Business objects DATA INTEGRATOR 11.7.3 RELEASE NOTES 10-2007 User Manual

BusinessObjects Data Integrator Release Notes

Data Integrator 11.7.3

October 31, 2007

© 2007 Business Objects. All rights reserved. Business Objects owns the following U.S. patents, which may cover products that are offered and licensed by Business Objects: 5,555,403; 6,247,008; 6,289,352; 6,490,593; 6,578,027; 6,768,986; 6,772,409; 6,831,668; 6,882,998 and 7,139,766. Business Objects and the Business Objects logo, BusinessObjects, Crystal Reports, Crystal Xcelsius, Crystal Decisions, Intelligent Question, Desktop Intelligence, Crystal Enterprise, Crystal Analysis, Web Intelligence, RapidMarts, and BusinessQuery are trademarks or registered trademarks of Business Objects in the United States and/or other countries. All other names mentioned herein may be trademarks of their respective owners.

Third-party Contributors

Business Objects products in this release may contain redistributions of software licensed from third-party contributors. Some of these individual components may also be available under alternative licenses. A partial listing of third-party contributors that have requested or permitted acknowledgments, as well as required notices, can be found at: http://www.businessobjects.com/thirdparty

About these Release Notes 7Chapter 1

Supported platforms and versions 9Chapter 2

Compatibility update..................................................................................10

Limitations update .....................................................................................15

Migration considerations 23Chapter 3

Logs in the Designer..................................................................................24

Data flow cache type.................................................................................24

Pageable cache for memory-intensive data flows.....................................24

Embedded data flows................................................................................25

Oracle repository upgrade.........................................................................25

Solaris and AIX platforms..........................................................................26

Data quality................................................................................................27

To migrate your data flow to use the new Data Quality transforms...........28

Distributed data flows................................................................................29

XML Schema enhancement......................................................................29

Password management.............................................................................30

Web applications.......................................................................................30

Web services.............................................................................................31

Supported products..............................................................................10

Unsupported products..........................................................................15

BusinessObjects Data Integrator Release Notes 3

Contents

Deployment on UNIX 33Chapter 4

Adapter interfaces 35Chapter 5

JMS adapter interface................................................................................36

Salesforce.com adapter interface..............................................................36

Documentation updates 37Chapter 6

New features in version 11.7.3..................................................................38

HP Neoview database support.............................................................38

Inxight integration for reading unstructured text ..................................38

base64_encode..............................................................................39

base64_decode..............................................................................40

Job Server enhancements...................................................................41

Getting Started Guide................................................................................41

Citrix Presentation Server version 4.0 .................................................41

AIX user resource limits.......................................................................42

Designer Guide..........................................................................................42

Changed-data capture for Microsoft SQL Server 2005 .......................42

To configure publications for Microsoft SQL Server 2005 CDC...........43

Creating Microsoft Excel workbook file formats on UNIX platforms.....44

To create a Microsoft Excel workbook file format on UNIX .................44

Reference Guide........................................................................................45

Maximum number of loaders................................................................45

Reserved words...................................................................................46

Microsoft Excel workbook format for UNIX..........................................46

Salesforce.com Adapter Interface Guide...................................................47

4 BusinessObjects Data Integrator Release Notes

Contents

Resolved issues 49Chapter 7

Known issues 83Chapter 8

SNMP copyright information......................................................................90

IBM ICU copyright information...................................................................92

BusinessObjects Data Integrator Release Notes 5

Contents

6 BusinessObjects Data Integrator Release Notes

About these Release Notes

Welcome to BusinessObjects Data Integrator XI Release 2 (XI R2) Accelerated version 11.7.3.0. This version is a functional release that includes the following new features:

• HP Neoview database support

• Inxight integration for reading unstructured text

• Job server enhancements that now allow up to 50 Designer clients with

no compromise in response time

See Documentation updates on page 37 for descriptions of how to use these new features.

Please read this entire document before installing your Business Objects software. It contains important information about this product release including installation notes, details regarding the latest resolved and known issues, and important information for existing customers.

To obtain the latest version of Data Integrator documentation including the most up-to-date version of these Release Notes, visit the Customer Support documentation download site (http://support.businessobjects.com/documen

tation/) and follow the appropriate product guide links.

8 BusinessObjects Data Integrator Release Notes

Supported platforms and versions

Compatibility update

This section describes changes in compatibility between Data Integrator and other applications. It also summarizes limitations associated with this version of Data Integrator.

For complete compatibility and availability details, see the Supported Platforms documentation on the Business Objects Customer Assurance Web site at:

http://support.businessobjects.com/documentation

Compatibility update

This section describes:

Supported products on page 10

•

Unsupported products on page 15

•

For complete compatibility and availability details, see the Supported Platforms documentation on the Business Objects Customer Assurance Web site at:

http://support.businessobjects.com/documentation

Supported products

Data Integrator now supports the following products as of this release.

• SAP BI 7

• Data Quality 11.7 and 11.7.1

• Microsoft Internet Explorer 7

• Sybase ASE 12.5 on Solaris and AIX

• HP Neoview 2.2 native support on Windows and Linux

• IBM DB2 9.1

• Netezza 3.1.4

• Oracle E-Business Suite Release 12

• Sybase IQ 12.7 (source and target)

• Teradata V2R6.2 on Windows and Linux

• Impact analysis will interoperate with BusinessObjects XI R2 SP3

universes, Web Intelligence documents, Desktop Intelligence documents, and Business Views

10 BusinessObjects Data Integrator Release Notes

Supported platforms and versions

Compatibility update

• Citrix Presentation Server version 4.0. For more information, see

documentation for Citrix support on the Business Objects developer community Web site at http://diamond.businessobjects.com/EIM/DataIn

tegrator.

• BusinessObjects Text Analysis for Data Integrator version 11.5. Check

the the Business Objects Customer Assurance Web site at http://sup

port.businessobjects.com/documentation/ for the availability of

BusinessObjects Text Analysis for Data Integrator version 11.5.

Data Integrator also supports the following products as of version 11.7.

• Support is available for the following platforms and configurations:

• DataDirect ODBC Driver version 5.2

• Data Federator XI R2 version 11.5

• Data Federator XI R2 Accelerated versions 11.6.2 and 11.7.0

• Salesforce.com AppExchange API for Enterprise version 7.0

• .NET support for Web services

• SAP ECC 6.0 via ABAP, BAPI and IDOC

Please note that this support is only available on the latest versions of Rapid Marts. Check your Rapid Marts documentation to get the latest support information for that product.

• Excel version 97, XP, 2000 and 2003 as a source (please review the

Limitations section for Excel support on UNIX)

• Impact analysis interoperates with:

• BusinessObjects universes 6.5.1, XI, and XI R2 SP2

• Web Intelligence documents versions 6.5.1, XI, and XI R2 SP2

• Desktop Intelligence documents versions 6.5.1, XI, and XI R2 SP2

• Business Views XI and XI R2 SP2

• The scheduling functionality is supported with BusinessObjects version

XI R2 or later versions of the BusinessObjects Enterprise scheduler

• The following Universe Builder versions are compatible with Data

Integrator 11.7:

• The stand-alone Universe Builder version 11.5 bundled with Data

Integrator 11.7. This stand-alone version of Universe Builder will interoperate with Business Objects version 6.5.1 and XI. Find Universe Builder documentation in the "Building Universes from Data Integrator metadata Sources" chapter of the Universe Designer Guide, which you can download from the Business Objects Customer Assurance Web site at http://support.businessobjects.com/documentation/. Select

BusinessObjects Data Integrator Release Notes 11

Supported platforms and versions

Compatibility update

• The Universe Builder version bundled with BusinessObjects Enterprise

• ODBC to generic database support has been validated with the following

database servers and ODBC drivers. For complete compatibility details, see the Business Objects Supported Platforms documentation on the Business Objects Customer Assurance Web site at: http://support.busi

nessobjects.com/documentation

• Microsoft SQL Server 2000 via DataDirect Connect for ODBC 5.2.

• MySQL version 5.0 via ODBC driver version 3.51.12 on Windows and

• Red Brick version 6.3 via ODBC driver version IBM 6.3

• SQLAnywhere version 9.0.1 via ODBC driver version Adaptive Server

• Sybase IQ version 12.6 requires ESD 4

• Log-based changed-data capture (CDC) works with the following database

versions

• Oracle version 9.2 and above compatible versions for synchronous

BusinessObjects Enterprise as the product and Universe Designer Guide as the document.

XI R2, XI R2 SP1, XI R2 SP2, and XI R2 SP3 (see the required BusinessObjects Enterprise XI R2 SP1 CHF-13 patch information in the Resolved issues section of this document for problem report ADAPT00534412).

MySQL version 4.1 via ODBC driver version 3.51.10 on Windows and UNIX.

Note: Driver version 3.51.12 is required in the multibyte environment.

Anywhere 9.0

CDC and Oracle version 10G and above compatible version for asynchronous CDC

Note: Changed-data capture for Oracle 10G R2 does not support

multibyte table and column names due to a limitation in Oracle.

• Microsoft SQL Server 2000 and 2005

• IBM DB2 UDB for Windows version 8.2 using DB2 Information

Integrator for Replication Edition version 8.2 (DB2 II Replication) and IBM WebSphere Message Queue version 5.3

• IBM DB2 UDB for z/OS using DB2 Information Integrator Event

Publisher for DB2 UDB for z/OS and IBM WebSphere Message Queue version 5.3.1

• IBM IMS/DB using DB2 Information Integrator Classic Event Publisher

for IMS and IBM WebSphere Message Queue version 5.3.1

12 BusinessObjects Data Integrator Release Notes

Supported platforms and versions

Compatibility update

• IBM VSAM under CICS using DB2 Information Integrator Classic Event

Publisher for VSAM and IBM WebSphere Message Queue version

5.3.1

• Attunity for mainframe sources using Attunity Connect version 4.6.1

• NCR Teradata with the following server versions, client versions, and

ODBC drivers:

ODBC driverClient versionTeradata server version

03.06.00TTU 8.2V2R6.2

03.05.00TTU 8.1V2R6.1

03.04.00TTU 8.0V2R6

03.03.00TTU 7.1V2R5.1

This support requires the following TTU patches:

• On Windows platforms:

• cliv2.04.08.00.01.exe

• tbld5000.05.00.00.01.exe

• psel5000.05.00.00.01.exe

• pdtc5000.05.00.00.01.exe

• npaxsmod.01.03.01.00.exe

• npaxsmod.01.03.00.02.exe

• npaxsmod.01.03.00.01.exe

• On Linux platforms:

• cliv2.04.08.00.01.exe

• mqaxsmod.01.03.00.00.exe

• npaxsmod.01.03.01.00.exe

• npaxsmod.01.03.00.01.exe

With Teradata V2R6, the named pipes mechanism for Teradata bulk loader is supported.

Note: Teradata does not recommend using Teradata Parallel Transporter

in TTU 7.0 and 7.1.

BusinessObjects Data Integrator Release Notes 13

Supported platforms and versions

Compatibility update

• Data Integrator uses Apache Axis version 1.1 for its Web Services support.

Axis version 1.1 provides support for WSDL version 1.1 and SOAP version

1.1 with the exception of the Axis servlet engine.

• Data Integrator connectivity to Informix servers is now only supported via

the native Informix ODBC driver. Version 2.90 or higher compatible version of the Informix client SDK is required.

• Designers on Japanese operating systems working with Job Servers on

Japanese Windows computers.

When using the Designer on a Japanese operating system and working with either an English Job Server on UNIX or a Job Server on Windows on a non-Japanese operating system:

• Messages from the back end display in English

• Business Objects recommends using a Japanese Administrator on

• Java Virtual Machine—Version compatibility with Data Federator and

Data Integrator:

Windows so that metadata Reports and Impact Analysis correctly display object descriptions in Japanese. If you use a Japanese Administrator on UNIX, "junk" characters display if any of the object descriptions have Japanese characters.

To access Data Federator data sources with Data Integrator, you must modify your OpenAccess environment to use the Java Virtual Machine that ships with Data Integrator. Data Federator installs a component called OpenAccess ODBC to JDBC Bridge that requires these modifications.

Please make the following modifications (where Data Federator is installed in D:\Program Files\DataFederatorXIR2, Data Integrator is installed in D:\Program Files\DataIntegrator, and OpenAccess is installed in D:\Program Files\OaJdbcBridge):

1. In the OpenAccess configuration file (D:\Program

Files\OaJdbcBridge\bin\iwinnt\openrda.ini), the path for the JVM must be changed from:

JVM_DLL_NAME=D:\Program Files\DataFedera torXIR2\jre\bin\client\jvm.dll

JVM_DLL_NAME=D:\Program Files\DataIntegra tor\ext\Jre\Bin\client\jvm.dll

14 BusinessObjects Data Integrator Release Notes

2. Add the following OpenAccess jar files to the CLASSPATH

environment variable:

D:\ProgramFiles\OaJdbcBridge\jdbc\thindriver.jar

D:\ProgramFiles\OaJdbcBridge\oajava\oasql.jar

Unsupported products

Data Integrator no longer supports the following products as of this release.

• Microsoft Windows 2003 with no service packs

• Teradata V2R5.0

• BusinessObjects Data Quality 11.6

Data Integrator does not support the following products as of version 11.7.

