tdsparseitem.cpp
26964 - ClassNotFoundException when using JNDI in Java node.
26970 - MRM-2 simple consecutive compute nodes cause BIP5502E
26974 - DECIMAL paramters in stored procs don't work
26977 - BIP2110 during thread cleanup
26986 - MRM- Regression in Null Handling of FLOAT
26991 - Hyphen can be typed in the message identifier, although it is n
26996 - Length Encoded Length flagged wrong
27004 - CREATE AS does not set MODIFIABLE on Field Ref
27008 - SegV When null datasource specifie
27019 - DTDImporter- Orphan elements' type member mishandling
27034 - Remove extraneous material from DTD Importer GUI
27041 - Looping in that while loop again
27045 - looks like a problem with tdsparseitem.cpp
27052 - Problems adding to an element of CT message
27063 - With Validate set back to None - wrong values at runtime
27065 - MRM-The vast majority of MRM testcases failing on all platforms
27073 - Remove redundent compile flag MTINLS for 390
27080 - MRM-All MRM testcases still failing on all platforms
27082 - MRM- Length of Encoded Length Fiel
27086 - REGR: BIP2329 in RCD node
27087 - mqsicompcif missing from 390 tape install
27128 - SELECT fails when called from within a procedure / function
27143 - Invalid xml in deploy causes invalid flow deployment
27144 - MRM- CWF 1 Byte Integer wont accept values > 127
27145 - SELECTed CLOBS on Oracle are corrupted when > 2000 Chars
27147 - Element Qualification tab disappering for Element under Message
27153 - Missing SqlParser.tables for feature 13552.1
27154 - Lots of Errors in /MTI/MTIforBroker/MtiImbParser2
27156 - Invalid Control Center title for Event Broker version
27161 - Default Values
27171 - ERRMSG BIP 5199 Insertion not working
27184 - Create bip5448.msg in S210
27190 - MRM- Validation exceeding maxOccurs gives 5371 instead of 5375
27191 - MRM- Validation BIP5374E Insertion for MinOccurs is out by 1
27199 - Create bip5447.msg in S210
27200 - InputDestinationList correlation name not working
27201 - MRM-"SET OutputRoot=InputRoot;" causes MRM XML msg to be parse
27204 - MTIParser.cpp: Operand for "->" must be a pointer or array
27207 - Incorrect handling of NULL's in + - * / operations
27208 - DFE abend with character class regular expression
27209 - spurious delimiters occur in variable elements delimited mess
27215 - wmqpsreporttrace -e returning unexpected default trace size
27217 - MRM- Validation not carried out for flow: MQInput->MQOutput
27218 - Correct path for MtiImbValueValidator
27222 - Fixed Al3 not outputting ?
27239 - Last thread returned to thread pool in plugin input node
27245 - Good Bye FUNCTION TRACE
27246 - MRM- Tagged Delimited discards Binary el data
27248 - MRM- Interaction of Datetime Template with "I" format string
27250 - Can't deploy Fixed Length parent with Use Data Pattern chil
27253 - Too many bipmsgs for missing mandatories on output
27257 - MRM- Control Center cursor control causes craziness
27258 - MRM Validation - incomplete TDS message written with validation
27261 - Still errors in MTIImbParser.cpp
27268 - "pattern" role icon is missing
27269 - Exception caught on changebroker when serviceid not correct
27274 - Problem using reclaimed syntax elements.
27277 - SELECT with WHERE and XML.attribute does not work
27280 - ImbAbend not initialize
27286 - Closed message working like open define
27296 - Merant driver bug fixes
27297 - SOLARIS Sybase 12.5 requires update to mqsi_setupdatabase.
27302 - MRM- Broker Abend in TDS Writer
27304 - Missing initial value for Default Time Zone ID (TDS & XML)
27308 - DFE abend when using regexp 'non-greedy' ? construct
27311 - Abend in TDSParser when testing closed validation
27317 - Fix up tdsmsgnumbers.h
27319 - Debugger gives message BIP0642 after breakpoint
27342 - New AIX UserGroup Plugins cannot be loade
27348 - MRM-Abend as TDS msg (with exponent. notation floats) is parse
27351 - Self defining element at beginning of message failing
27352 - Extraneous elements in MRM tree
27360 - Choice dictionary error
27373 - New DomainAware functionality does not work as expected.