• AIX 32-bit, SOLARIS 32-bit

• DataDirect ODBC Driver version 5.1

• Firstlogic Data Cleansing products including:

• International ACE versions 7.25c, 7.50c, 7.60, and 7.70

• DataRight IQ versions 7.10c, 7.50c, 7.60, and 7.70

• BusinessObjects Data Quality 11.5.1

• IBM Connector for the mainframe

• Citrix MetaFrame XP is not supported. To see the latest supported version,

go to the Supported Platforms documentation on the Business Objects Customer Support Web site at: http://support.businessobjects.com/docu

mentation

• Informix (as a repository)

• Context-sensitive Help for Data Integrator

• Data Mart Accelerator for Crystal Reports is no longer supported; however,

it is still available on the Business Objects developer community Web site at http://diamond.businessobjects.com/EIM/DataIntegrator.

Supported platforms and versions

Limitations update

The following limitations apply to Data Integrator version 11.7.3.

• Files read by an adapter must be encoded in either UTF-8 or the default

encoding of the default locale of the JVM running the adapter.

BusinessObjects Data Integrator Release Notes 15

Supported platforms and versions

Limitations update

• All Data Integrator features are available when you use an Attunity

Connector datastore except:

• Bulk loading

• Imported functions (imports metadata for tables only)

• Template tables (creating tables)

• The datetime data type supports up to 2 sub-seconds only

• Data Integrator cannot load timestamp data into a timestamp column

The Adapter SDK no longer supports Native SQL or Partial SQL.

•

Unsupported data type:

•

• Unsupported data type is only implemented for SQL Server, Oracle,

• Data Integrator can read, load, and invoke stored procedures involving

• Data Integrator might have a problem loading VARCHAR to a physical

in a table because Attunity truncates varchar data to 8 characters, which is not enough to correctly represent a timestamp value.

Teradata, ODBC, DB2, Sybase ASE, Sybase IQ, Oracle Applications, PeopleSoft, and Siebel.

unknown data types for SQL Server, Oracle, Teradata, ODBC, DB2, Informix, Sybase ASE, and Sybase IQ assuming these database servers can convert from VARCHAR to the native (unknown) data type and from the native (unknown) data type to VARCHAR.

CLOB column (for example, bulk loading or auto-correct load could fail).

Note: Use the VARCHAR column in the physical schema for loading.

• PeopleSoft 8 support is implemented for Oracle only.

Data Integrator jobs that ran against previous versions of PeopleSoft are not guaranteed to work with PeopleSoft 8. You must update the jobs to reflect metadata or schema differences between PeopleSoft 8 and previous versions.

• Stored procedure support is implemented for DB2, Oracle, Microsoft SQL

Server, Sybase ASE, Sybase IQ, and ODBC only.

• Teradata support is only implemented for Windows and Linux.

• On Teradata, the named pipe implementation for Teradata Parallel

Transporter is supported with Teradata Tools and Utilities version 8 or later compatible version. Teradata Tools and Utilities version 7.0 and 7.1 are not supported with named pipes.

• Bulk loading data to DB2 databases running on AS/400 or MVS systems

is not supported.

16 BusinessObjects Data Integrator Release Notes

Supported platforms and versions

Limitations update

• Data Integrator Management Console can be used on Microsoft Internet

Explorer version 6.0 SP1, 6.0 SP2, or 7.0 only. Earlier browser versions may not support all of the Administrator functionality.

• Data Integrator's View Data feature is not supported for SAP R/3 IDocs.

For SAP R/3 and PeopleSoft, the Table Profile tab and Column Profile tab options are not supported for hierarchies.

• Data Integrator now supports multibyte metadata for table names, column

names, file names, and file paths. The following table lists which sources support multibyte and single-byte metadata and which support single-byte only:

Multibyte and single-byte metadata supported*

BusinessObjects Enterprise Data Federator DB2 Informix MySQL ODBC Oracle Siebel Microsoft SQL Server Sybase ASE Sybase IQ Teradata XML

Single-byte metadata supported

Attunity connector for mainframe databases

Data Quality HP Neoview JD Edwards Netezza Oracle Applications PeopleTools SAP R/3 SAP BW Server

* Support for multibyte metadata is dependent on comparable support in the applications, databases, and technologies with which Data Integrator interoperates.

• Support for the Data Profiler is provided for sources as follows:

BusinessObjects Data Integrator Release Notes 17

Supported platforms and versions

Limitations update

Not supportedSupported

Attunity Connector for mainframe databases

DB2 Data Federator Flat file HP Neoview Informix Microsoft SQL Server MySQL Netezza ODBC Oracle Oracle Applications PeopleSoft SAP R/3 Siebel Sybase ASE Sybase IQ Teradata

COBOL copybooks Excel IDOC JDE Memory Datastore SAP BW XML

• Support for the LOOKUP_EXT function is provided for sources as follows:

18 BusinessObjects Data Integrator Release Notes

Supported platforms and versions

Limitations update

Not supportedSupported

DB2 Data Federator Flat file JDE HP Neoview Memory Datastore Microsoft SQL Server MySQL Netezza ODBC Oracle Oracle Applications PeopleSoft Siebel Sybase ASE Sybase IQ Teradata

COBOL copybook Excel IDOC SAP BW SAP R/3 XML

• DB2 Java Library limitations

All Web applications in the Management Console will not work with a DB2 repository under any of the following conditions:

• db2java library is incompatible with DB2 client. For example, DB2

Client is version 8.1 and db2java library is version 8.2, or

• db2java library is not generated for JDBC 2 driver, or

• the java class path variable is not properly set to the corresponding

java library, or

• the DB2 JDBC shared library (e.g. libdb2jdbc.so on AIX) is not

compatible with the Java JDBC 2 driver

Under these conditions, you might see the following behavior:

• The Administrator stops working or crashes after you configure the

DB2 repository. Find any error and warning messages related to the DB2 repository configuration in the log file.

BusinessObjects Data Integrator Release Notes 19

Supported platforms and versions

Limitations update

• When testing your DB2 connection from the Administrator, the following

In this release, the db2java JDBC 2 driver version of DB2 8.1 and DB2

8.2 are provided. On Windows, find the different version of these libraries LINK_DIR/ext/lib

• db2java.zip (by default) DB2 8.2 version

• db2java_8.1.zip DB2 8.1 version

By default, the db2java.zip of version 8.2 will be used. If you run with a DB2 8.1 Client, you must:

• replace with version 8.1 of db2java.zip

• make sure the compatible DB2 JDBC shared library is in the shared

• restart the Web server service on Windows or restart the job service

If you run with other DB2 Client versions, you must obtain the corresponding java library with the JDBC 2 driver. Please refer to IBM DB2 documentation for how to obtain the correct java library.

errors appear in the Administrator log:

• BODI-3016409: fail to connect to repository.

• BODI-3013014: didn't load DB2 database driver.

library path

on UNIX

• Informix native-driver support on UNIX requires a fix in the Informix Client

SDK for IBM bug number 167964. Please contact IBM for a patch that includes this bug fix.

Note: IBM is targeting inclusion of this bug fix in Client SDK version 2.90.

• Data Integrator supports files greater than 2 GB only in the following

cases:

• Reading and loading large data files of type delimited and positional.

• Generating large files for bulk loader staging files for subsequent bulk

loading by a native bulk loader utility (such as SQL Loader).

• Previewing data files from the file format page in Designer when the

large files are on a UNIX Job Server.

• Reading COBOL copybook data files.

All other files generated by Data Integrator, such as log files and configuration files, are limited to 2 GB.

• For the Microsoft Excel as a data source feature, the following limitations

apply:

20 BusinessObjects Data Integrator Release Notes

Supported platforms and versions

Limitations update

Concurrent access to the same Excel file might not work. For example,

•

View Data might not display if the file is currently open in Excel.

• Because an Excel column can contain mixed data types, some data

type conversions could produce unexpected results; for example, dates might convert to integers.

• Boolean formulas not supported.

• Workbooks with AutoFilter applied are not supported. Remove the

filter before importing the workbook.

• Workbooks with hidden rows and/or columns are not supported.

• Stored procedures for MySQL are not supported due to limitations in

the MySQL ODBC library.

• The Data Integrator Salesforce.com adapter can connect to the

Salesforce.com version 7 API; however, the new functionality added in API version 7 and above is not supported. The custom lookup field is the only exception (new in API version 7) and is supported in the Salesforce.com adapter. To ensure that your existing Salesforce.com adapter works properly, you must update the URL in the datastore so that it points to Salesforce.com 7.0 API, the new URL should be: https://www.salesforce.com/services/Soap/u/7.0.

BusinessObjects Data Integrator Release Notes 21

Supported platforms and versions

Limitations update

22 BusinessObjects Data Integrator Release Notes

Migration considerations

Logs in the Designer

Note: To use this version of Data Integrator, upgrade all existing Data

Integrator repositories to version 11.7.0.0.

This section lists and briefly describes all migration-specific behavior changes associated with this version of Data Integrator. To view migration-specific behavior changes among previous releases of Data Integrator, see the Data Integrator Migration Behavior Changes guide available from the Business Objects Customer Support documentation Web site (http://support.busines

sobjects.com/documentation/).

Logs in the Designer

In Data Integrator 11.7.3, you will only see the logs (trace, error, monitor) for jobs that started from the Designer, not for jobs started via other methods (command line, real-time, scheduled jobs, or Web services). To access these other log files, use the Administrator in the Data Integrator Management Console.

Data flow cache type

When upgrading your repository from versions earlier than 11.7 to an 11.7 repository using version 11.7.3.0, all of the data flows will have a default Cache type value of pageable. This is different from the behavior in 11.7.2.0, where the upgraded data flows have a default Cache type value of in-memory.

Pageable cache for memory-intensive data flows

As a result of multibyte metadata support, Data Integrator might consume more memory when processing and running jobs. If the memory consumption of some of your jobs were running near the 2-gigabyte virtual memory limit in a prior version, there is a chance that the same jobs could run out of virtual memory. If your jobs run out of memory, take the following actions:

Set the data flow Cache type value to pageable.

•

• Specify a pageable cache directory that:

24 BusinessObjects Data Integrator Release Notes

Contains enough disk space for your data. To estimate the amount of

•

space required for pageable cache, consider factors such as:

• Number of concurrently running jobs or data flows

• Amount of pageable cache required for each concurrent data flow

• Exists on a separate disk or file system from the Data Integrator system

and operating system (such as the C: drive on Windows or the root file system on UNIX).

Embedded data flows

In this version of Data Integrator, you cannot create embedded data flows which have both an input port and an output port. You can create a new embedded data flow only at the beginning or at the end of a data flow with at most one port, which can be either an input or an output port.

However, after upgrading to Data Integrator version 11.7.2 or later, embedded data flows created in previous versions will continue to run.

Migration considerations

Embedded data flows

Oracle repository upgrade

If you previously upgraded your repository to Data Integrator 11.7.0.0 and you open the Object State Report on the Central repository from the Administrator, you might see the error ORA04063 view ALVW_OBJ_CINOUT

has errors. This occurs if you had an Oracle central repository prior to

version 11.7.0.0 and you upgraded the central repository to 11.7.0.0.

Note: If you upgraded to 11.7.0.0 from a prior version of Data Integrator and

you are now upgrading to version 11.7.3 (this release), this issue might occur and you must follow the instructions below. Alternatively, if you upgraded from a version prior to 11.7.0.0 to 11.7.3 without upgrading to version 11.7.0.0, this issue will not occur because it has been fixed in 11.7.2.0.

To fix this error, you must manually drop and recreate the view

ALVW_OBJ_CINOUT using an Oracle SQL editor, such as SQLPlus.

Use the following SQL statements to perform the upgrade:

DROP VIEW ALVW_OBJ_CINOUT; CREATE VIEW ALVW_OBJ_CINOUT (OBJECT_TYPE, NAME, TYPE, NORMNAME,

VERSION, DATASTORE, OWNER,STATE, CHECKOUT_DT, CHECKOUT_REPO, CHECKIN_DT,

CHECKIN_REPO, LABEL, LABEL_DT,COMMENTS,SEC_USER,SEC_USER_COUT)

BusinessObjects Data Integrator Release Notes 25

Migration considerations

Solaris and AIX platforms

( select OBJECT_TYPE*1000+TYPE,NAME, N'AL_LANG' , NORMNAME,VER SION,DATASTORE, OWNER, STATE, CHECKOUT_DT, CHECKOUT_REPO, CHECKIN_DT, CHECKIN_REPO, LABEL, LABEL_DT,COMMENTS,SEC_USER ,SEC_USER_COUT

from AL_LANG L1 where NORMNAME NOT IN ( N'CD_DS_D0CAFAE2' , N'XML_TEMPLATE_FORMAT' , N'CD_JOB_D0CAFAE2' , N'CD_DF_D0CAFAE2'

, N'DI_JOB_AL_MACH_INFO' , N'DI_DF_AL_MACH_INFO' , N'DI_FF_AL_MACH_INFO' ) union select 20001, NAME,FUNC_TYPE ,NORMNAME, VERSION, DATASTORE, OWNER, STATE, CHECKOUT_DT, CHECKOUT_REPO, CHECKIN_DT, CHECKIN_REPO, LABEL, LABEL_DT,COMMENTS,SEC_USER ,SEC_USER_COUT

from AL_FUNCINFO F1 where FUNC_TYPE = N'User_Script_Function'

OR OWNER <> N'acta_owner' union select 30001, NAME, N'PROJECT' , NORMNAME, VERSION, N'' , N''

, STATE, CHECKOUT_DT, CHECKOUT_REPO, CHECKIN_DT, CHECKIN_REPO, LABEL, LABEL_DT,COMMENTS,SEC_USER ,SEC_USER_COUT

from AL_PROJECTS P1 union select 40001, NAME,TABLE_TYPE, NORMNAME, VERSION, DATASTORE, OWNER, STATE, CHECKOUT_DT, CHECKOUT_REPO, CHECKIN_DT, CHECKIN_REPO, LABEL, LABEL_DT,COMMENTS,SEC_USER ,SEC_USER_COUT

from AL_SCHEMA DS1 where DATASTORE <> N'CD_DS_d0cafae2' union select 50001, NAME, N'DOMAIN' , NORMNAME, VERSION, DATASTORE,

N'' , STATE, CHECKOUT_DT, CHECKOUT_REPO, CHECKIN_DT, CHECKIN_REPO, N'' ,to_date( N'01/01/1970' , N'MM/DD/YYYY' ), N'' ,SEC_USER ,SEC_USER_COUT from AL_DOMAIN_INFO D1

);

Solaris and AIX platforms

Data Integrator 11.7.3 on Solaris and AIX platforms is a 64-bit application and requires 64-bit versions of the middleware client software (such as Oracle and SAP) for effective connectivity. If you are upgrading to Data Integrator

11.7.3 from a previous version, you must also upgrade all associated middleware client software to the 64-bit version of that client. You must also update all library paths to ensure that Data Integrator uses the correct 64-bit library paths.