27380 - COBOL Importer fieldname containing Group
27385 - Adding PSService node to workspace
27390 - Ensure that a XML delete message always works
27401 - Writer stuck in infinite loop
27403 - Problem resolving choices with embedded seq
27404 - Validation of embedded compound types
27407 - Error handler does not print le message inserts
27413 - ConfigMgr HeapSize should check for 2045 not 4045
27422 - lookahead parameter not being rest
27428 - MRM- BIP5354 MTI Internal Error shouldnt be put in usertrace
27431 - Broker gives STL port error for SET... = UUIDASBLOB();
27439 - bip5502 internal error if DBCS data passed to TDS with DataPatt
27466 - Skip count omitted from choice length calculations
27467 - Skip count not extracted for compound elements
27468 - EndMessageContent sent before root tag encountere
27470 - MRM BIP2160 exercising EDIFACT TDS message
27471 - Error found in zh_TW mrmtoxmlmsgs.rc file
27476 - BIP2110 during thread cleanup
27485 - Abend when finding string length in PWFDateTimeFormatter
27487 - Error found in Pt-BR msg file bip9206
27491 - Sun's java symbol appears in Warning message box
27509 - validation tab in wrong place
27510 - Empty elements not editable in debugger
27525 - cpi/bsutils/RegExp... should be in Release CP210
27533 - Sample odbc file needs updating
27541 - MRM-Validation causes ELEM ON TREE WITHOUT VALUE internal error
27564 - Validation - Remove Timing from Compute node & alter defaults
27569 - Validate Escape Character atrribute
27585 - Repeated Java exc after multi EG startup or seq num recovery
27597 - Escaped Seg Terminator should not be interprete
27603 - MRM- REG Choice Variable Length All Elements Delimite
27619 - MRM- TDS Extractor should warn if Boolean reps are dift lengths
27624 - Memory cleanup and MQe multibroker fix
27626 - MRM- REG TDS Tagged Delimited repeating children within choice
27630 - Memory leaks in convertData in tdsparser.cpp
27631 - GetCompoundLength not coping with embedded types
27633 - min and max occurs not validated at compute node
27636 - STRESS: Oracle8.1.7 fails XA test (HY000)
27637 - Treat booleans as fixed in VED if all lengths are equal
27639 - Data pattern field wiping incorrectly.
27661 - Problem with wrong assertion check in MTIDict
27662 - Regular Expressions do not seem to be working well on AIX
27673 - Wrong index used in stddict getvalue calls
27674 - Change unsupported database message to a warning (BIP2112)
27683 - mqsidelete command with -q option fails
27686 - Incorrect .jar file on mqsilccsedm.bat file
27695 - MRM TDWF Length of Encoded Length not correctly greyed out
27702 - fields out of sequence following regular expression choice
27705 - Incorrect behaviour for CCI_SQL_TRANSACTION_COMMIT
27716 - NullPadFill Input for TDS Fixed Length not recognise
27718 - pattern role icon is missing
27722.1 - Disable CASE statement
27723 - FOR All/Some fails to deploy (FILTER)
27731 - Null permitted not being extracted for compounds with a base ty
27733 - MRM BIP5374 Message text
27751 - Configmgr does not start with BIP1161S
27759 - Tagged encoded lengths not output correctly
27761 - Dynamic Schema Name Resolution (DSNR)
27766 - Reduce occurance of BIP2061, BIP2062, BIP2063, BIP2064
27772 - Add new files to UNIX packages
27808 - STRESS: cannot connect to NNSY db in full mixed test on oracle
27811 - Java problems connecting to cfgmgr from control centre AGAIN.
27814 - XML/TDS prperties added in F26066 are inconsistent with CWF
27816 - Errors in ImbContextSwitch
27821 - Errors in track 27705
27822 - ABEND: EXCEPTION_ACCESS_VIOLATION
27824 - VDB BIP0642 with large message
27834 - ODE pkg.parse.mk rules error in 390 service packaging
27836 - BIP4280: Control center losses connection to visual debugger
27868 - Broker internal error when DELETE cursor followed by ATTACH
27883 - pattern role icon is missing
27899 - Setting on service trace causes 0C4 abend.
27975 - MRM TDS Boolean Extractor warnings
28004 - 390 Stored Procedure test failure
28017 - MRM- Abending TDS writer
28029 - Creation of EG - UNS cores
28039 - 0c4 abend in libMessageservices.a
28058 - Japanese install bipihlpw.chm has error
CSD02 - U482016
________________
This is CSD02 for WebSphere MQ Integrator Version 2.1 for Sun
Solaris. It is delivered via PTF U482016 for APAR IY28082.
It contains fixes for ALL the problems listed below:
Name - Abstract
IY28082 - CUMULATIVE MAINT #02 FOR WEBSPHEREMQ INTEGRATOR V2.1 FOR SOLARIS
IC31345 - COBOL IMPORT PROBLEM WHEN LEVEL NUMBERS FIELDS NOT ASCENDING
IC31419 - FORMATTER GUI CLOSING ABNORMALLY
IC31766 - MQSIREPORTTRACE DOES NOT WORK CORRECTLY WITH USERNAMESERVER COM
IC31806 - PROBLEM IMPORTING COBOL COPYBOOK. IMPORTER DOESN T LIKE EXTRA
IC31916 - MESSAGE CONVERSTION TROUBLE (SI/SO HANDLING) WITH MQSI
IC32037 - MQSI ERROR RFH2 MESSAGE FAILS TO CONVERT. REASON 2116 -
IC32047 - CONTROL CENTER HANG DURING IMPORT AND/OR DEPLOY LARGE NUMBER OF
IC32440 - ASSURED DELIVERY OF MESSAGES TO MQE NODES IN MQSI BROKEN
IC32689 - ESQL CAUSES EG TO USE 100% CPU 202 CSD02
IC32690 - BIP5502E FAILURE UNDER MQSI V210.