26 BusinessObjects Data Integrator Release Notes

Data quality

Data Integrator now integrates the BusinessObjects Data Quality XI (formerly known as Data Cleansing) application for your data quality needs, which replaces Firstlogic's RAPID technology.

Note the following changes to data cleansing in Data Integrator:

Depending on the Firstlogic products you owned, you previously had up

•

to three separate transforms that represented data quality functionality: Address_Enhancement, Match_Merge, and Name_Parsing.

Now, the data quality process takes place through a Data Quality Project. To upgrade existing data cleansing data flows in Data Integrator, replace each of the cleansing transforms with an imported Data Quality Project using the Designer.

You will need to identify all of the data flows that contain any data cleansing transforms and replace them with a new Data Quality Project that connects to a Data Quality blueprint or custom project.

Data Quality includes many example blueprints that are sample projects

•

that can serve as a starting point when creating your own customized projects. (You may need to modify the blueprints to work with your installation. Please see your Data Quality Documentation, including the Release Notes, for updated blueprint information.) If none of these blueprints work to your satisfaction, you can either save these blueprints as a project and edit them, or you can create a project from scratch.

Migration considerations

Data quality

You must use the Project Architect, Data Quality's graphical user interface,

•

to edit projects or create new ones. Business Objects strongly recommends that you do not attempt to manually edit the XML of a project or blueprint.

Each imported Data Quality project in Data Integrator represents a

•

reference to a project or blueprint on the data quality server. The Data Integrator Data Quality projects allow field mapping.

BusinessObjects Data Integrator Release Notes 27

Migration considerations

To migrate your data flow to use the new Data Quality transforms

1. Install Data Quality XI, configure the server, and make sure it is started

before you can use it in Data Integrator. Please refer to the Data Quality XI documentation for installation instructions.

2. In the Data Integrator Designer, create a new datastore of type Business

Objects Data Quality and connect to your Data Quality server.

3. Import the Data Quality projects that represent the data quality

transformations you want to use. Each project will appear as a Data Quality project in your datastore. For the most common data quality transformations, you can use existing blueprints (sample projects) in the Data Quality repository.

4. Replace each occurrence of the old data cleansing transforms in your

data flows with one of the imported Data Quality transforms. You will also need to reconnect the input and output schemas with the sources and targets used in the data flow.

When opening a data flow containing one of the old data cleansing transforms (address_enhancement, name_parsing, match_merge), you will still be able to see the old transforms in this release (although they are not available anymore in the object library). You can even open the properties and see the details for each transform.

When validating a data flow that uses one of the old data cleansing transforms, you will get an error such as:

[Custom Transform:Address_Enhancement] BODI-1116074: First Logic support is obsolete. Please use the new Data Quality feature.

It is not possible to execute a job that contains data flows using the old data cleansing transforms (you will get the same error).

Contact Business Objects Customer Support at http://support.businessob

jects.com/ if you need help migrating your data cleansing data flows to the

new Data Quality transforms.

Note: Data Integrator 11.7.3 allows you to reconcile metadata for Data

Quality datastores.

28 BusinessObjects Data Integrator Release Notes

Distributed data flows

After upgrading to this version of Data Integrator, existing jobs have the following default values and behaviors:

Job Distribution level: Job

•

All data flows within a job will be run on the same job server.

The default for Collect statistics for optimization and Collect statistics

•

for monitoring is cleared.

The default for Use collected statistics is selected.

•

Since no statistics are initially collected, Data Integrator will not initially use statistics.

Every data flow runs as a process (not as a sub data flow process).

•

New jobs and data flows you create using this version of Data Integrator have the following default values and behaviors:

Job Distribution level: Job

•

The Cache type for all data flows: pageable.

•

The default for Collect statistics for optimization and Collect statistics

•

for monitoring is cleared.

Migration considerations

Distributed data flows

The default for Use collected statistics is selected.

•

If you want Data Integrator to use statistics, you must collect statistics for optimization first.

Every data flow is run as a single process. To run a data flow as multiple

•

sub data flow processes, you must use the Data_Transfer transform or select the Run as a separate process option in transforms or functions.

All temporary cache files are created under LINK_DIR/log/PCache

•

directory. This option can be changed from the Server Manager.

XML Schema enhancement

Data Integrator 11.7 adds the new Include schema location option for XML target objects. This option is selected by default.

BusinessObjects Data Integrator Release Notes 29

Migration considerations

Password management

In the Designer option Tools > Options > Job Server > General, Data Integrator 11.5.2 provided the key XML_Namespace_No_SchemaLocation for the section AL_Engine. The default value, FALSE, indicates that the schema location is included. If you upgrade from 11.5.2 and had set

XML_Namespace_No_SchemaLocation to TRUE (indicates that the schema

location is NOT included), you must open the XML target in all data flows and clear the Include schema location option to keep the old behavior for your XML target objects.

Password management

All password fields are encrypted using two-fish algorithm starting in this

•

release of Data Integrator.

To simplify the process of updating new passwords for the repository

•

database, this version of Data Integrator introduces a new password file feature. If you have no requirement to change the password to the database hosting your repository, you may not need to use this optional feature.

If you must change the password (for example, security requirements stipulate that you must change your password every 90 days), then Business Objects recommends that you migrate your scheduled or external job command files to use this feature.

Migration requires that you regenerate every job command file to use the password file. After migrating, when you update the repository password, you need only regenerate the password file. If you do not migrate using the password file feature, then you must regenerate every job command file every time you change the associated password.

Web applications

The Data Integrator Administrator (formerly called the Web Administrator)

•

and metadata Reports interfaces have been combined into the new Management Console in Data Integrator 11.7. Now, you can start any Data Integrator Web application from the Management Console launch pad (home page). If you have created a bookmark or favorite that points to the previous Administrator URL, you must update the bookmark to point to http:// computername : port /diAdmin.

30 BusinessObjects Data Integrator Release Notes

If in a previous version of Data Integrator you generated WSDL for Web

•

service calls, you must regenerate the WSDL because the URL to the Administrator has been changed in Data Integrator 11.7.

Web services

Data Integrator is now using Xerces2 library. When upgrading to 11.7 or above and configuring the Web Services adapter to use the xsdPath parameter in the Web Service configuration file, delete the old Web Services adapter and create a new one. It is no longer necessary to configure the xsdPath parameter.

Migration considerations

Web services

BusinessObjects Data Integrator Release Notes 31

Migration considerations

Web services

32 BusinessObjects Data Integrator Release Notes

Deployment on UNIX

Please observe the following requirements for UNIX systems.

Install JDK version 1.4.2 as described in the vendor's documentation. (If

•

the Daylight Savings Time update affects you, install the JDK patch, version 1.4.2_13 or later, 1.4.x compatible version).

The AIX maintenance level must be at least 5200-05. For all versions of

•

AIX, the following file sets must be installed:

DescriptionStateLevelFile set

COMMITTED6.0.0.13xlC.aix50.rte

COMMITTED6.0.0.0xlC.rte

COMMITTED6.0.0.0xlC.msg.en_US.rte

To find these xlC file sets and their levels on your AIX system, use the command lslpp -l xlC*

Refer to the $LINK_DIR/log/DITC.log file for browser-related errors.

For Red Hat Linux 4, install the following patches:

•

glibc-2.3.4-2

•

libgcc-3.4.3-9.EL4

•

compat-libstdc++-296-2.96-132.7.2

•

compat-libstdc++-33-3.2.3-47.3

•

For Solaris 9, install the appropriate patch:

•

Install 111721-04 (for both 32-bit and 64-bit SPARC; see the vendor's

•

latest patch updates)

C Set ++ Runtime for AIX 5.0

C Set ++ Runtime

C Set ++ Runtime Messages, U.S.English

For Solaris 10 SPARC, install the following patch:

•

120470-01

•

34 BusinessObjects Data Integrator Release Notes

Adapter interfaces

JMS adapter interface

This section contains notes related to installation, configuration, and use of the adapter interfaces provided with this release.

For installation details, see the Data Integrator Getting Started Guide, JMS and Salesforce.com interface integration subsection.

JMS adapter interface

Find the full technical documentation for the Data Integrator Adapter for JMS in the same directory as your other Data Integrator documentation.

The JMS adapter is generic and can work with the JMS libraries of any

•

JMS provider.

This version of the JMS adapter has been tested using Weblogic JMS

•

libraries (from BEA Systems) as the JMS provider.

If you are running the JMS adapter with any other JMS provider, you

•

should include the location of the third-party jar files associated with the specific JMS provider.

On the Adapter Instance Configuration page, the classpath field contains

•

the list of Data Integrator-provided jar files. Append the location of the JMS jar files to the classpath field.

Salesforce.com adapter interface

Find the full technical documentation for the Data Integrator Salesforce.com Adapter Interface in the same directory as your other Data Integrator

documentation.

The Salesforce.com adapter Interface is compatible with Data Integrator version 11.6 and later.

36 BusinessObjects Data Integrator Release Notes

Documentation updates

New features in version 11.7.3

Please note the following updates to the Data Integrator 11.7.2 technical documentation. These changes will be incorporated into a future release of the manual set.

Note: The numbers are ADAPT system Problem Report tracking numbers.

New features in version 11.7.3

HP Neoview database support

This version supports the HP Neoview database as a source or target via a new datastore option in the Designer. Ensure the HP Neoview ODBC driver is installed and configured on the client computer (where the Data Integrator Designer and the Job Server are located).

For information on creating datastores, see the Data Integrator Designer Guide, Datastores section and the Data Integrator Reference Guide, Data Integrator Objects section, Datastore subsection.

For details on the ODBC options available for the HP Neoview datastore, see the Data Integrator Reference Guide, Data Integrator Objects section, Datastore subsection, ODBC subsection.

Inxight integration for reading unstructured text

Data Integrator version 11.7.3 provides support for extracting and transforming contents from unstructured text by calling the Inxight SDX (SmartDiscovery Extraction Server) via Web services. Two new functions have been added to Data Integrator that enable you to pass data to Inxight using the required base64 encoding.

The following procedure describes how to use Data Integrator with Inxight SDX Web services.

1. In the Designer, create an adapter datastore that connects to the Inxight

SDX Web service.

2. From the Inxight adapter datastore, import an operation, which becomes

the adapter Web service function.

38 BusinessObjects Data Integrator Release Notes

3. Open the adapter function and edit the schema as necessary, for example

4. In the data flow query, use the base64_encode function to encode the

5. Use another query transform to call the Inxight Web service function and

6. Use the base64_decode function to decode the returned result as plain

base64_encode

Returns the base64-encoded data in the engine locale character set.

Syntax

base64_encode(input data, 'UTF-8')

Documentation updates

New features in version 11.7.3

increase the column length of the base64 text column.

input text to base64.

pass the encoded text as input.

text again.

Return Value

varchar

Returns base64-encoded data. If the input data is NULL or the size is 0, Data Integrator returns NULL. Otherwise, it returns the base64-encoded data that conforms to RFC 2045.

Where

input data

UTF-8

The input data that needs to be encoded to base64. Does not support long or BLOB data types.

The code page of the input data. UTF-8 is required for Data Integrator version 11.7.3.

BusinessObjects Data Integrator Release Notes 39

Documentation updates

New features in version 11.7.3

Example:

You want to extract home address, city, state, and zip code information from email content using the Inxight SDX Web service. One of the requirements for the Inxight SDX Web service is to send the content in base64-encoded format and to specify the character set of the content. Inxight SDX Web service returns the extract type and name in base64-encoded format.

For example, create a data flow with a flat file source that contains the data. The column name of the content data is CONTENT.