IC32828 - RUNTIME ERROR EXPERIENCED BY SETTING AN EXTENDED DECIMAL FIELD
IC32868 - ERROR BIP1812S IS RECEIVED WHEN IMPORTING C HEADER FILE.
IC32871 - USING ALT+CTRL+SUPP DOEN NOT REBOOT THE MACHINE DUE TO 2 EG
IC32896 - IMPORTING A C HEADER FILE ON A REMOTE CM FAILS IF A LOCAL
IC32918 - BIPCONFIGMGR.EXE TERMINATES ABNORMALLY WHEN MACHINE IS REBOOTED
IC32929 - ESQL STATEMENT IS DELETED IF COMMENT DOESN'T MATCH CC LOCALE
IC33040 - INSUFFICIENT DETAIL ON IMPLEMENTING THE PASSTHRU FUNCTION.
IC33053 - ERROR BIP0067E WHEN SWITCHING THE CONTROL CENTER ON THE MACHINE
IC33066 - MQE NODES SENDING DUPLICATE MESSAGES
IY16738 - START OF BROKER CONFIGURED TO USE ORACLE GIVES AN ERROR - UNABL
IY21734 - MQSI CRASHED WHEN ADDITIONAL INSTANCES OF A MESSAGE-FLOW WERE
IY26160 - BIP2088E AND BIP2106E AT DEPLOYMENT OPERATION.
IY26427 - A BROKER RUNNING WITH A SCADA NODE LOCKS UP
IY28136 - PART OF DBCS CHARACTERS CHANGED TO DIFFERENT CHARACTERS WHEN
IY28181 - NEONRULES AND NEONFORMATTER WERE NOT ABLE TO BE IMPORTED.
IY28384 - CONTROL CENTER FOR V2.01 INCORRECTLY IMPORT VIRTUAL
IY29050 - ORA-03113: DFE TERMINATES AFTER A LOSS OF COMMUNICATION WITH DB
PQ57196 - MESSAGES COMING FROM OS/390 TO AIX/NT THROUGH NETVIEW FAIL TO
PQ57697 - ERROR OCCUR WHEN MQSICHECKSETUP IS CHECKING THE DSNAOINI FILE.
PQ57991 - DATAFLOWENGINE GOES INTO LOOP IF TRACE NODE IS USED AND THE
f13625 - Output Parameters And Stored Procedures
f18699 - Multipart Message support
f21571 - Optimize CC when performing commands on list of entities
f21825 - Enhanced Select Function
f22439 - NULL Handling - MRM Parsers
f23840 - Enhanced MOVE statement
f23842 - Enhanced CREATE function
f24317 - Allowing expressions in field references
f24543 - Further implementation of MM messages for runtime broker
f24706 - Relax delimiter limitations in TDS messages.
f24859.1- Add new Data Element Separation method "Undefined"
f25032 - Natural ESQL support for local elements & attributes
f25127.1- EUI changes for Enhanced THROW statement
f25730 - Modifications to DTD Importer handling of PCDATA Elements
s25795 - C,D,F packed flags in CWF
19930 - MRM BIP5167 error message bad substitution
20220 - BIP5117: Message full of meaningless numbers and terms
20685 - MRM/CWFLayer: CT with base type elements not properly rendered
20737 - Event ID 3100 not found
20787 - ERRMSG: MRM BIP5294 message content
21542 - Netview not seeing all response messages from MVS command
21989 - MRM MRMB0406 padding char SPACE
22232 - MQSI2.1:Default value of Transaction mode in data nodes
22831 - Unhelpful BIP5170 when input CWF msg does not match definition
22962 - GUI:User Role 'Op. Domain Controller' has misleading help text
22963 - GUI:User Role 'Topic security administrator' misleading help
23120 - MRM: Event 5199 Description inaccurate
23293 - MRM-problems writing simple types to bitstream.
23334 - CCSID and ENCODING clauses not reported in User Trace
23377 - MRM- CC online Help needs "Check-Out List" explanation adding
23381 - MRM BIP1929 message content
23536 - BIP2801E on broker startup if DB2 not installed
23589 - Trace node properties - File Path
23755 - MRM-DTD Importer imports an element with a strange icon
23855 - OVERLAY function causes unexpected truncation
23899 - missing repetitions incorrectly handled
23916.1 - MRM-Element Value value should not be mandatory
24225 - PubSub RC 3073 to ReqUpdate after successful RegSub
24284 - MRM-Imported DTD cannot be deployed
24294 - ERRMSG: 5339 Dictionary Iterator Error Event Viewer Insertions
24296 - SQLCODE and SQLNATIVEERROR wrong when TEODBE=NO
24297 - EXCEPTION_ACCESS_VIOLATION (C0000005).
24359 - ERRMSG: Wellformed XML but 5285 gives insufficient diagnostics
24364 - Repeating choice not working for Tagged Fixed Length 24364.1
24379 - MRM-Extractor logs only show last message set
24385 - Can't get rid of warning box in MRM after changing a choicetype
24416 - Oracle 8.1.6 SEGVing in Malloc
24417 - ERRMSG: MRM Type composition sequence error diagnostics poor
24420 - CWF Repeats not working with Choice Type
24421 - STRESS: SQLstate HY010 on aggregation test
24438 - Choice with compound elements not working for some TDS formats
24450 - ERRMSG: TDS 5436 missing Insertion for Element Name
24456 - TDS-XML : MRM give an existing identifier who can't be altered
24461 - MRM incomplete messageset errormessages need substitution vars
24469 - Rules can no longer reference fieldnames containing parentheses
24477 - BIP5339: Confusing error caused by simple mistake
24481 - STRESS: Extra added by mrm parser
24484 - SEGV in broker when DB2 terminated
24488 - MRM- Base type data causes following child lost from bitstream
24496 - Invalid MF copy/assignment while broker copy/paste
24522 - output of ''asdf'' instead of 'asdf'
24523 - MRM Cobol import of extended decimal
24524 - impexp with -x swaps public and system ids.
24542 - ERRMSG: BIP 5512 Insertion not working
24550 - Message Format on Input Node supplies incorrect value
24558 - In the XML plugin, the DOCTYPE text property currently has maxl
24559 - Parsing of unresolved choice not working for tds formats.
24564 - Generic XML : Bad DOCTYPE causes DataFlowEngine to loop
24568 - Cannot parse TDS repeating embedded compound types
24592 - Error with coordinated transaction value
24602 - BIP2605E should not be issued for OS/390 (as use RRS for TC)
24619 - GUI Check Out list does not complete
24620 - MRM Invalid datetime allowed on output TDS
24621 - MRM-DATETIME Bizarre week in month behaviour when handling the
24624 - MRM - XML as reserved word stops you defining an XML namespace
24627 - At the moment when the extractors run, the log files for it are
24628 - DFE failure with large MRM message
24638 - MRM-DATETIME Make binary and ICU DATETIME parsers have common b
24674 - Ongoing TDS Parser work with EDIFACT Imp
24697 - Break point infront of AggregateReply node corrupts message
24709 - TDS Problems with EDIFACT smaple from IMP
24710 - Embedded message support for compute node customizer.
24716 - Multiple calls to hasnext cause TDS to fail
24739 - Can't deploy messageset with embedded msg and C lang binding.
24746 - Nested repeats for zero occurrences not working correctly.
24758 - Do not handle Nested message from within MatchEmbeddedCompound
24810 - Database SELECT does not work any more
24831 - Gen XML: setElementChanged is inefficient
24834 - Root message name being lost
24846 - MRM- Base Type BINARY for XML not working
24850 - BIP1836E is showing wrong log names.
24870 - AIX NNSY messages are not being found
24877 - AggregateReplyNode binds incorrectly to VARCHARs
24878 - Reference variables in a SELECT's FROM clause does not work
24889 - Inserts in exceptions thrown by the THROW statement are wrong
24906 - MRM- XML CDATA tag for BINARY element (Null Handling related?)
24908 - MRM BIP5521 output extended decimal in codepage 1200
24961 - Uuid generation failures bring down Config Mgr
24968 - Unable to go from one message set to another
24979 - MRM/Length and Repeat Count Types are wrongly editable combo
24986 - Improve performance of CC<->CM MRM interaction
25010 - Broker restart fails with Aggregation on SQL Server
25028 - Config manager gets CPU bound when deploying large message sets
25029 - Parsing of unresolved choice not working for tds formats.
25037 - Repeating choice failing for TDS when choice is compound type
25041 - Choice with compound elements not working for some TDS formats
25060 - MRM-Simple message brings down the execution group
25089 - MRM- NullPadFill shouldnt be offered for certain PhysicalTypes
25123 - Mnemonic escape character not translated properly in TDS
25152 - XML Extractor doesn't complain when attributes occur after elem
25168 - Incorrect string in Interpret Element Value attribute dropdown
25216 - Attributes cannot be specified on the XML Root Tag Name
25226 - Negative value in 'numeric' field causes DFE to exit()
25227 - MRM - TDF & XML parser does not handle a Message Type Path
25232 - TDS defaults not displayed correctly
25243 - Rules for REPEAT,Min/Max Occurs on connection tab are incorrect
25244 - Repeating choice not working for Tagged Fixed Length
25247 - Message/explanation/user response text in :MSG./:EMSG. tags.