Map the source in a query transform as follows. In the column mapping editor for the CONTENT column, specify:

base64_encode(CONTENT, 'UTF-8')

Map this query to another query where you invoke the Inxight SDX Web service and it gets a response in an NRDM schema called Extract Entities Response. The extract fields have values in base64 encoding.

Map to another query transform and decode the base-64-encoded data as in the following example:

base64_decode(extract_name, 'UTF-8')

base64_decode

Returns the source data after decoding the base64-encoded input.

Syntax

base64_decode(base64-encoded input, 'UTF-8')

Return Value

varchar

40 BusinessObjects Data Integrator Release Notes

Documentation updates

Getting Started Guide

Returns the source data after decoding the base64-encoded input. If the input is NULL or the size of the data is 0, Data Integrator returns NULL. Otherwise, it returns the base64-decoded data that conforms to RFC 2045.

Where

base64-encoded input

UTF-8

Job Server enhancements

Using multithreaded processing for incoming requests, each Data Integrator Job Server can now accommodate up to 50 Designer clients simultaneously with no compromise in response time. (To accommodate more than 50 Designers at a time, create more Job Servers.)

In addition, the Job Server now generates a Job Server log file for each day. You can retain the Job Server logs for a fixed number of days using a new setting on the Administrator Log retention period page.

Finally, each Designer client only displays the logs for jobs executed from that Designer, not from jobs executed using the Management Console, command line, or Web services.

Getting Started Guide

Citrix Presentation Server version 4.0

The documentation that describes Citrix was removed from the Data Integrator Getting Started Guide in a prior version. Data Integrator now supports Citrix

BusinessObjects Data Integrator Release Notes 41

Documentation updates

Designer Guide

Presentation Server version 4.0. For more information, see documentation for Citrix support on the Business Objects developer community Web site at

http://diamond.businessobjects.com/EIM/DataIntegrator.

AIX user resource limits

ADAPT00871860

In the section (chapter) Installing Data Integrator in UNIX Systems in the AIX user resource limits subsection, change the following line in the User resource limits table:

From:

To:

CommentsValueUser resource limit

At least 2GB2097151data (kbytes)

CommentsValueUser resource limit

(blank)unlimiteddata (kbytes)

Designer Guide

Changed-data capture for Microsoft SQL Server 2005

Data Integrator now supports changed-data capture (CDC) for Microsoft SQL Server 2005.

In chapter 19, Techniques for Capturing Changed Data, the section "Setting Up SQL Replication Server for CDC" was written for Microsoft SQL Server

2000.

The following procedure applies to Microsoft SQL Server 2005.

42 BusinessObjects Data Integrator Release Notes

Documentation updates

Designer Guide

To configure publications for Microsoft SQL Server 2005 CDC

1. Start the Microsoft SQL Server Management Studio.

2. Select the SQL Server, right-click Replication menu, then select New >

Publication. The New Publication Wizard opens.

3. In the New Publication Wizard click Next.

4. Select the database that you want to publish and click Next.

5. Under Publication type, select Transactional publication, and then click

Next to continue.

6. Click to select tables and columns to publish as articles. Then click to

open Article Properties.

7. Set the following to False: - Copy clustered index - Copy INSERT,

UPDATE and DELETE - Create schemas at subscriber

8. Set the "Action if name is in use" to "keep the existing table unchanged"

9. Set Update delivery format and Delete delivery format to XCALL <stored

procedure>. Click OK to save the article properties.

10. Configure Agent Security and specify the account connection setting.

Click Security Settings to set the Snapshot agent.

11. Configure the Agent Security account with system administration privileges

and click OK.

12. Enter the login password for the Log Reader Agent by clicking Security

Settings. Note that it has to be a login granting system administration privileges.

13. In the Log Reader Agent Security window, enter and confirm password

information.

14. Click to select Create the publication then click Finish to create a new

publication.

15. To Complete the Wizard, enter a Publication name and click Finish to

create your publication.

BusinessObjects Data Integrator Release Notes 43

Documentation updates

Designer Guide

Creating Microsoft Excel workbook file formats on UNIX platforms

Note: The following new section will be added to Chapter 6: File Formats

in the Data Integrator Designer Guide.

This section describes how to use a Microsoft Excel workbook as a source with a Job Server on a UNIX platform.

To create Microsoft Excel workbook file formats on Windows, see “Excel workbook format” in the Data Integrator Reference Guide.

To access the workbook, you must create and configure an adapter instance in the Administrator. The following procedure provides an overview of the configuration process. For details about creating adapters, see Chapter 10, “Adapters,” in the Data Integrator Management Console: Administrator Guide.

To create a Microsoft Excel workbook file format on UNIX

1. Using the Server Manager ($LINK_DIR/bin/svrcfg), ensure the UNIX Job

Server can support adapters. See “Configuring Job Servers and Access Servers” in the Data Integrator Getting Started Guide.

2. Ensure a repository associated with the Job Server has been added to

the Administrator. To add a repository to the Administrator, see “Adding repositories” in the Data Integrator Management Console: Administrator Guide.

3. In the Administrator, add an adapter to access Excel workbooks. See

“Adding and configuring adapter instances” in the Data Integrator Management Console: Administrator Guide.

You can only configure one Excel adapter per Job Server. Use the following options:

• On the Installed Adapters tab, select MSExcelAdapter.

• On the Adapter Configuration tab for the Adapter instance name,

type BOExcelAdapter (required and case sensitive).

You may leave all other options at their default values except when processing files larger than 1 MB. In that case, change the Additional

44 BusinessObjects Data Integrator Release Notes

Documentation updates

Reference Guide

Java Launcher Options value to -Xms64m -Xmx512 or -Xms128m

-Xmx1024m (the default is -Xms64m -Xmx256m). Note that Java memory management can prevent processing very large files (or many smaller files).

4. Start the adapter.

5. In the Designer on the Formats tab of the object library, create the file

format by importing the Excel workbook. For details, see “Excel format” in the Data Integrator Reference Guide.

Note:

• To import the workbook, it must be available on a Windows file system.

You can later change the location of the actual file to use for processing in the format source editor. See “Excel workbook source options” in the Data Integrator Reference Guide.

• To reimport or view data in the Designer, the file must be available on

Windows.

• Entries in the error log file might be represented numerically for the date

and time fields.

Additionally, Data Integrator writes the records with errors to the output (in Windows these records are ignored).

Reference Guide

The following updates apply to the Data Integrator Reference Guide.

Maximum number of loaders

ADAPT00667119

Data Integrator now supports a maximum number of loaders of 10.

The following sections describe the option Number of Loaders.

Chapter 2, Data Integrator Objects, Section "Target" in Table 2-30: Target

•

table options available in all datastores

Chapter 5, Transforms, section "Data_Transfer" in subsection "Target

•

table options"

BusinessObjects Data Integrator Release Notes 45

Documentation updates

Reference Guide

Replace the first paragraph in these descriptions of Number of Loaders with the following paragraph:

Loading with one loader is known as "single loader loading." Loading when the number of loaders is greater than one is known as "parallel loading." The default number of loaders is 1. The maximum number of loaders is 10. If you specify a number greater than this maximum value, Data Integrator uses 10 loaders.

Reserved words

ADAPT00621691

Reserved words should not be used as the user name when you create a Data Integrator repository.

In chapter 10, Reserved Words, the section "About Reserved Words" lists the reserved words that should not be used as names for design elements. Replace the first paragraph with the following paragraph:

The following words have special meanings in Data Integrator and therefore should not be used as names for work flows, data flows, transforms, or other design elements that you create. They should also not be used as user names when you create a Data Integrator repository. They are reserved with any combination of upper- and lower-case letters.

Microsoft Excel workbook format for UNIX

In chapter 2, Objects, "Descriptions of objects," the "Excel workbook format" entry, in the Notes section replace the fifth bulleted item with the following.

For workbook-specific (global) named ranges, Data Integrator would name

•

a range called range as range. However for worksheet-specific (local) named ranges, Data Integrator would name a range called range that belongs to the worksheet Sheet1 as range!Sheet1.

In UNIX, you must also include the worksheet name when defining a workbook-specific (global) named range.

46 BusinessObjects Data Integrator Release Notes

Documentation updates

Salesforce.com Adapter Interface Guide

When you create a new Custom table in Salesforce.com and attempt to fetch CDC data from that table from the Data Integrator Designer, if the CDC table does not specify a starting date, Data Integrator will throw an error stating:

Error reading from <custom table name>: <There was an unexpected

error. Salesforce.com message is startDate before or replica

tion enabled date>.

Therefore, the "Using the CDC table source default start date" section of the Salesforce.com Adapter Interface Guide should state the following:

When you do not specify a value for the start date:

Data Integrator uses the beginning of the Salesforce.com retention period

•

as the start date if a check-point is not available (during initial execution).

Data Integrator uses the check-point as the start date if a check-point is

•

available and occurs within the Salesforce.com retention period. If the check-point occurs before the retention period, Data Integrator uses the beginning of retention period as the start date.

Data Integrator may throw an error message. Business Objects

•

recommends that you specify a default start date for your CDC tables.

Please note the following correction to the Salesforce.com Adapter Interface Guide version 11.7.0.0. These changes will be incorporated into a future release of the manual set.

In Chapter 3, page 29, change the text under "Using the CDC table source default start date" as follows:

When you do not specify a value for the start date:

Data Integrator uses the beginning of the Salesforce.com retention period

•

as the start date if a check-point is not available (during initial execution). However, if a table is created within the Salesforce.com retention period and a check-point is not available, the execution will return an error message. Drill into the source object and enter a value for the CDC table source default start date. The value must be a date that occurs after the date the table was created to work around this problem.

BusinessObjects Data Integrator Release Notes 47

Documentation updates

Salesforce.com Adapter Interface Guide

48 BusinessObjects Data Integrator Release Notes

Resolved issues

Please refer to http://support.businessobjects.com/documentation for the latest version of these release notes, which includes resolved issues specific to version 11.7.3.

The numbers in the following table are sorted in ascending order and refer to the Business Objects ADAPT system Problem Report tracking number.

ADAPT00534412

ADAPT00613822

ADAPT00619275

ADAPT00619336

When creating a BusinessObjects Universe from Data Integrator using either Microsoft SQLServer or Sybase datastores, the table names are not qualified with the proper database name and owner name in the BusinessObjects Universe. This issue is resolved. Customers using Universe Builder must apply the BusinessObjects Enterprise CHF-13 patch level or later for this fix to take effect. Contact Customer Support for information on how to download CHF patches.

See also ADAPT00534412 in the CHF-13 documentation for additional information regarding this fix.

When a DTD schema contains a standard XML header, adapters could not import the schema. This issue has been resolved.

In this release, Data Integrator uses a new version of the Xerce2 library that, when using Data Integrator Web services, allows you to import a WSDL file that previously caused the following error: XML parser failed: Error <An exception occurred! Type:NetAccessorException.

This issue has been resolved.

When editing column and table descriptions in a datastore, the descriptions sometimes got lost. This issue has been addressed.

ADAPT00619350

50 BusinessObjects Data Integrator Release Notes

Data Integrator always works in binary collation for the Table Comparison transform. If the comparison table sort order is not binary, Data Integrator will generate inconsistent results between cached, row-by-row, and sorted comparison methods.

Resolved issues

ADAPT00619352

ADAPT00619355

ADAPT00619357

ADAPT00619387

ADAPT00619393

ADAPT00619403

ADAPT00619413

When using regular loading and bulk loading for Microsoft SQL Server targets, decimal data rounding was not the same with the two loading methods. This problem has been addressed.

In metadata Reports, the start timestamp for data flows displayed incorrectly. This problem has been fixed.

In a data flow with an XML source containing a join with a nested query, the WHERE clause and ORDER BY clause now compile properly without getting an Unknown type error.

The push-down capability of Informix sources has been improved. Data Integrator now pushes down more queries to the Informix database for evaluation, and you can expect better performance in those scenarios.

Metadata Reports was not showing column-mapping information. This is no longer an issue.

When importing Microsoft SQL Server tables from the Designer, locks would not release on SQL Server's tempdb database. This problem has been resolved.

When working with the Query transform in the Data Integrator Designer, outer join entries sometimes disappeared from the Query transform when modifying other parts of the Query transform. This problem has been fixed.

ADAPT00619414

ADAPT00619421

The Profiler can now profile Microsoft SQL Server database tables that use NT authentication.

When importing a SQLServer table that contained an index with the index name size greater than 64, the table import operation failed. This problem has been resolved.

BusinessObjects Data Integrator Release Notes 51

Resolved issues

ADAPT00619427

ADAPT00619438

ADAPT00619479

For Microsoft SQL Server datastores, if the table owner name is DBO and the CMS connection is also Microsoft SQL Server with a login user name of sa, the updated Universe will determine the new table schema to be new because it compares DBO to sa as the owner name.

Therefore, do not use Microsoft SQL Server as a CMS connection with sa as the login name.

Data Integrator generated incorrect outer join syntax when accessing DB2 on AS/400. This issue has been fixed.

Due to a known Informix issue, two phase commit (XA) transactions in IBM® Informix® Dynamic Server could cause Data Integrator to fail when loading the data to an IBM In formix Dynamic Server or profiling data with Informix. The typical Informix errors are:

Could not position within a table

and

Could not do a physical-order read to fetch next row

The reference number from the IBM site is Reference# 1172548

http://www-1.ibm.com/sup port/docview.wss?uid=swg21172548

ADAPT00619756

ADAPT00619781

ADAPT00619824

52 BusinessObjects Data Integrator Release Notes

An issue with post-load commands in the table loader getting corrupted has been fixed.

After upgrading Data Integrator from 11.0 to 11.5 or 11.7, metadata Reports failed with a Java error. This issue has been fixed.

When using the Profiler to profile a Microsoft SQL Server table containing columns of text data type, the profiling server now works properly.

Resolved issues

ADAPT00619835

ADAPT00619871

ADAPT00619893

ADAPT00619934

ADAPT00619969

When running metadata Integrator with a clustered CMS environment, CMS computer names were used when gathering metadata. This caused the CMS computers to appear as separate systems in Impact and Lineage reports. In this release, the metadata integrator gathers the name of the cluster. Because of this change in behavior, if you are using a clustered environment and have already run the metadata Integrator against the CMS system, you need to delete the data previously collected. Please contact Business Objects Customer Support for details on how to perform this task.

When using an Informix version 7.3 datastore and client software 2.90, the Informix datastore now correctly imports tables into Data Integrator.

If the NLS_LENGTH_SEMANTICS flag was set to char in a UTF-8 encoded repository, the profiler did not work properly for Oracle databases. This problem has been fixed.

The Data Integrator function is_valid_datetime() function returned true if the hour contained a value greater than 24. This problem has been fixed.

Data Integrator jobs sometimes exited abnormally on the Linux platform if there was a lookup_ext function call to an Oracle database with a datetime column in the condition list. This fix applies to all UNIX platforms.

ADAPT00620002

ADAPT00620025

ADAPT00620158

ADAPT00620250

On Windows XP operating systems, columns in the query editor now appear correctly highlighted.

In the Administrator, you can now sort schedules by job name.

The Difference Viewer no longer shows that a column's data type and nullable attribute change after changing the owner name of the table.

Data Integrator did not support any non-ASCII characters in which the eighth bit was being used in the metadata (for example the umlaut in German). This is no longer an issue.

BusinessObjects Data Integrator Release Notes 53

Resolved issues

ADAPT00620297

ADAPT00620308

ADAPT00620314

ADAPT00620336

ADAPT00620338

ADAPT00620345

ADAPT00620352

Fixed-width file formats were not being created correctly from existing flat files. This problem has been fixed.

When the Data Integrator-imported table schema contains a long column that does not match its associated data type in the database, it could cause an access violation error in the Data Integrator engine. This problem has been resolved.

When a real-time job terminated, two more instances of the jobs started. This problem has been fixed. Real-time jobs now start only once if terminated abnormally.

The documentation for Catch error types and groups was unclear. The Reference Guide and Designer Guide have been corrected to clarify that you specify exception groups, instead of individual error numbers, in a Try/Catch block.

The Designer could not import materialized views on an Oracle 10g database. This issue has been addressed.

When validating a large job, the Designer sometimes aborted because of depleted memory available on the system. This problem has been fixed.

When monitoring job execution, on the Monitor tab, the columns Row count and Elapsed Time were not sorted in ascending order. This problem has been fixed.

ADAPT00620375

ADAPT00620377

ADAPT00620665

54 BusinessObjects Data Integrator Release Notes

Data Integrator license keys sometimes did not work with dual Ethernet cards. This problem has been resolved.

When a DTD schema contained a standard XML header, adapters could not import the schema. This issue has been resolved.

When data type conversion was required in a Query mapping (for example when mapping a varchar column to an integer column), the Designer displayed warnings during validation, but an error occurred at runtime. This issue has been addressed.

Resolved issues

ADAPT00620689

ADAPT00620882

ADAPT00620958

ADAPT00621069

ADAPT00621143

The Data Integrator index() function was not pushed down to Oracle. The fix addresses this problem. Please note that there are slight differences between executing the index function within Oracle and executing the index function within Data Integrator. Check the instr() function documentation in the Oracle documentation for details.

Using the functions lpad and rpad sometimes lead to crashes such as an access violation. This problem has been fixed.

Jobs that were running did not get displayed in the Administrator if the log retention period had been reached. This problem has been fixed.

The following error was encountered when working with ABAP programs:

ABAP SYNTAX ERROR Without the addition "CLIENT

SPECIFIED", you cannot specify...

This problem has been resolved.

When using a variable of type datetime in the Reverse Pivot transform in the Pivot Axis, the following warning is no longer thrown: "Unexpected axis value <2003-08-15> found in transform". This issue is resolved in this release by providing variable support in the Reverse Pivot transform.

ADAPT00621184

ADAPT00621198

ADAPT00621229

ADAPT00621278

The passwords in some of the page of the Administrator are now encrypted properly. The text of the password cannot be viewed under any circumstances now.

When using multiple datastore configurations for SAP datastores, the folder names between the configurations are no longer switched.

WSDLs generated by Data Integrator can be imported into a .NET environment.

In this release, Data Integrator has been enhanced to generate DTDs that can handle multiple rows.

BusinessObjects Data Integrator Release Notes 55

Resolved issues

ADAPT00621334

ADAPT00621340

ADAPT00621996

ADAPT00622074

ADAPT00622121

ADAPT00622188

Data Integrator can now import an XML file as a large varchar column. Please note that when loading this data into a target database, most databases have a maximum size restriction for their character data columns. Note the restrictions for your target database and design the Data Integrator jobs accordingly.

When a job has multiple validation transforms, the labels that appear in the DI_ERRORCOLUMN target table are now correct.

When entering Japanese characters in the Case Transform labels, you will no longer get an "Internal Application Error" when saving the dataflow.

If you entered an invalid user name or password in the Administrator, the values remained in the fields. The password field now clears.

When using a variable in the Error file name field in the flat file editor, a runtime error occurred when running the job. This problem has been resolved.

LONG and TEXT data type columns do not get profiled.ADAPT00622137

It is now possible to enter a text delimiter in the flat file format. The format rules and limitations for the text delimiter are the same as the rules applicable to the field and column delimiters.

ADAPT00622190

ADAPT00625180

ADAPT00627739

56 BusinessObjects Data Integrator Release Notes

If load triggers were set up in a table loader, sometimes Data Integrator did not execute the regular load. This problem has been resolved.

Sometimes Data Integrator wrote incorrect data to the target loader overflow file if using the data type char. This issue has been solved.

The following error sometimes occurred when Data Integrator processed a complex nested COBOL copybook: Wrong sequence number for field ...

This problem has been addressed.

Resolved issues

ADAPT00629178

ADAPT00631055

ADAPT00632204

ADAPT00633094

ADAPT00633872

ADAPT00633885

In a Case transform, when entering a tab character in the expression field and saving the data flow, the text following the tab character disappeared. This issue has been fixed.

While loading data from a source database to a target database, Data Integrator generated the ORA-01461 error when using an Oracle database as repository. The NLS_CHARACTERSET of Oracle database is AL32UTF8. This problem has been fixed.

During the export of a datastore, you could not change the owner of a table or database stored procedure. This problem has been fixed.

If a SQL transform was used in a data flow and the transform did not define an output schema, the data flow was corrupted. This problem has been resolved.

The JMS interface is not supported with IBM MQ Series 6.0.ADAPT00633774

The Data Integrator job failed when it contained large COBOL copybook schemas. This issue was addressed by modifying the internal Data Integrator language generation to eliminate unnecessary fields and attributes.

Designer crashed when performing View data with a filter condition on file formats like COBOL copybook. The issued has been fixed.

ADAPT00633919

ADAPT00634484

ADAPT00635201

If a Cobol copybook contained a REDEFINES clause and it was used together with the OCCURS clause, Data Integrator interpreted the data incorrectly.

If a WSDL from a third party already contains xmlns:xsd=\"http://www.w3.org/2001/XMLSchema\", the Data Integrator Web Service could not import it. This problem was caused by Data Integrator automatically adding the string while getting the XSD for the message. This problem is fixed.

In the Data Integrator Designer, the trace logs are now sorted properly by date when clicking on the Date column header.

BusinessObjects Data Integrator Release Notes 57

Resolved issues

ADAPT00635698

ADAPT00635475

ADAPT00637736

ADAPT00638268

ADAPT00641650

ADAPT00642513

In the DB2 table loader, if the Use input key option is selected and if any generated UPDATE statements updated more than one row, the load failed. This problem has been fixed.

A new option has been added for the Sybase IQ bulk loader that allows you to specify the row delimiter in the data file generated by Data Integrator.

In a real-time job, using the same data flow more than once caused an incorrect SQL INSERT statement to be generated. This issue has been addressed.

When running Data Integrator with SAP R/3 version 3.x, the following error could occur:

BODI-1112339:Error:Cannot import the metadata

table <name=*-CCSS-KOSTL>. R/3CallReceive er ror <function Z_AW_TREE_IMPORT40: Call (PER FORM) to a non-existent routine.

This problem has been fixed.

When using the Ignore row marker option in the flat file format, data truncation could occur. This problem has been fixed.

When importing a database link created in Microsoft SQLServer 2005, the Designer did not recognize the database link as a valid link. This issue has been addressed.

ADAPT00642531

ADAPT00642556

58 BusinessObjects Data Integrator Release Notes

When a COBOL copybook contained REDEFINES sections, the data for those fields was not processed properly by Data Integrator. This issue has been resolved.

When a data flow or a work flow runs in an infinite loop, the Data Integrator engine process (al_engine) leaked memory. This problem has been fixed. Data Integrator also now submits a slightly different SQL statement to Microsoft SQL Server to improve performance when retrieving changed data from the SQL Server replication server distribution database.

Resolved issues

ADAPT00644740

ADAPT00645348

ADAPT00645757

ADAPT00646862

When defining a custom function in the Designer, the order of the input parameters was reversed in the function. This problem has been fixed.

In Auto Documentation, the Case Transform did not display correctly. This problem has been fixed.

The Repository Manager failed if you upgraded a repository from a computer with a Designer-only Data Integrator installation. This problem has been fixed.

For Microsoft SQLServer CDC readers, there is a new option in the source editor called Automatically delete rows after reading. This option lets you delete rows after reading them from the replication server to manage server volume. For example, if many rows collect on the replication server during a given retention period, performance could suffer.

If you select this option, Data Integrator deletes only the rows that have already been read by all Data Integrator CDC readers and other named subscriptions (and also observes the retention period setting). However, if there are additional anonymous subscribers, Data Integrator is unable to identify them and could delete rows regardless of whether they have been read by those subscribers.

A CDC reader subscription is not established until the first time a job with such a subscription is successfully executed. If in the CDC reader Automatically delete rows after reading option is selected, and this reader with this subscription has not been executed before in any other data flow, Data Integrator will assume that there is only one subscription and the data will be purged after reading. This behavior will stop after the job has been executed at least once. If that is not the desired behavior, try these workarounds:

Do not enable Automatically delete rows after

•

reading for the first run of the Data Integrator job to establish all subscriptions.

If a CDC table has more than one subscription, check

•

the Delete after reading option in one CDC source only and execute the corresponding job last.

BusinessObjects Data Integrator Release Notes 59

Resolved issues

ADAPT00647533

ADAPT00647923

ADAPT00648523

ADAPT00648631

ADAPT00652463

ADAPT00655407

The function get_domain_description() did not return a value when used with PeopleSoft 8.9 domains. This problem has been fixed.

The Performance Optimization Guide incorrectly stated that if a data flow was not valid or saved, Data Integrator alerted you when you tried to view optimized SQL for the data flow. The guide has been corrected to state "If a data flow is not valid, Data Integrator alerts you."

Fixed-width file formats were not generated correctly if the data file was located on a remote server. This problem has been fixed.

When logging off from Windows where the Job Server is running, the Data Integrator jobs launched against this Job Server will now continue to run.

The following error:

Unknown exception in rdr::open()

occurred when reading an XML file using the Designer. This problem has been fixed.

The combination of UTF8 engine code page and MAP_CDC transform no longer generates the following warning: invalid Value <> found for row operation column. Previously, this problem only appeared on UNIX platforms.

ADAPT00655691

ADAPT00655863

ADAPT00656102

60 BusinessObjects Data Integrator Release Notes

Data Integrator sometimes generated virtual memory errors while executing jobs containing Oracle sources with varchar columns. This issue has been addressed.

Data Integrator Web Services only supports the UTF-8 code page.

When reading a COBOL copybook that contains a REDEFINES clause, the data was not aligned properly when viewing the data from the Designer. This issue has been addressed.

Resolved issues

ADAPT00656803

ADAPT00657456

ADAPT00658163

ADAPT00658378

ADAPT00658811

ADAPT00659178

Dataflow now works correctly when a column in the Map Transform is designated as "optional" and the column is not mapped.

Validation conditions were not displayed in Auto Documentation. This issue has been addressed.

If there were a large number of logs in a Data Integrator instance, the Designer might have appeared to freeze when launched. If there were real-time services configured for this instance, the services could fail due to the delay. This issue has been resolved in this release.

Data Quality transforms were not supported for real-time jobs.

When creating a Data Integrator repository on IBM DB2, only the UTF-8 code page is supported.

When executing the Table Comparison transform in rowby-row mode with the History Preserving transform, Data Integrator added an extra current row to the output when encountering a certain type of input row. This situation occurred when the input of the Table Comparison contained rows where the primary key data and the compare column data were the same as the current row in the target. This issue has been addressed.