25248 - MRM CWF Null handling datetime NULLLogicalValue causes BIP5168
25257 - MRM/Length and Repeat Count Types are wrongly editable combo
25260 - MRM CWF Length Value Of eligible fields not displayed
25262 - Can no longer use CWF Repeat Count Value of field in GUI
25300 - MRM- Base type Decimal causes Internal MTI Error 5354E
25302 - Apply button wrongly greyed on CompoundType TDS view
25312 - MRM- TDS Writer Problem with Simple Types
25316 - MRM- CWF Datetime String not being recognised
25322 - MRM-CWF Tab, Boolean Value fields are not editable.
25323 - Console output not coming back to USS via oeconsol command
25338 - MRM BIP1951 Wrong substitution
25339 - GUI Check Out list does not complete
25340 - DATABASE exception shouldn't be accepted in THROW syntax
25341.1 - Increase timeout in MQConnection for MRM_WAIT_INTERVAL
25341.3 - Config. Mgr. gets knotted after 1350 timeout. BIP1350E
25341.3.1 - CC sends to CM several times same request after Time-Out.
25341.8 - Add filtering for MRM member collections
25342 - Backend still won't allow ^ in an identifier when it is checked
25347 - TDS input, empty values not being recognised as NULL
25355 - XML undefined attributes on 390 not rendered correctly
25356 - TDS Validator - ensure DEST=Undefined if CT composition=MESSAGE
25358 - Sybase broker hangs on restart w/multiple msg sets
25361 - Trace updates for improved MOVE command
25362 - nullPointerException when trying to check a Message Flow
25364 - MRM - Absolute Message Type Path doesn't work with Message Type
25366 - MRM Check out attempted even though messageset is frozen
25370 - Incorrect data element separation for type cmposition = message
25375 - Default DateTime ElementValue causing an exception on deploy.
25378 - Unable to SET a REFERENCE to a DECLARED variable
25385 - MRM COBOL import fails if >1 space after OCCURS
25386 - DTD Importer - Type content should be closed to model valid DTD
25391 - UNA and ISA written out with extra delimiter
25391.1 - GUI must allow non empty delimiter in some cases
25394 - CacheManager should be symmetric about UUID usage
25409 - CC not allows duplicate name on message set in special case
25414 - Parsing of repeated nesting items not working
25419 - MRM-DATETIME Bizarre week in month behaviour when handling the
25424 - Generation of dictionaries on delta deploy
25428 - ERRMSG: CWF 5181 Insertions not working
25432 - MRM BIP5440 incorrect substitution
25434 - CC model optimisations are not reset when exceptions occur
25445 - Error introduced after GA for bip8120.msg
25447 - Bracket flagged in error in function call
25451 - TDS Validator - ensure DEST=Undefined if CT composition=MESSAGE
25457 - EUI MRM Tagged Delimited requires Tag Length/Separator
25482 - Recursive multipart mess causes DFE abend TDS->XML
25493 - TDS Writer is too memory intensive
25509 - Tagged Delimited Choice with compound elements failing.
25525 - bip5503.msg, error in the text of the message.
25532 - Storage not released by DataFlowEngine after PROPAGATE
25533 - DataFlowEngine and ConfigMgr terminate on NT user logoff
25547 - MRM TDS unresolved choice
25550 - Inconsistent format TDS->XML vs XML->TDS with embedded messages
25558 - MRM-Using Message Type Path with CWF causes broker abend
25566 - In the EUI, Attributes cannot be specified on the XML Root Tag
25585.1 - The EUI needs to validate dateime values entered.
25602 - Bad expiry value causes MQe message flow failure
25610 - uid(0) causing problems in changebroker and
25617 - Modification of message identifiers and XMLName by DTD Importer
25662 - XML msg without top level causes DFE process to go into spin
25666 - Wrong path is generated for embedded message.