ADAPT00659451

In Auto Documentation, if a data flow contained a While loop, the While condition and the data flow did not display correctly. This issue has been addressed.

BusinessObjects Data Integrator Release Notes 61

Resolved issues

ADAPT00660212

ADAPT00661720

ADAPT00667007

ADAPT00668842

ADAPT00671056

A new utility FixupDIMapText is included in this release to allow you to clean up a repository that could have corrupted mappings in the data flows. If you have noticed that when you change mapping values that the changes are not reflected after saving the data flows, you need to run this utility. The usage of the utility is as follows:

fixupUIMapText.exe -U<user> -P<passwd> S<server> -N<database_type> -Q<database_name>

-g<Windows Authentication (Microsoft SQL

Server only)

Valid database types are: <Informix>, <DB2>, <Ora

cle>, <Sybase>, <Microsoft_SQL_Server>.

The Salesforce.com adapter occasionally encountered the Java error OutofMemory. This problem has been fixed.

When using the bulk loader option in a Netezza target table, Data Integrator does not truncate spaces to null for character columns.

When performing Impact and Lineage analysis with Business Objects 6.5, the Universe impact does not contain WEBI document information.

The metadata Integrator generated an error when the configuration instance name was set to something other than the default. This problem has been fixed.

ADAPT00672346

ADAPT00673438

ADAPT00673927

ADAPT00674212

62 BusinessObjects Data Integrator Release Notes

When using the smtp_to function to send e-mail, the send time and receive time were not synchronized. This problem has been fixed.

Data Quality Project descriptions are not imported into Data Integrator Object Library.

In some cases, Data Integrator did not release connections to the Job Server and depleted system resources. This problem has been resolved.

When using the View Data functionality with a SQL Server 2005 CDC table, the following error no longer occurs if the last column in the table is a numeric type: "cannot convert data <0> into type <DECIMAL> Context".

Resolved issues

ADAPT00674474

ADAPT00675249

ADAPT00674212

ADAPT00678466

ADAPT00681355

ADAPT00682563

ADAPT00682686

When editing Job Server information using the Server Manager, all of the Job Servers in the Server Groups appear.

On Linux, completed exec() function processes in a job stayed in a defunct state. This issue has been fixed.

When using the View Data functionality with a SQL Server 2005 CDC table, the following error no longer occurs if the last column in the table is a numeric type: Cannot convert data <0> into type <DECIMAL> Context Column.

When using the Oracle char size semantic to define a column data type as char or varchar, Data Integrator no longer invokes an Oracle error at runtime when performing the binding process for the column.

In an Salesforce.com adapter, a CDC table can handle changes that are equal to or greater than 2000 records.

Importing an existing Data Quality project does not invalidate the data flow.

When inserting a string of blank characters to an Oracle column of data type char or nchar, blanks are inserted.

ADAPT00683039

ADAPT00683167

ADAPT00685077

ADAPT00685406

When a Job creates hundreds of DB2 connections to evaluate lookup function calls, Data Integrator no longer unnecessarily replicates them.

When creating sources or targets from tables from a datastore that uses persistent cache, view data operation is now available.

Changes to the Data Quality datastore parameters were not reflected in the properties page of the Data Quality transform. This issue has been resolved.

When exporting a data flow containing lookup_ext functions to a flat file, validation of the data flow sometimes failed when importing the data flow back into the Designer. This problem has been fixed.

BusinessObjects Data Integrator Release Notes 63

Resolved issues

ADAPT00685442

ADAPT00686674

ADAPT00687185

ADAPT00687873

ADAPT00688040

ADAPT00688861

ADAPT00688976

If you use a non-US time zone with the Salesforce.com adapter, you will no longer encounter a problem when inserting/updating records.

In the Windows environment, Data Integrator Web services no longer shut down when users log off.

A Java error is not encountered when using the print functionality in Auto Documentation.

When performing View Data on a column of type CHAR, the data is no longer double-quoted to the full length of the column. In this release, the double quotes are not displayed when viewing the CHAR data.

When you view monitor logs in the Administrator, the memory usage of the job server process no longer increases.

If the Job Server dies unexpectedly, all the adapters still keep on running. This problem is resolved. The adapters now exits gracefully in this situation.

If the Web Services adapter does not work due to an incorrect keystore path, specify the correct keystore path then restart the Tomcat Web server.

ADAPT00689618

ADAPT00689651

ADAPT00691549

ADAPT00691896

ADAPT00692414

64 BusinessObjects Data Integrator Release Notes

You must have the same schema name as the MYSQL datastore user name to make data transfer transforms work with the automatic option.

When using a function inside the decode function, the parameter list of the nested function is processed properly. The ordering of the parameter is not reversed.

If Data Integrator functions are used in the WHERE clause in the Query transform, outer joins now evaluate properly.

When calling a SQLServer 2000 stored procedure from Data Integrator, the varchar parameter value is now passed correctly.

When viewing complex jobs in Auto Documentation, the print functionality no longer fails.

Resolved issues

ADAPT00694262

ADAPT00695487

ADAPT00695492

ADAPT00695837

ADAPT00696321

ADAPT00696333

ADAPT00696434

Auto Documentation for Data Quality transform now shows the description for the Data Integrator input fields. The unused Data Quality output fields are not displayed without showing the data type.

Both WIndows and UNIX server configurations now allow central repositories to be associated to a Job Server.

A warning message is not thrown when you attempt to login to the central repository.

When running a real-time job that contains an XML message format as a source as well as an XML format in the transform, you might encounter an error with WSDL generation if the real-job is published via the Administrator. This problem has been addressed in this release. Open the job in the Designer and re-save it for this fix to take effect.

In German regional settings, Teradata bulkloading loads the correct decimal data to target.

Data Integrator correctly reads data from a COBOL copybook that is in a German regional setting operating system.

When using the validation transform on Solaris, the job no longer exits incorrectly.

ADAPT00700221

ADAPT00700254

ADAPT00700991

ADAPT00701078

You can now use a lookup_ext function as part of the input schema to the pivot transform as a non-pivot column.

Table comparison now provides the correct results when the input row and/or target table row have null values.

This release of Data Integrator includes fixes to address the new U.S. Daylight Savings Time rule change. For this fix to take effect, apply a DST patch from the operating system vendor to the computers that host Data Integrator components.

Data Integrator correctly imports XMI files using metadata Exchange functionality.

BusinessObjects Data Integrator Release Notes 65

Resolved issues

ADAPT00702076

ADAPT00702312

ADAPT00702353

ADAPT00702817

ADAPT00704503

ADAPT00706671

Data Integrator is now compatible with Salesforce.com version 7.0 API. With this API upgrade, Data Integrator can now support custom lookup functions. Any other new functionality specific to Salesforce.com 7.0 is not supported. To ensure that your existing Salesforce.com adapter works properly, update the URL in the datastore so that it points to Salesforce.com 7.0 API. The new URL is https://www.salesforce.com/services/Soap/u/7.0.

Data Integrator can now use the SQL Server to load tables with an owner name that is different from the one used in the datastore.

The metadata Integrator does not fail when collecting Universes stored in a subfolder.

When using Auto Documentation to generate a PDF file, the java error with NullPointer exception no longer occurs.

When you specify the row delimiter '/10' (line feed character) in a flat file, Data Integrator no longer interprets this character on the Windows and UNIX platforms differently.

When reading a flat file containing varchar data that does not match the schema of the file format, the Job no longer exits abnormally.

ADAPT00709026

ADAPT00709500

ADAPT00711853

ADAPT00712157

66 BusinessObjects Data Integrator Release Notes

When comparing a local repository object to a central repository object in a multiuser environment, the Designer no longer exits abnormally if you are using custom SQL in the column mapping.

When Data Integrator reads containing Occurs Depending On Items with multiple counters for a single record, Data Integrator recognizes only the first counter item. In this release, Data Integrator is able to handle multiple Occur Depending On clauses when reading COBOL Copybooks.

In chapter 4 of the Reference Guide, the information for varchar data type is now correct.

The "view from end" functionality in the Web Administrator is restored.

Resolved issues

ADAPT00712275

ADAPT00713935

ADAPT00716264

ADAPT00716293

ADAPT00716886

ADAPT00717175

ADAPT00717659

For a Microsoft SQL Server repository created In Data Integrator version 11.0, upgrading to 11.6.0 or 11.7.0.0 would fail if the table AL_CMS_REPORTS was not empty. This problem has been fixed.

Oracle repository upgrade now works with SQL when upgrading from 11.5.1.5 to 11.7.0.0.

MySQL version 5.0 can now be used as a repository for Data Integrator. Creation of repositories is now supported.

In the Management Console under Impact and Lineage, Universes and reports belonging to a user now appear automatically in the "Objects to analyze" window.

When using the Oracle CDC subscriber user as the login user of the Oracle CDC datastore you can now access CDC data. The owner name for the internal views are correctly generated.

Data Integrator Designer issues cursors with hold on DB2 system catalog tables. This issue has been resolved by submitting a commit statement after read access.

metadata Integrator does not fail when collecting Desktop Intelligence report information.

ADAPT00718127

ADAPT00720453

ADAPT00720989

When there was a mismatch between the Sybase server version and the Sybase client version (such as 12.5.3 for Sybase ASE server and 12.0.0 for Sybase client), the Data Integrator job could fail. This problem was from incorrect error message processing sent from Sybase to Data Integrator. This issue has been resolved.

When using a file in the lookup_ext function, Data Integrator now cleans up the file handle properly so files are correctly locked.

When viewing data in the Designer and it is running in a Japanese locale, error messages from the View Data module were not legible. This issue has been resolved.

BusinessObjects Data Integrator Release Notes 67

Resolved issues

ADAPT00721616

ADAPT00722498

ADAPT00723254

ADAPT00724995

ADAPT00726764

ADAPT00729550

Repository upgrades work when starting from repository version 6.1 to 11.x if the repository contains a datastore with ms874 code page.

The to_date function was not pushed down to the Oracle database. This issue could happen if in the format string of the to_date function, you had specified FF instead of ff to indicate fractional second. This issue has been resolved.

After upgrading an Oracle pre-11.7.0 central repository to

11.7.0, you can now access the Object State Report via the Data Integrator Management Console.

On Windows 2003, the Data Integrator engine no longer fails when launched from the Designer and the Management Console.

When using the table comparison transform with duplicate input rows and sorted input options together, UPDATE rows are now detected.

The following warning message no longer appears when the Rows per commit option has been set to 1:

Warning: Rows per commit for table loader has

been reset to 1 because table <T_LONG_TEST> contains a LONG column

This issue has been resolved.

ADAPT00729957

ADAPT00730595

ADAPT00730703

ADAPT00731785

68 BusinessObjects Data Integrator Release Notes

Web servicesnow handle outbound calls properly if the message name is the same as the referenced element name.

When Data Integrator is running in UTF-8 mode, the Lookup function now returns the correct result when comparing a decimal number with a database that is also in UTF-8 mode.

When Japanese characters are used in function raise_exception(), the messages displayed when the function is called are now correct.

The Data Integrator Profiler now correctly detects distinct rows if there is a large number of distinct rows in the table.

Resolved issues

ADAPT00733805

ADAPT00736500

ADAPT00738912

ADAPT00738936

ADAPT00740642

If the input to the Data Quality transform is not mapped and the output of the transform is mapped, Data Integrator freezes when the executing the job. Ensure the transform is properly mapped before running the job.

Under the following circumstances, the Profiler could stay in the pending state indefinitely:

The repository is not connected to a Job Server.

•

The associated Job Server is not running.

•

When performing an outer join with the "IN" keyword in the WHERE clause, Data Integrator now correctly generates the correct SQL to push-down to the database.

There is a known issue with Data Quality projects that contain periods in the field names. Although the projects can be imported from the Data Quality server, Data Integrator cannot correctly map these fields. The project can be used in a data flow as long as the fields containing periods are not used. A workaround for this problem is to rename any Data Quality project fields to use another character instead of the period. Business Objects suggests using the underscore character (_).

When using the Sybase IQ bulk loader with a target table schema that does not have the same number of columns as the source table schema, Data Integrator jobs could fail with an Access Violation error. This problem has been resolved.

ADAPT00741303

ADAPT00741570

The Message Client sample (client_sample) program in the AccessServer directory has not been updated for this release. Please check the Business Objects Knowledge Base for any updates.

When an object such as work flow was renamed in the local repository, Data Integrator threw an error indicating that the dependent objects did not exist. This issue has been resolved.

BusinessObjects Data Integrator Release Notes 69

Resolved issues

ADAPT00742231

ADAPT00742242

ADAPT00743816

ADAPT00743879

ADAPT00746019

When the order of the columns in the query transform is different from the input columns of a Microsoft SQL Server bulk loader, the load operation now works correctly.

If you have the bulk loader specified in the target, you change the order of columns in the query, and you execute the job without trace enabled, the job execution no longer fails.

When the SQL SERVER NCHAR column is used in the primary key and compared column, the correct SELECT statement is selected for table comparison transform rowby-row option.

Data Integrator might not behave correctly when using the table comparison transform with the row-by-row option. This could happen if the compared table was in an Oracle UTF8 database and an NCHAR column was used as a primary key or compared column. In that case, Data Integrator generated an incorrect SQL SELECT statement. This issue has been resolved.

When publishing a job as a Web service with the Enable job attributes option, you might have encountered the following error when importing the WSDL:

DIBODI-1112015: adapter metadata import failed:

error parsing

This issue has been resolved.

ADAPT00749771

ADAPT00750372

70 BusinessObjects Data Integrator Release Notes

When the escape character in a flat file appears as part of the data, Data Integrator now escapes the character properly.

A new flag has been introduced to provide read-uncommitted capability for Informix datastores. To use read uncommitted data for Informix datastores, from the Designer, choose Tools > Options > Job Job Server > General. Enter

al_engine in the section field, InformixReadUncommitted

for the key, and TRUE for the value.

Resolved issues

ADAPT00751136

ADAPT00751141

ADAPT00751731

ADAPT00752299

ADAPT00755702

ADAPT00755958

When exporting a data flow to another repository and the owner of the datastore changed in the process, the new owner name was not reflected in the lookup function in the exported data flow. This issue has been resolved.

Creatiing embedded data flows failed when the content contained a validation transform. This issue has been resolved.

When loading to a flat file field defined as a DECIMAL(12,0) or DECIMAL(15,0) with formatting #,##0.0 enabled, Data Integrator truncated the digits to the left of the decimal point. This issue has been resolved.

Data Integrator generated the incorrect PL/SQL for an Oracle loader if the column name was quoted. This problem has been resolved.

In the Validation transform editor, dragging a column into the editor from a datastore and adding a condition invoked the BODI-1112468 error. This issue has been resolved.

When using Auto Documentation to view a column mapping that contained a Key Generation function, the mapping text did not display properly in the HTML format. This issue has been resolved.

ADAPT00755960

ADAPT00761033

ADAPT00761605

ADAPT00763665

When launching multiple jobs at the same time, some jobs might not run and produced an error Could not find the

GUID. This issue has been resolved.

When using SFDC Boolean data in the WHERE clause of a Data Integrator query transform, Data Integrator did not interpret the Boolean values properly. This issue has been resolved in this release.

When exporting a Cobol copybook, the atl file did not show the output schema. The problem has been solved.

When Data Integrator pushed down a query to DB2 that contained the to_char function, the DB2 syntax generated by Data Integrator was incorrect. This issue has been resolved.

BusinessObjects Data Integrator Release Notes 71

Resolved issues

ADAPT00764861

ADAPT00765394

ADAPT00765441

ADAPT00765502

ADAPT00767564

ADAPT00770720

In this release, Data Integrator supports Sybase IQ 12.7 as a source and target. This support is available on all platforms. Note that in 11.7.2.2, support for this version of Sybase IQ was available on Windows only.

In this release, Data Integrator supports SAP BI 7.0 (formerly known as SAP BW). This connectivity is available on all platforms. Note that previous releases of Data Integrator already provided supported to some of the platforms.

In this release, Oracle E-Business Suite Release 12 support has been added on all platforms. Note that in 11.7.2.3 this support was already available on Windows and Linux.

When turning on the trace for the loader from the Designer, the row level data did not display. This issue has been addressed.

When using Metadata Exchange functionality to export metadata to XML ERWin 4.x format, the NULL option values of the columns were reversed. This issue has been resolved.

Data Integrator did not allow spaces between column names and the ! character. This issue has been resolved in this release.

ADAPT00770958

72 BusinessObjects Data Integrator Release Notes

After modifying the input or output parameter of a stored procedure, Data Integrator did not retain the change after restarting the Designer. This issue has been addressed.

Resolved issues

ADAPT00773839

When using table sources from a DB2 datastore or ODBC datastore with a WHERE clause in the Query transform, the WHERE clause for that source might not be pushed down to the database. This situation could happen if the WHERE clause contains a concatenation operation. The WHERE clause will be evaluated inside Data Integrator. Note that in Data Integrator, the evaluation of concatenation operator with NULL operands might be different from the DB2 or the ODBC data source evaluation rule. In Data Integrator, a value concatenated with NULL value results in the value itself. This might not be the case with DB2 or generic ODBC sources.

This release includes a flag to allow the WHERE clause in the Query transform to be pushed down to DB2 or ODBC when possible. Note that depending on the data flow construction, Data Integrator may or may not push down the WHERE clause. If this flag is turned on, the concatenation of NULL values in the WHERE clause may have different behavior in different parts of the data flow depending on whether Data Integrator is evaluating the expression. In order to avoid the discrepancy in this evaluation, Business Objects recommends that you use this flag only when there is no concatenation operator used with DB2 sources and ODBC sources in your repository.

To enable the WHERE clause containing a concatenation operator for DB2 and ODBC sources to be pushed down, in the Designer choose Tools > Options > Job Job Server > General. Enter al_engine in the section field, enter

PushdownODBCDB2Concat for the key, and enter TRUE for

the value.

ADAPT00774266

ADAPT00774804

When saving a data flow in the Designer with the Calculate column mapping option enabled, the memory consumption of the Designer could increase to the point that all available virtual memory was consumed and the Designer exited abnormally. This problem has been resolved by eliminating unnecessary memory copy operations.

In this release, Data Integrator is cerfieid to work with Microsoft Internet Explorer version 7. This certification is available on Windows only.

BusinessObjects Data Integrator Release Notes 73

Resolved issues

ADAPT00774812

ADAPT00774892

ADAPT00774914

ADAPT00774940

ADAPT00777105

ADAPT00777107

In this release, Data Integrator supports Sybase ASE version

12.5 on Solaris 64-bit and AIX 64-bit platforms as source, target, and repository. Note that on other platforms Sybase ASE version 12.5 was already supported in prior versions of Data Integrator.

In this release, Data Integrator supports DB2 version 9.1 as a source, target, and repository. This support is available on all platforms. Note that in 11.7.2.3, this certification was already available on Windows and Linux platforms.

In this release, Data Integrator has been certified to be compatible with BusinessObjects Enterprise version XI R2 SP3.

In this release, Data Integrator has been certified on Microsoft Windows Server 2003 R2 SP2.

When using Metadata Exchange to generate XML metadata in ERWin 4.x format, the size of the CHAR data type columns were not showing the actual length of the column definition. This issue has been resolved.

In the XML file reader, you could not specify a network path in the file name. This issue has been resolved.

ADAPT00777109

ADAPT00779045

ADAPT00784324

74 BusinessObjects Data Integrator Release Notes

When running a Data Integrator job that contained nested schemas, the job sometimes consumed all of the available memory and stopped execution abnormally. This issue has been resolved.

When compacting a Data Integrator repository, you could get a Unique constraint error from the database. This issue has been resolved.

Using Auto Documentation to display a file format that contained non-printable characters in the row marker invoked the java.lang.NullPointerException error. This issue has been resolved by printing a space in place of the nonprintable character.

Resolved issues

ADAPT00786004

ADAPT00788762

ADAPT00793560

ADAPT00796163

ADAPT00797003

In this release, Data Integrator supports Teradata V2R6.2 via TTU8.2 clients. This support is provided on the Windows and Linux platforms only.

When using aliases in Data Integrator datastores and creating a Business Objects Universe with Data Integrator metadata, information on Data Integrator objects did not display properly in the Business Objects Designer. This issue has been resolved.

When using the float data type, Data Integrator rounded the data at approximately one-tenth precision. This problem has been addressed in this release. Note that the float data type is a imprecise data type, which means rounding the data occurs if the value has a large precision. Business Objects suggest using a different data type, such as decimal, if your application requires a consistent numeric result.

As of this release, Data Integrator Metadata Integrator can retrieve metadata from CMS systems in SSL mode.

When loading into an Oracle table that contained multiple CLOB columns, if the data being loaded to the column contained data larger than 4,000 bytes and if another CLOB column contained NULL, Data Integrator invoked the error

String literal too long. This issue has been resolved.

ADAPT00799347

ADAPT00802253

When using the Table Comparison transform in sorted input mode, if the input data set is considerably smaller than the compare table data set, the transform could take extra time to complete after the input rows had been exhausted. This issue has been resolved.

When reading level 3 of an XML file source, an Access

Violation error occurs. This issue has been resolved.

BusinessObjects Data Integrator Release Notes 75

Resolved issues

ADAPT00804492

ADAPT00804532

ADAPT00806765

ADAPT00816263

ADAPT00816382

The Display Optimized SQL functionality sometimes caused the Designer to crash. This bug is due to an implicit data type conversion in the WHERE clause in the Query transform when the conversion applied to a non-column reference. The ordering of the WHERE clause could also be a factor. Running a job with this scenario icould terminate due to error <170101>. This issue has been resolved in this release.

In a data flow that joins multiple R/3 data flows, the job could fail. This problem has been resolved.

When using the Salesforce.com adapter, if using the DI_PICKLIST_VALUE virtual table and the size of the pick list is 100 or greater, the job failed with the error message

Error reading from <AIView1>: <Unknown object type. Salesforce.com message is Object type cannot be null>. This problem has been resolved, and

now Data Integrator can handle batches of any size.

The database reader leaked memory if the reader contained varchar columns and the native data type of the column was non-varchar (such as nvarchar, nchar, graphic, etc.). This issue has been addressed.

Data Integrator did not recognize the computational field value correctly for some Cobol compilers. This issue has been fixed.

ADAPT00818697

ADAPT00819587

76 BusinessObjects Data Integrator Release Notes

When running Data Integrator jobs on the Solaris platform, the parent engine froze intermittently in some environments. This issue was caused by a problem in the process forking mechanism from the operating system. This issue has been resolved by using a different process forking mechanism on the Solaris platform.

Job Server enhancements on page 41.

Resolved issues

ADAPT00820475

When calling an SAP function, the AL_RFC_RETCODE parameter did not return correctly if the Data Integrator engine was running in UTF-8 mode. This issue has been resolved.

BusinessObjects Data Integrator Release Notes 77

Resolved issues

ADAPT00820896

As of Data Integrator 11.7.2.3 there are two changes in behavior from prior versions:

• If you upgrade a repository from a version prior to

11.7.0.0, all of the data flows will have a cache type of pageable. Previously the cache type was in-memory.

• When running jobs in Collect statistics for optimization

mode, the cache type for the data flows was always pageable. Previously, Data Integrator honored the cache type setting at the individual data flow level. This change in behavior was made to ensure that the job will run to completion regardless of the amount of data being processed.

To ensure that Data Integrator uses the optimal cache type when running the jobs in a production environment, Business Objects strongly suggests that you run jobs with the option Collect statistics for optimization set at least once with a data volume similar to the production environment. Once Data Integrator has the statistics on the job memory usage, it will use the collected information to determine the best cache type to use (pageable or in-memory).

To collect statistics,you can enable theCollect statistics for optimization option when running individual jobs. Alternatively, you can enable a global flag to always collect statistics for optimization for all jobs regardless of the job level setting, which is a new feature in 11.7.2.3. After the statistics have been collected for all jobs, disabling the same global flag ensures that the job level settings are again honored.

To enable this global flag, in the Designer, select Tools > Options > Job Job Server > General. Enter engine in the section field, enter collect_statistics_for_optimiza

tion for the key, and enter TRUE for the value.

To disable the global flag in order to honor the individual job level settings, set the value of collect_statis

tics_for_optimization to FALSE.

For proper maintenance of your jobs, Business Objects recommends that you collect statistics for your jobs whenever there is a significant change in data volume to ensure that Data Integrator continues to make the right cache decision when running jobs.

78 BusinessObjects Data Integrator Release Notes

Resolved issues

ADAPT00823175

ADAPT00823263

ADAPT00823272

ADAPT00823545

ADAPT00823899

ADAPT00823904

When dragging Data Quality transforms into Data Integrator Designer and the Job Server is on the HP-UX platform, you might get the error BODI-1112436. This issue has been resolved.

The Job Server could not start when the Data Integrator was installed with a Spanish code page and if the Windows regional setting was also Spanish. This issue has been addressed.

The sql() function call used with DB2 datastore resulted in a small memory leak. Depending on where the function call was used, the memory leak could compound and cause the memory usage of the job to increase. This issue has been resolved.

When Japanese characters were used in an Annotation and the data flow was exported and imported in XML mode, the Japanese characters did not display properly. This issue has been resolved.

Creating a Data Integrator repository on a MySQL database sometimes resulted in a syntax error. This issue has been resolved.

When using table names containing the $ character in a nested lookup_ext function, Data Integrator could not validate the function after restarting the Designer. This issue has been addressed.

ADAPT00823958

Under certain circumstances, the repository displayed twice in the Server Manager. This issue has been resolved.

BusinessObjects Data Integrator Release Notes 79

Resolved issues

ADAPT00824010

ADAPT00829000

ADAPT00829117

ADAPT00829859

This release of Data Integrator supports Netezza ODBC driver 3.1.4 on all platforms. Note that this feature was available in Windows and Linux platforms as of 11.7.2.3.

Note that due to a change in Netezza's ODBC driver, by default Netezza's internal log files for bulk loading will be written to Data Integrator's root directory. You may delete these temporary log files upon job completion. If you want to store the temporary Netezza log files in a different location for easier management, specify the desired log file directory location in the Netezza ODBC driver.

In addition, grant the CREATE EXTERNAL TABLE privilege to the Netezza user, which is necessary to use Netezza table loaders.

The Convert R/3 null to null option in the reader did not treat NULLs or spaces properly. This issue has been resolved.