25668 - MRM- Null Handling CWF String input blank NullLogicalValue
25675 - TypePrefix incorrectly handled
25696 - keywords CREATE FUNCTION, PROCEDURE, REPEAT case sensitive
25697 - MRM Check in all with new 'message' compound type
25699 - MRM- Broker Abend with XML Input
25723 - XML: encoding null NUMERIC, fails to parse on NULLElement
25727 - DTD Importer doesn't support recursive elements
25741 - MRM finalize messageset not reflected in GUI
25745 - CWF length units End of Bitstream not working
25753 - TDS Parser does not support syntax level B defaults
25761 - Unable to set (XML.Attribute)Type using SELECT
25763 - MTiImbparser - support for string -datetime conversion
25783 - First week of year not used in runtime
25794 - Input node parser name attributes not being set
25802 - Valid syntax gets flagged in error
25827 - MRM- Hex Delimiter Substitution wrong in tdscommonutils
25842 - MRM- Null Handling TDS Fix Length Float NullLog / Pad confusion
25868 - Add Message List for a closed compound type is not filtered
25869 - Incorrect Value set for ISA INdexes
25885 - STRESS: Core File from UserNameServer in Mixed test
25895 - Repeating Tagged Fixed Length choice type is still failing.
25917 - Contol Centre loops trying to expand message with lots of child
CSD01 - U481438
________________
This is CSD01 for WebSphere MQ Integrator Version 2.1 for Sun
Solaris. It is delivered via PTF U481438 for APAR IY26924.
It contains fixes for ALL the problems listed below:
Name - Abstract
IY26924 - CUMLULATIVE MAINTENANCE #01 FOR WEBSPHEREMQ INTEGRATOR FOR SUN
IC30350 - CONFIGMGR NOT DETECTING A COMM LINK FAILURE TO DB2, GIVES
IC31802 - GETTING PARSING ERRORS WHILE USING REPEATING FIELDS WITH SECOND
IC32018 - WHEN A REFORMAT OPEARTION IN A NEON-MQSI-ENVIRONMENT DOES NOT
IC32309 - WITHIN THE COMPUTE NODE, WHEN CODING A PASSTHRU FUNCTION
IY16738 - START OF BROKER CONFIGURED TO USE ORACLE GIVES AN ERROR - UNABL
IY24362 - CONVERSION INCORROUT FROM OS/390 TO OS/400.
IY25465 - SCADA NODE STOPS TRANSMITTING MESSAGES THROUGH FLOW
f16275 - Command line deploy utilities
f22449.2- Add non loaded skip node to the workspace when it is check out.
f23369.2- ESQL Syntax Checker:Removing field names checking in path
f24217 - Complete set of diagnostic streams written under single JOBLOG.
readme - README updates for CSD01
17377 - mqsilist shows old 'default' execution group label after rename
18007 - Cannot set breakpts in main flow whilst in subflow
18405 - mqsichange with -a only does not work
18578 - mqsicreate / mqsichange -t on AIX has no error msg
19673.4 - Handling errors during lil loading on OS390 (and related)
20308 - MRM-mqsimrmimpexp gives wrong errMsg when MsgSet is incomplete
20408 - report trace command does not produce output on OS/390
20847 - MRM-BIP1930E is not displayed correctly
20946 - DateTime - MRM Parser Support
20955 - MRM Data Converters - handling overflow
21208 - The control centre users don't accurately reflect real users
21208.1 - Primary group not enumerated on Unix Platforms.
21569 - OS390 Java Plugin Node Message Flow Failure
21724.1 - New message for 21724.
21843 - MRM-DATETIME processing ignores Era
22144 - MRM-self-defining messages unacceptable to MRM.
22233 - The call ImbWstring MtiImbParser::getTypeProperty()
22249 - MRM TDS Writer is writing TagDataSeparator when not required
22521 - Control Centre unacceptably slow when adding types to workspace
22548 - MRM-Modified Self defining elems cause DI prob.
22548.1 - XML - Parsing of compound types with attribute values results
22654 - 0 occurrences of repeating elem/comptype failing for CWF input.
22687 - Java trace not working from registry setting
22751 - GMTTIMESTAMP gives incorrect results
22836 - MRM-XWF Encoding not working
22925 - BIP1802: error message gives wrong name for database.
23037 - The time zone in the JVM is being set explicitly to GMT.
23147 - Typing 'NNRie' produces help text, core dump.
23232 - TDS strings containing just spaces fail on AIX
23283 - Poor performance of TextBufferReader::getStringSection
23309 - Change description on STEPLIB_NAME
23316.1 - Maintain list of delimiters to improve TDS performance
23316.2 - In TDSParser SetReservedChars not reset for service string
23318.1 - Memory Leaks in MRM (part 2)
23341 - MRM: CWF Parsing of Choice Message Fails
23351 - 390 debugger will not start
23359 - Swift importer fails with "attribute name is not defined"
23404 - Unnecessary bracket in syntax description of "mqsichangetrace"
23429 - Dataflow engine terminates on large (cust mfs) debug deploys
23453 - Trace formatting-ImbMqOutputNode::evaluate
23476 - Expiry event not available
23543 - Boolean default values not being written out on Unix and 390
23545 - MtiImbParserFactory in stdout
23558 - Ambigious behaviour for TDS messages with choice type.
23568 - Repeating choice not working correctly
23568.1 - Some problems with choices.