The error Job server not part of server group displayed if you changed the repository password through the Management Console. This issue has been resolved.

The loading performance to MySQL tables has been improved in this release. The insert/update/delete statements of the MySQL loader are now parameterized, which allows Data Integrator to submit batches of data to the database for faster processing on the database side. This enhancement applies to straight INSERT/UPDATE/DELETE rows to the MySQL target table.

ADAPT00832129

ADAPT00843602

ADAPT00847284

80 BusinessObjects Data Integrator Release Notes

When importing Sybase ASE tables containing date or time data type columns, those columns were not imported into Data Integrator. This issue has been resolved.

Two functions have been added to this release of Data Integrator: base64_encode and base64_decode. For details on these functions, see Inxight integration for reading unstruc-

tured text on page 38.

When joining two different data sources and the function gen_row_num is used in the Query transform, the row generation might not generate the correct result. This issue has been addressed.

Resolved issues

ADAPT00847565

ADAPT00852830

ADAPT00853579

ADAPT00853583

When loading to an Oracle partitioned table using the Oracle bulk loader in API mode with enabling partition, the index of the table is now automatically maintained by Data Integrator within a loading transaction by default. If you specified not to maintain the index via the OracleBlkldrAPIParallelLoadEnableIndexMaint parameter, your index maintenance preference will continued to be honored. However, if the option has not been set, the behavior of this release has been changed to automatically maintain the index. This new behavior could have a performance implication. However, this new behavior safeguards the Oracle table indexes from being corrupted.

This release includes the new database type HP Neoview version 2.2. This support is available on Windows and Linux platforms only. In future releases, the HP Itanium 64-bit support will also be added.

When logging in to an instance of Data Integrator that has connections to multiple repositories and each is associated with a different central repository, the central repositories did not get automatically activated when logging into the local repositories. This issue has been addressed.

When installing Data Integrator, if the Management Console was installed without the Data Integrator Designer component, Auto Documentation did not work properly. This issue has been addressed.

ADAPT00853591

ADAPT00855478

ADAPT00855856

In this release, Data Integrator has been certified to work with Data Quality 11.7. This certification is available on all platforms.

Data Integrator sometimes extracted the data from Oracle CDC datastore tables in the incorrect sequence. This problem has been resolved.

When two repositories are associated with a Job Server, the default repository option was not editable from the Server Manager. This issue has been resolved.

BusinessObjects Data Integrator Release Notes 81

Resolved issues

ADAPT00856745

ADAPT00858554

ADAPT00866776

ADAPT00867379

ADAPT00868134

ADAPT00871130

When defining a WHERE clause in a Query transform using the LIKE operator with a constant such as %MONDAY%, rows containing only blank characters were incorrectly matched. This issue has been resolved.

When loading to a Microsoft SQLServer table containing a column of text data type, the Data Integrator job failed. This issue has been resolved.

In the XML pipeline transform, the UNC path name can now be used.

The Data Integrator Web Service message was missing the

/localtypes: tag in the response SOAP message body

for batch jobs. This issue has been resolved.

When using the Netezza bulk-loading option on any UNIX or Linux platform, the bulk loader could fail with a SQL syntax error if the schema of the target table was large. Data Integrator generates an INSERT...SELECT SQL statement as part of the bulk-loading process. If the length of this statement was larger than 1024 bytes, the generated SQL might be corrupted with extra spaces at every 1024 bytes. This problem has been resolved.

When using the wildcard character to read multiple files from the same file reader, Data Integrator failed to read the files that have a header row only. This problem has been resolved.

ADAPT00871857

ADAPT00874472

82 BusinessObjects Data Integrator Release Notes

If you checked out an object such as a data flow from the central repository and one of its dependent objects was already checked out by another user to a different local repository, the check-out operation of object failed. This issue has been resolved.

When using Table Comparison with the Sorted input option, Data Integrator compared the sort ordering of UTF-16 characters incorrectly, which sometimes resulted in incorrect output. This problem has been resolved.

Known issues

This section lists known issues in ascending order by Business Objects ADAPT system Problem Report tracking number.

ADAPT00619350

ADAPT00619427

ADAPT00619479

Do not use Microsoft SQL Server as CMS connection with sa as the login name.

Could not position within a table

and

Could not do a physical-order read to fetch next row

The reference number from the IBM site is Reference# 1172548

http://www-1.ibm.com/sup port/docview.wss?uid=swg21172548

ADAPT00619744

84 BusinessObjects Data Integrator Release Notes

To correctly configure metadata Integrator for a BusinessObjects Enterprise XI Release 2 CMS instance that is not on the same computer as Data Integrator, refer to the Business Objects Customer Support Web site and locate Knowledge Base article c2019073.

Known issues

ADAPT00620063

ADAPT00620112

ADAPT00620168

ADAPT00620210

When comparison tables are on Microsoft SQL Server, the Table Comparison transform's Sorted input option generates incorrect results. The Data Integrator engine runs in binary collation while the Microsoft SQL Server database runs in a Latin1 collation sequence. As a result, Microsoft SQL Server generates the data in a different sort order. This issue will be addressed in a later release.

Use the Row-by-row select or Cached comparison table option in the Table Comparison transform.

When you import a table by name and this table is the target table of a Table Comparison transform, the table comparison does not refresh the database changes.

Open the Table Comparison transform to refresh the output schema.

When adding a very large project to a central repository, a large memory leak occurs. After adding large projects to a central repository, restart the Designer.

Due to an issue in Attunity software, when using Attunity Connect to load into a DB2 table on MVS that contains varchar columns larger than 255, the data inserted into those columns is always an empty string regardless of what you actually insert.

ADAPT00620227

ADAPT00620654

ADAPT00620815

When using Attunity CDC, import from browser and import by name behave differently. For example, when you type either CDC_DSNX4612:PUBLIC or CDC_DSNX4612 as the owner name, the same table will be imported into the repository.

When you import a table by name, always use the full owner name such as CDC_DSNX4612:PUBLIC and also define the alias for this full name.

The Import By Name functionality for the Web Services adapter is not working. The work-around is to use metadata Explorer (browser) to import any metadata from the Web Services adapter.

Changed-data capture for Oracle 10g R2 does not support multibyte table or column names due to an Oracle limitation.

BusinessObjects Data Integrator Release Notes 85

Known issues

ADAPT00620821

ADAPT00621063

ADAPT00644390

ADAPT00655863

ADAPT00658811

ADAPT00661876

Windows NT authentication for Microsoft SQL Server doesn't work for the Administrator and metadata Reports. Use Microsoft SQL Server authentication instead.

When using the Get by label option of the multiuser development interface with a non-existent label, Data Integrator returns the latest version of the objects.

Auto-correct loading might fail for Microsoft SQL Server, ODBC, DB2, Informix, Sybase IQ, or Teradata databases if the data contains NULL termination characters at the end of varchar data. The auto-correct SQL does not generate properly. To work around this issue, use the rtrim_blanks_ext functions on columns that you know will contain NULL termination characters to trim the blanks.

This issue also occurs with Sybase and Sybase IQ targets in regular load mode. Use the rtrim_blanks_ext to work around this issue for those databases as well.

Data Integrator Web Services only supports the UTF-8 code page.

When creating a Data Integrator repository on IBM DB2, only the UTF-8 code page is supported.

A Data Quality project cannot be imported if the project path name is longer than 230 characters.

ADAPT00673438

ADAPT00688056

ADAPT00688210

ADAPT00688976

86 BusinessObjects Data Integrator Release Notes

Data Quality Project descriptions are not imported into Data Integrator Object Library.

Stopping a Data Integrator job will not stop the Data Quality process.

Data Integrator should not import a project with more than one socket writer.

If the Web Service Adapter does not work due to an incorrect keystore path, you must restart the Tomcat server after you specify the correct keystore path.

Known issues

ADAPT00695056

ADAPT00733805

ADAPT00736500

The Adapter SDK has the following syntax limitation for DTDs that can be imported for use as an adapter source or target.

Conditional syntax cannot be included in the DTD file. Sample syntax that fails:

<!ENTITY % IN CLUDE_EBHEADER "INCLUDE"><![ %INCLUDE_EBHEAD ER;[<!ENTITY % eb:MessageHeader SYSTEM "ebHeader.dtd"> %eb:MessageHeader; ]]

The following syntax is supported and produces the same result:

<!ENTITY % eb:MessageHeader SYSTEM "ebHead er.dtd"> %eb:MessageHeader;

This limitation only applies to DTD files read by an adapter, not those read by the Designer. It is important to note that the only file the adapter reads is the one configured in the adapter operation. Adapters do not read embedded DTD files.

If the input to the Data Quality transform is not mapped and the output of the transform is mapped, Data Integrator will hang when the Job is executed. Please ensure that the proper input mapping is done before running the Job.

Under the following circumstances, the Profiler may stay in the pending state indefinitely:

• The repository is not connected to a job server.

• The associated job server is not running. Please check

the above before running the Profiler again.

ADAPT00738936

There is a known issue with Data Quality projects that contain periods (".") in the field names. Although the projects can be imported from the Data Quality server, Data Integrator cannot correctly map these fields. The project can be used in a data flow as long as the fields containing periods are not used. A workaround for this problem is to rename any Data Quality project fields to use another character instead of the period. Business Objects suggest using the underscore character ("_").

BusinessObjects Data Integrator Release Notes 87

Known issues

ADAPT00741303

ADAPT00743879

ADAPT00794024

The Message Client sample (client_sample) program in the AccessServer directory has not been updated for this release. Please check the Business Objects Knowledge Base for any updates.

Data Integrator does not correctly handle Oracle UTF8 database the NCHAR data type in a table comparison rowby-row select option if the column is used in the primary key or compared column. In such a case, Data Integrator generates an incorrect SQL SELECT statement. As a result, the table comparison result is not correct.

When using Data Integrator to load to a Teradata V2R6.2 target table via the ODBC driver, you will receive the following error:

error message for operation <SQLExecute>: <[NCR][ODBC Teradata][Teradata Database] The source parcel

length does not

match data that was defined.

This issue is due to a bug in the Teradata ODBC driver. To work around this issue, either set the commit size of the target table to 1, or use one of the bulk-loader methods to load your data

88 BusinessObjects Data Integrator Release Notes

Copyright information

SNMP copyright information

This section includes third-party copyright information.

SNMP copyright information

Portions Derivative Work - 1996, 1998-2000

Permission to use, copy, modify and distribute this software and its documentation for any purpose and without fee is hereby granted, provided that the above copyright notice appears in all copies and that both that copyright notice and this permission notice appear in supporting documentation, and that the name of CMU and The Regents of the University of California not be used in advertising or publicity pertaining to distribution of the software without specific written permission.

CMU AND THE REGENTS OF THE UNIVERSITY OF CALIFORNIA DISCLAIM ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL CMU OR THE REGENTS OF THE UNIVERSITY OF CALIFORNIA BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES

WHATSOEVER RESULTING FROM THE LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

90 BusinessObjects Data Integrator Release Notes

SNMP copyright information

Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

Neither the name of the NAI Labs nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANT ABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

The name of Cambridge Broadband Ltd. may not be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER "AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANT ABILITY AND

BusinessObjects Data Integrator Release Notes 91

IBM ICU copyright information

FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS

INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

Specifications subject to change without notice. Not responsible for errors or admissions.

IBM ICU copyright information

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, provided that the above copyright notice(s) and this permission notice appear in all copies of the Software and that both the above copyright notice(s) and this permission notice appear in supporting documentation.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF THIRD PARTY RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS INCLUDED IN THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR CONSEQUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.

92 BusinessObjects Data Integrator Release Notes

IBM ICU copyright information

Except as contained in this notice, the name of a copyright holder shall not be used in advertising or otherwise to promote the sale, use or other dealings in this Software without prior written authorization of the copyright holder.

BusinessObjects Data Integrator Release Notes 93

IBM ICU copyright information

94 BusinessObjects Data Integrator Release Notes

Business objects DATA INTEGRATOR 11.7.3 RELEASE NOTES 10-2007 User Manual

Specifications and Main Features

Frequently Asked Questions

User Manual

Contents

About these Release Notes

Supported platforms and versions

Compatibility update

Supported products

Unsupported products

Limitations update

Migration considerations

Logs in the Designer

Data flow cache type

Pageable cache for memory-intensive data flows

Embedded data flows

Oracle repository upgrade

Solaris and AIX platforms

Data quality

To migrate your data flow to use the new Data Quality transforms

Distributed data flows

XML Schema enhancement

Password management

Web applications

Web services

Deployment on UNIX

Adapter interfaces

JMS adapter interface

Salesforce.com adapter interface

Documentation updates

New features in version 11.7.3

HP Neoview database support

Inxight integration for reading unstructured text

base64_encode

base64_decode

Getting Started Guide

Job Server enhancements

Citrix Presentation Server version 4.0

Designer Guide

AIX user resource limits

Changed-data capture for Microsoft SQL Server 2005

To configure publications for Microsoft SQL Server 2005 CDC

Creating Microsoft Excel workbook file formats on UNIX platforms

To create a Microsoft Excel workbook file format on UNIX

Reference Guide

Maximum number of loaders

Reserved words

Microsoft Excel workbook format for UNIX

Salesforce.com Adapter Interface Guide

Resolved issues

Known issues

Copyright information

SNMP copyright information

IBM ICU copyright information