23577 - STRESS: Code dump in mixed test with NNSY nodes on HP
23581 - Command line deploy - misleading message
23583 - CWF output not working for nested choice
23584 - Cmd line deploy - no message returned when -f and -a flags used
23599 - Currently the Java node API does not have an onDelete( ) method
23603 - STRESS: Resource Manager errors in XMLtoMRMCWF
23605 - Trace entries missing in textbffr.cpp
23614 - MRM-Errors handling binary DATETIMEs
23622 - mqsiimportmsgflows - no msg flow in Control Center
23627 - dump formatter abends no memory
23651 - Add fields in the ESQL and the new field gets concatenated
23667 - STRESS: SQL server error in aggregation tests
23670 - MRM-DTD Generator does not append * and + to appropriate elems
23690 - Update .bat files
23697 - Some console messages have old format
23701 - cb command converts mixed case lilpath to lower case -
23703 - MRM-problem with encoding for base type binary.
23713 - bipjrst reset broker has problems with clearing one queue
23719 - wmqpslist coredumps on Unix
23722.1 - Implement MRM implicit casts at the tree level
23723 - Wrong description of console command is displayed
23738 - Shipped Merant ODBC driver still "available" after uninstall
23765 - abend in ImbUGI because of wrong linkage
23784 - Unable to debug failing flow
23786 - MRM-can assign any string value to Boolean elements.
23796 - MRM BIP2160 trying SWIFT -> CWF
23805 - EXCEPTION_ACCESS_VIOLATION (C0000005)
23819 - FI events during TDS parser destruction
23821 - MRM BIP1951 if no length on CWF endofbitstream
23852 - STRESS: Core File from UserNameServer in Mixed test
23856 - Refresh New Era SysMgmtGuide in docs pkg
23861 - MRM Add mapping for element in nested type
23882 - TDS:LengthValueOf is enable if embedded simple(element or type)
23892 - MRM: Deleting types can cause database corruption.
23894 - MRM rounding of CWF packed decimal on input
23896 - MRM CWF null chars in fixed length strings
23899.2 - TDS Writer related fixes for defect
23905 - MRM-DTD Imprtr,typng non existent filenme doesn't throw an err
23906 - OS390: mqsicreatebroker allows invalid broker name.
23907 - Disconnecting Scada client on HP-UX causes EG failure
23915 - MRM-DTD Imprtr creates uncheckinable cmpts names frm valid DTDs
23921 - MRM-all inline DTD data dropped on output.
23922 - MRM BIP5285 with TDS unresolved choice
23929 - Parsing BaseType in TDSParser
23947 - Config manager and broker JVM's share a CLASSPATH
23961 - Abend during MbSQLStatement object usage
23966 - MRM BIP1950 because EDIFACT Boolean Null not set
23972 - STL exception in TDSValidator::ValidateMessageTypeMembers
23974 - Memory leak in MRM2 with TDS fix
23983 - IC01 ref to the jar files jfaceall and argo
23994 - No error messages for out of range timeoutvalues
23997 - CoreDump on Mqsistop + unexpected syslog messages.
24001 - MRM BIP0068 on save to shared
24005 - Stopping and starting broker causes database errors
24038 - Unable to connect from English CC to Chinese ConfigMgr
24044 - Problem with TDS repeating structures.
24047 - Unable to change type composition from sequence to choice.
24055 - Database deadlocks on SQLServer when unknown timeout set
24061 - MRM2 Data Datatype test failing
24068 - MRM BIP5339 input EDIFACT/TDS output XML
24073 - STRESS: Memory Leaks shown in Bitstream Parsing tests
24075 - CWF Repeats not working with Choice Type
24077 - Unexpected response to invalid input
24078 - Problem with compound elements/types within a choice type.
24089 - MRM: TDS Mnemonics aren't used by the TDS Parser or Writer.
24089.1 - MRM: TDSWriter not writing mnemonics
24098 - Memory used after being freed in choice parse
24102 - MRM-error message 5503 when copy input root to output root.
24104 - MRM-problem rendering compound element with base type.
24110 - Can't check in repeating choice with Repeat Type as "Value of"
24112 - SIGSEGV raised but DFE did not core dump
24115 - MRM: TDS Parser causes execution group restart when RFH2 type t
24126 - DataCnv::IntFromInt32 and DataCnv::IntFromInt64 need to check f
24133 - MRM: STL Errors & exec group restart on Debug builds
24144 - MRM - INT64_MAX is not defined properly on AIX
24148 - MRM TDS fixed length output max occurrences instead of actual
24152 - MRM-broker abend following TDS length delimited message input.
24175 - NNRie and the Rules GUI rearrange long rules
24176 - MRM TDS BIP5425 on fixed length repeat structure
24177 - RC=9 from BIP$JCVP verification job.
24180 - TDS Fixed Length Choice padding is incorrect.
24181 - MVS: Broker will not start with DB2_CONVERSION=MIXED
24203 - MRM EDIFACT UNA not processed in runtime
24206 - MRM-standalone clause in xml declaration not suppressed.
24220 - TDS read error, integer -99 out of bounds
24223 - The XL plugin doesn't return the default for for anonymoud simp
24229 - NNFie loses data from Substitute output operations
24230 - Error with TDS decimal->CWF packed decimal when precision set
24235 - MRM-MRMB0425 is failing on all platforms
24235.1 - Same problem as 24235 but on parsing.
24240 - The defaults for XML SYSTEMID and PUBLICID are the wrong way ar
24252 - MRM BIP5426 Tagged fixed length repeating elements
24260 - Using mqsideploy the wrong BIP message is displayed
24261 - MRM-DTD Importer produces AQM7610E when importing a DTD
24263 - TDS fixed length compound choice type not working.
24265 - Parameters in call to IMBDate are reversed
24273 - CPI/MTIIMBParser Error messages
24277 - MRM-MRMB0815 & 1115 fail on OS/390
24280 - NNSY Formatter GUI does not import format definitions
24288 - ERRMSG: TDS Missing Group Indicator BIP5437 insertion
24290 - MRM TDS X12 default settings wrong
24293 - Some CWF exceptions logged even though they are not errors
24303 - Incorrrect message number used for incomplete message set
24319 - MRM-MRMB1815 is failing on Windows
24321 - Pointer returned from cniGetEnvironment is wrong.
24326 - IC01 export return codes (nonexistant flow)
24327 - Parsing of EDI problem
24340 - Requested changes to NNInstall-Driver.sh
24346 - ERRMSG: TDS Incorrect Issuing of 5422
24354 - definition of UINT32_MAX has a ... at the end of it
24361 - ERRMSG: CWF 5181 Insertions not working
24368 - MRM BIP1949 trying to check in Type
24381 - ERRMSG: Generation of new BIP msg req'd for use by TDS Parser
24392 - ERRMSG: CWF Expected BIP5520 Not Being Thrown
24395 - ERRMSG: BIP5519 Insertions not working
24402 - MRM- Generate Message Book HTML contains nonsense min/max info
24406 - mqsifindandreplace replaces keyword with quoted lowercase equiv
24411 - MRM-elements rendered as xmlAttribute not found.
24412 - MRM-XML/Perf: broker keeps opening ccsid 0
24413 - MRM: TDSWriter has problems writing out message, and gives litt
24431 - XML message in Double Byte char set causes XML ParserException
24437 - Syntax error reported from quoted field references
24453 - MRM BIP8184 message content unclear
24457 - IC01: -r flag used in mqsideploy causes timeout
24462 - onDelete method java call stops msg flow deletion.
24474 - Dictionary delete memory that was allocated with malloc.
24475 - MRM BIP1812 manipulating large message
24560 - Field in ENVFILE not updated
24581 - Missing line feed escape in BIP0001E error message
24656 - MRM-DATETIME CSD1 regression. Packed decimal DATETIMEs not pars
24661 - wrong description of Compound Type attribute Type Composition
24675 - MRM - CWF->CWF Cannot write out to different output msg
24698 - MRM-Irrelevant ESQL makes message fail inexplicably
GA PTF - U481172
________________
This is the GA PTF for WebSphere MQ Integrator Version 2.1 for Sun
Solaris. It is delivered via PTF U481172 for APAR IY26071.
It contains fixes for ALL the problems listed below:
Name - Abstract
IY26071 - CUMULATIVE MAINTENANCE #00 FOR WEBSPHEREMQ INTEGRATOR FOR SUN
17516.1 - MRM BuildTimeDictionary S202 Compatibility
19673.3 - Handling errors during lil loading on OS390
20950.2 - MRM - Integer Backward compatibility support - RTD Extractor
21724 - S000-L010828_D Java exception when starting execution group
21930 - Oracle XA on NT makes the execution group unresponsive.
22534 - MVS: Data corrupted when Inserting DBCS string in Database node
23218 - MRM MRMB5914 extended decimal dbcs codepage
23351.2 - 390 debugger will not start
23379 - Choice has stopped working for CWF messages
23683 - STRESS:Abend file produced during destruction of a locale objec
23813 - STRESS: Memory corruption with large repeating message
24013 - MRM-cwf error writing to bitstream causes broker abend.
24031 - NNSY Updates to CIF file ( mqsicompcif)
24056.1 - Rename WMQI_REGISTRY
24079 - Allow import to proceed even if importing attributes fails.
24255 - Update classpath within envnt.bat
24256 - BIP$JCVP job should return Java
24323 - Sample nnsyreg.dat needs changing to include NNFIE session
####### ------- end of file MEMO.PTF ------------#######################